Greetings,
I would like to announce the following beta program offered by 3Com. We
are presently
developing a utility which will aide in the conversion of Netserver chasses
to HiperARC.
If any of you are interested, please browse to http://totalservice.usr.com
and apply online
for the beta program.
Description of utility:
This utility will automatically configure a HiPer ARC 4.1 card based on a
configuration derived from an
existing (running) NetServer 3.7 card. While it is known that the
functionality of these two platforms do
not completely overlap, it is believed that an automated process can
relieve 90% of the burden of
migration.
The basic approach will be a 3-step process:
1.Extractor: Extract the NetServer configuration from a series of show
screens. The configuration
will be stored in a binary file on the application client machine.
2.Translator: Derive a set of HiPer ARC CLI configuration commands from
the extracted
NetServer configuration. The mapping between the NetServer
configuration parameters and the
HiPer ARC CLI command set
3.Configurator: Configure the target HiPer ARC network parameters
(through the NMC) and apply
the derived configuration commands.
Available on Solaris 2.5.1, Windows 95/NT.
The ArcSwitcher gets the configuration data from the NETServer by
establishing a telnet session to
the card (using the !root user) and issuing a series of CLI show commands
and capturing/parsing the
returned data. This information is translated into a series of HARC CLI
commands which are "pushed"
to the HARC card over another telnet session (using the adm user), i.e.:
1.ARC Switcher strips data from the NETServer Card
2.Remove card and places a HARC into the chassis
3.HiPer ARC is "primed" by autoconfigure routine
4.ARC Switcher dumps NETServer data to HARC
The HARC is autoconfigured (step 3 above) with a network address using the
following procedure:
1.Set autoconfiguration parameters for the HARC via the NMC
2.Turn on autoconfiguration on the NMC
3.H/W reset the HARC
4.Turn off autoconfiguration on the NMC
Feedback we are looking for:
At this point we are mostly interested in hearing feedback on:
1.Installation problems
2.Fatal execution problems
3.Translation mistakes or omissions
The intent of the ArcSwitcher is that we automate transition between the
NETServer and HiPer ARC
as much as possible, while at the same time pointing out areas which cannot
be converted or handled
automatically.
It would be good to keep an eye out for any failed commands during the
HiPer ARC configuration step
of the transition process. These errors show up in the log file and are
noted as having occurred by the
top level arcswitch or arcswitch.pl script. Ideally these errors should not
occur at all.
Warnings regarding the translation phase are detailed in the translator.log
log file. It is probably
prudent for the user to read through this log to take note of any
limitations related to the translation.
Subject:(usr-tc) Accounting and Radius problem... From: Rick Payne <rickp@ops.netcom.net.uk> Date: 1998-10-06 17:22:27
Dragan D. Vecerina writes:
> Hi,
> Is there any workaround or suggestions for loss of accounting records
> from HiperArc Card (4.1.11).
> Problem is that it doesn't send (sometimes) ACCT OFF information
> to radius server when user disconnect.
Indeed, and does this explain why the ARC loses track of how many IP's its
assigned from the pool?
On one of ours today, we had no-one connected, but it still reckoned 2
IP's from the pool were in use. *sigh*.
Rick
Subject:Re: (usr-tc) Pt to Pt Test Link From: Mark Lemmert <techmgr@athenet.net> Date: 1998-10-07 16:14:00
>I'd like to setup a in-house 56k line that would be just to test some
>compression ideas. I'm trying to avoid installation fees if I can help it
>just to test. Can I set up something using 2 56k CSU's and routers? If
>not, is there a cheap box that I can buy that I can set this up with to
>test?
>______________________________________________________
>Thanks,
>Greg Coffey 307-234-5443 Fax 307-234-5446
>CoffeyNet v.90 56k Access for Casper & Douglas
>142 S. Center St.
>Casper, WY 82601 http://www.coffey.com
If you made an cat5 cable /w RJ45 ends with
a special wiring scheme you can simulate make
the two CSUs talk as though you had a telco circuit
between them.
See your CSU manual to find out which pins are
for transmit/receive and tip/ring. You will want
to have the transmit/tip matched with the receive/tip
on the other end and the transmit/ring matched with
the receive/ring on the other end.
Hope this helps.
Mark Lemmert techmgr@athenet.net
Chief Technical Officer (920)954-9799
AthEnet Data Exchange http://www.athenet.net
A Cisco AS5100 unit (more or less a 3CTC Chassis with Cisco cards) has
been dropped on my lap, and I have a somewhat confusing problem:
When connected dialling an analog line, the modem responds,
authenticates, and (could presumably) enter PPP mode. When dialling the
T1 number (these are x2/V.90 modems), there is no response -- only a
brief pause, then a fast busy signal.
The telco providing the T1 has never done such a circuit before -- could
the problem be there?
When going to Programmed Settings (for T1 card) > DS0 <-> Modem
Configuration, I cannot assign a modem to the slots -- Get returns
nothing, the modem cells are white, indicating that they are
unavailable, but the DS1 channels are aqua, indicating available.
How do I make these modems available? Have I missed a setting?
This is what I need to associate each modem to its own timeslice,
correct?
BTW, I'm using TCM 5.5.1
________________________________________________
Following the help file given, I'm supposed to select a cell for the
modem and a trunk on the T1, then press Connect. It does nothing. Of
course, according to the help file legend, *gray* indicates unavailable,
which disagrees with the visual legend...
_-~=_-~=_-~=_-~=_-~=_-~=_-~=_-~=_-~=_-~=
Andrew Champion andrew@lakes.com
507.386.0408
Lakes Internet 888.386.0444
:-_=:-_=:-_=:-_=:-_=:-_=:-_=:-_=:-_=:-_=
Subject:(usr-tc) Excessive data transferred / DNS resolution From: david.perrin@mnco.com Date: 1998-10-13 14:20:32
Working with the TCS 3.5 release using the Hiper ARC version 4.1.11 ,
I'm running into a couple of significant
issues.
We are currently using a NetServer card in production and are planning
to move to the Hiper ARC card in a new
chassis, but I'm having the following problems:
1. There's an excessive amount of data transferred during a session
with the Hiper ARC card versus the
Netserver card connection. It's on order of 5 times the data
transferred with a Hiper ARC connection versus
the Netserver connection.
2. DNS connectivity is acting strangely. DNS seems to be active and
working OK when telnetted into the
Hiper ARC console itself. The problem occurs when attempting to
use DNS via a Windows 95 session
connected remotely via PPP. For example I can ping a DNS host
name successfully, when using the
hostname along with the domain name (ie. abcdef.corp.mnco.com),
but receive a Bad IP address response
when trying to use the the hostname (ie. abcdef) . The DNS
domain name is set properly on the server
and as I said earlier I can sucessfully use DNS with just a
hostname when telnetted into the Hiper ARC
card.
Any ideas or suggestions would be helpful...
Subject:(usr-tc) using round robin & fixed assignment From: Taino d Johnston <usr-list@accesscom.com> Date: 1998-10-14 16:23:22
We recently switched telcos, which then switched us from channelized T1
lines to PRI lines. The telco switch was from PacBell to ICG. Upon
switching we found ourselves facing a lot of problems, however, it was this
last problem that we felt was quite strange and worth noting.
In order to do some tests with the new service, we changed the
configuration of our equipment from 'round robin' to 'fixed assignment'.
After we thought the experienced problems were corrected we started getting
reports of busy signals -- during time periods when there were modems
available. The problem turned out to be that the PRI lines were resetting
faster than our modems were. Essentially, this meant calls would be routed
back down to a newly opened line on the PRI but would hit a modem that had
not been reset. Our testing found that we were giving busy signals out
about 20% of the time. Switching back to 'round robin' seems to have since
corrected the problem and we have not given out any more busy signals.
What I am trying to find out is why when configured for 'fixed assignment'
we were giving out busy signals? The problems went away once we switched
to 'round robin'.
We are running a few TC chassis containing Quad Digital & Digital/Analog
modem cards with PRI lines. All our Quad modem cards, NETserver PRI cards,
Dual PRI cards and NMC cards are running the current versions of the
software.
Taino Johnston
Manager, Technical Support
Access Internet Communications
+----------------------------------------------------------------+
| Taino d Johnston | Phone: (408) 777-8190 |
| Manager, Technical Support | Tech Support: (408) 342-0551 |
| | Fax: (408) 777-8191 |
| Access Internet Communications | http://www.accesscom.com/ |
| tdj@accesscom.com | support@accesscom.com |
+----------------------------------------------------------------+
Subject:(usr-tc) V.90 is it safe to put on non-hyper dsp Total Control From: Mike Hamrich <mhamrich@drfast.net> Date: 1998-10-14 23:25:13
It's been a while since I have seen a V.90 post.
Any advice for taking a X2 enabled year and a half old 2.5.X release Total
Control unit up to V.90 successfully?
The one thing I did hear was to reset modem to factory default. Have not
read if you need to upgrade in steps with new then what in the unit but
older then what current. If you need to put interim release on. How do you
get them?
If this has been covered. Please point me to an archive.
Thanks
Dave Winston
Support
DrFast.Net, Inc.
Subject:(usr-tc) ISDN: Netserver vs. Modems From: Mark Lemmert <cto@athenet.net> Date: 1998-10-15 15:58:33
Currently I am terminating the ISDN calls on the Netserver (I-ports)
and am having problems with routes getting 'jammed' in the system
for ISDN callers with static addresses.
Essentially what happens is sometimes the caller will disconnect
from a Netserver and the route for their static IP won't be flushed
out, so they won't be connected but the route will remain. The client
will then connect to a difference Netserver and a new route will
be created on that server. I am broadcasting the routes via RIP so
the border routers end up with two routes for the clients static IP
when this happens and the client has problems.
3com has suggested that I terminate the ISDN calls on the quad modems
instead of on the Netserver to solve this problem. Has any body else
had this problem? Can anybody give an opinion on whether 3coms suggested
fix is likely to have an affect on this? Thanks!
Mark Lemmert cto@athenet.net
Chief Technical Officer (920)954-9799
AthEnet Data Exchange http://www.athenet.net
Subject:(usr-tc) Filters for local users on HARC From: Alex Substanley <das@gol.com> Date: 1998-10-19 18:08:19
Hi,
I'm trying to setup filters for local users rather than interfaces on a HipeARC running version 4.1.11. I've looked through the previous messages in this list, but couldn't find the information. Could just be me :)
I've set up two filters test.in and test.out on the HARC, I've added them and they verified without any problems. I've turned filter_access to on to use with user specific filters (according to the documentation) I've also assigned the filter to the user, but it doesn't seem to be binding the filter to the user so far as I can tell.
When I changed the filter to interface instead of user, it seemed to work fine.
I'm sure that this has been addressed in previous messages, but if anyone can shed some light on this...
The filter I am using for the incoming filter is a simplified filter I am using for testing.
I expect that this filter will block everything, but it appears that the user retains full access.
#filter
IP:
005 REJECT src-addr=<ip address>;
010 DENY;
and when I display the user info. I get this:
INFORMATION FOR USER: <userid>
Status: INACTIVE
Type: NETWORK
Expiration: NONE
Message: (D)
Phone Number: (D)
Alternate Phone Number: (D)
Input Filter: test.in
Output Filter: test.out
Modem Group: all (D)
Session Timeout in seconds: 0 (D)
Idle Timeout in seconds: 0 (D)
Tap Status: DISABLED
Tap Format: ASCII
Tap Output: SYSLOG
Tap Facility: INVALID
Tap Loglevel: CRITICAL
Tap Address: 0.0.0.0 (D)
Chat Script Name:
--
____________________________________________
Alex Substanley Global OnLine Japan
Engineering Department
Das Man TEL: 81-3-5334-1700
Systems Engineer FAX: 81-3-5334-1711
____________________________________________
Subject:(usr-tc) Filters for local users on the HARC From: Alex Substanley <das@gol.com> Date: 1998-10-20 20:59:30
Hi,
I'm trying to setup filters for local users rather than interfaces on a HipeARC running version 4.1.11.
I've looked through the previous messages in this list, but couldn't find the information. Could just
be me :)
I've set up two filters test.in and test.out on the HARC, I've added them and they verified without any
problems. I've turned filter_access to on to use with user specific filters (according to the
documentation) I've also assigned the filter to the user, but it doesn't seem to be binding the filter
to the user so far as I can tell.
When I changed the filter to interface instead of user, it seemed to work fine.
I'm sure that this has been addressed in previous messages, but if anyone can shed some light on
this...
The filter I am using for the incoming filter is a simplified filter I am using for testing.
I expect that this filter will block everything, but it appears that the user retains full access.
#filter
IP:
005 REJECT src-addr=<ip address>;
010 DENY;
and when I display the user info. I get this:
INFORMATION FOR USER: <userid>
Status: INACTIVE
Type: NETWORK
Expiration: NONE
Message: (D)
Phone Number: (D)
Alternate Phone Number: (D)
Input Filter: test.in
Output Filter: test.out
Modem Group: all (D)
Session Timeout in seconds: 0 (D)
Idle Timeout in seconds: 0 (D)
Tap Status: DISABLED
Tap Format: ASCII
Tap Output: SYSLOG
Tap Facility: INVALID
Tap Loglevel: CRITICAL
Tap Address: 0.0.0.0 (D)
Chat Script Name:
--
____________________________________________
Alex Substanley Global OnLine Japan
Engineering Department
Das Man TEL: 81-3-5334-1700
Systems Engineer FAX: 81-3-5334-1711
____________________________________________
Subject:(usr-tc) PRI vs T1 ? From: Robb Bryn <robb@cape-fear.net> Date: 1998-10-21 09:04:21
This is a multi-part message in MIME format.
------=_NextPart_000_0004_01BDFCD1.CA8ACB60
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
We are looking at expanding our current line count (which is using H-DSP =
cards with PRI lines) and are faced with a small dilema. We have always =
used PRI lines in the past for our incoming lines and are now pricing =
PRI vs T1. In our area T1's are priced about $600/month cheaper than a =
PRI. Aside from PRI's being easier to setup, I've caught rumors that =
T1's are not as fast as PRI (when considering a v.90 call & all other =
factors are equal). I have only one question... are these rumors true?
Thanks
Robb Bryn
------=_NextPart_000_0004_01BDFCD1.CA8ACB60
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN">
<HTML>
<HEAD>
<META content=3Dtext/html;charset=3Diso-8859-1 =
http-equiv=3DContent-Type>
<META content=3D'"MSHTML 4.72.3510.1400"' name=3DGENERATOR>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT size=3D2>
<P>We are looking at expanding our current line count (which is using =
H-DSP=20
cards with PRI lines) and are faced with a small dilema. We have always =
used PRI=20
lines in the past for our incoming lines and are now pricing PRI vs T1. =
In our=20
area T1's are priced about $600/month cheaper than a PRI. Aside from =
PRI's being=20
easier to setup, I've caught rumors that T1's are not as fast as PRI =
(when=20
considering a v.90 call & all other factors are equal). I have only =
one=20
question... are these rumors true?</P>
<P>Thanks</P>
<P>Robb Bryn</P></FONT></DIV></BODY></HTML>
------=_NextPart_000_0004_01BDFCD1.CA8ACB60--
Subject:(usr-tc) Interested in Linux on the Netserver? From: Mark R. Lindsey <mark@vielle.datasys.net> Date: 1998-10-21 21:52:47
I'd like to get a show of hands, so to speak, for everyone interested in
Linux on the Netserver. If you are interested, please send the message
`subscribe' to
lon-request@datasys.net
I see this as a relatively important venture that's good for every
Netserver user. And this venture is good for 3Com, too -- it will provide
an extended lifetime to those of us who plan to stick with our Netservers,
and will make 3Com the first vendor I know of with an open-source
high-density telecom system. That, friends, is a radical step forward.
We're talking about a fresh start for the TC Netserver.
---
Mark R. Lindsey, mark@datasys.net
Internet Engineering, DSS Online LLC
Voice: 912.241.0607x200, Fax: 912.241.0190 (US)
Subject:Re: (usr-tc) Interested in Linux on the Netserver? From: Allen Marsalis <am@shreve.net> Date: 1998-10-22 00:09:40
At 12:09 AM 10/22/98 -0400, Mark R. Lindsey wrote:
>I'd like to get a show of hands, so to speak, for everyone interested in
>Linux on the Netserver. If you are interested, please send the message
>`subscribe' to
> LON-request@datasys.net
a list? so soon? ok, I'll subscribe to see what's up. But here is
my .02 on the subject..
Lets say you get USR to cooperate (1st feat). Then you get linux up
on the netserver (2nd feat). What if Quake still lags? I mean Quake
Lag is a symptom of a much bigger problem. And it might be hardware
related and that has been hinted at.
So if someone was able to pull off both these feats and our customers
still bitched, it would not do us much good and we would still use
100% arcs, albeit more regretfully..
Of course if there are any real hardware limits, they could be
minimized and dealt with so udp performance would be as good or better
than it is now, which is good enough for many.. just not us I
suppose.. We house alot of Quake players..
How about getting 3COM to give us specs and info on HiperARC's
too while you're at it.. ;->
Allen
I have a problem with Radius Accounting packets reporting the wrong NAS
port. When the Radius server recieves a packet for authentication it will
report the correct NAS port but when it sends the Accounting start/stop
packet it always reports NASPort=1.
This problem only seemed to occur after upgrading TCS 3.3
Config is:
Total Control Chassis w/
NMC v5.5.5
HARC v.4.1.11
Hiper DSP v.1.2.5
Radius = IEA Software's RadiusNT v2.5
Thanks
Robb Bryn
Subject:Re: (usr-tc) Problems with brackets in session_start_message From: Rick Payne <rickp@ops.netcom.net.uk> Date: 1998-10-26 15:45:44
Mike Wronski writes:
> Have you tried escaping the '(' with back slashes??
Yes.
> set slip session_start_message "SL/IP session from \(%server_ip\) to
> %client_ip beginning..."
I entered it as above. Same thing:
SL/IP session from ( 194.42.231.28 to 255.96.0.0 beginning..
(assuming I don't need to relaod after setting the parameter).
Code is 4.1.78 on the HiperARC.
Rick
Subject:(usr-tc) HELP: need to route IP addresses to a customer over TC Hub From: C Thompson <cthompson@wingnet.net> Date: 1998-10-30 17:22:56
We have not had to do this before, but I need to route a Class C address
across an ISDN connection on a USR/3com Total Control hub to one of
our customers. Also, for future reference, it would be helpful to know
what to do for a subnet as well.
Can anyone provide helpful pointers on what I need to do
1) in RADIUS
2) in the TC Hub (if anything)
3) in my router
4) anything else?
Thanks for any help you can offer. If you reply to the list, please also e-
mail me privately, too, as it's the weekend.
Thanks,
Craig Thompson
WingNET Internet Services,
P.O. Box 3000 // Cleveland, TN 37320-3000
423-559-LINK (v) 423-559-5444 (f)
http://www.wingnet.net
Thought for the day:
Concerto (n): a fight between a piano and a pianist.
Ascend now has both a global switch you can throw to check incoming
packets source addresses against the assigned address and a radius
RADIUS attribute to do it on a per-session basis.
How hard can this be? Why isn't it in 3com products?
--
Aaron Nabil
Subject:Re: (usr-tc) Ascend version 7 contains anti-spoofing code, 3com? From: bryan s. blank <bryan@supernet.net> Date: 1998-11-01 10:41:20
|o| Ascend now has both a global switch you can throw to check incoming
|o| packets source addresses against the assigned address and a radius
|o| RADIUS attribute to do it on a per-session basis.
|o|
|o| How hard can this be? Why isn't it in 3com products?
no kidding ... one solution is to use the super cool radiator
radius server from http://www.open.com.au and use it to
dynamically build filters for a hiperarc-based system ...
failing that, i don't think you have any other options ;(
this is definetly something that people should demand from nas
vendors
|o|----------------------------------------------------------------------|o|
|o| bryan s. blank (203)-351-1178 voice |o|
|o| senior systems analyst (203)-351-1186 fax |o|
|o| discovernet, incorporated (203)-979-5126 emerg |o|
Subject:Re: (usr-tc) Ascend version 7 contains anti-spoofing code, 3com? From: Mark R. Lindsey <mark@vielle.datasys.net> Date: 1998-11-01 13:04:05
: |o| Ascend now has both a global switch you can throw to check incoming
: |o| packets source addresses against the assigned address and a radius
: |o| RADIUS attribute to do it on a per-session basis.
:
: no kidding ... one solution is to use the super cool radiator
: radius server [...]
: this is definetly something that people should demand from nas
: vendors
A knob is nice, but don't forget that we *can* solve the problems
without their help. I would prefer that they let us solve such problems
in higher layers (e.g., with Radiator); the alternative is for them to
complicate their code.
In general, problems should be solved at as high a layer as possible.
I want high-quality bolts from 3Com: leave the bridge-building to me.
Subject:Re: (usr-tc) Ascend version 7 contains anti-spoofing code, 3com? From: bryan s. blank <bryan@supernet.net> Date: 1998-11-01 13:19:52
|o| A knob is nice, but don't forget that we *can* solve the problems
|o| without their help. I would prefer that they let us solve such problems
|o| in higher layers (e.g., with Radiator); the alternative is for them to
|o| complicate their code.
|o|
|o| In general, problems should be solved at as high a layer as possible.
|o| I want high-quality bolts from 3Com: leave the bridge-building to me.
this is a very interesting way of looking at things, i hadn't
considered it this way, and i like it. thanks!
|o|----------------------------------------------------------------------|o|
|o| bryan s. blank (203)-351-1178 voice |o|
|o| senior systems analyst (203)-351-1186 fax |o|
|o| discovernet, incorporated (203)-979-5126 emerg |o|
Subject:(usr-tc) DS1 MIB support for HDMs From: Mark R. Lindsey <mark@vielle.datasys.net> Date: 1998-11-01 20:34:05
Howdy,
Has anyone read if the HDMs (HiPer DSPs) will support the DS1 MIB?
They don't seem to now, or, at least, one of my NMCs doesn't think so.
Is there something else I should be using to monitor for alarms and
line conditions?
Thanks.
---
Mark R. Lindsey, mark@datasys.net
Internet Engineering, DSS Online LLC
bryan s. blank was heard to say:
>
>|o| A knob is nice, but don't forget that we *can* solve the problems
>|o| without their help. I would prefer that they let us solve such problems
>|o| in higher layers (e.g., with Radiator); the alternative is for them to
>|o| complicate their code.
>|o|
>|o| In general, problems should be solved at as high a layer as possible.
>|o| I want high-quality bolts from 3Com: leave the bridge-building to me.
>
> this is a very interesting way of looking at things, i hadn't
> considered it this way, and i like it. thanks!
That's why I like 3Com's RADIUS server so well... I have complete control
over how each radius packet is handled.
--Ricky
Subject:Re: (usr-tc) DS1 MIB support for HDMs From: Charles Sprickman <spork@inch.com> Date: 1998-11-02 01:32:48
On Sun, 1 Nov 1998, Mark R. Lindsey wrote:
> Has anyone read if the HDMs (HiPer DSPs) will support the DS1 MIB?
>
I'm a little foggy on how I found this, but David Bolen pointed me in the
right direction. In a little test program I was futzing with I polled
this entry to get line status on the HDMs:
$oid=".iso.org.dod.internet.private.enterprises.usr.nas.rds1.usrds1StatT
able.usrds1StatEntry.usrds1StatE1PhysicalState"
This is from one of the usr mib files...
Charles
> They don't seem to now, or, at least, one of my NMCs doesn't think so.
> Is there something else I should be using to monitor for alarms and
> line conditions?
>
> Thanks.
>
> ---
> Mark R. Lindsey, mark@datasys.net
> Internet Engineering, DSS Online LLC
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
=-----------------= =
| Charles Sprickman Internet Channel |
| INCH System Administration Team (212)243-5200 |
| spork@inch.com access@inch.com |
= =----------------=
Subject:(usr-tc) HELP: need to route IP addresses to a customer over TC From: C Thompson <cthompson@wingnet.net> Date: 1998-11-02 08:31:10
We have not had to do this before, but I need to route a Class C address
across an ISDN connection on a USR/3com Total Control hub to one of
our customers. Also, for future reference, it would be helpful to know
what to do for a subnet as well.
Can anyone provide helpful pointers on what I need to do
1) in RADIUS
2) in the TC Hub (if anything)
3) in my router
4) anything else?
Thanks for any help you can offer. If you reply to the list, please also
e- mail me privately, too.
Thanks,
Craig Thompson
WingNET Internet Services,
P.O. Box 3000 // Cleveland, TN 37320-3000
423-559-LINK (v) 423-559-5444 (f)
http://www.wingnet.net
Thought for the day:
O, give thanks unto the Lord for He is good,
for His mercy endureth forever.
Subject:(usr-tc) NetServer/16 I-Modem Plus for sale From: TieUs System Administrator <admin@tieus.com> Date: 1998-11-02 12:40:16
This is a multi-part message in MIME format.
------=_NextPart_000_00AC_01BE065D.F1B097C0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Dear Friends:
We have a NetServer/16 I-Modem Plus for sale. We have this machine =
in about March, so this machine is about 7 month old. We only used it =
for about 2 month for testing our VPN usage. Since now we are using =
HyperArc as our enterprise solution, We want to get rid of the =
NetServer/16 I modem. It is X2 enabled and V.90 code will release on =
January, 1999 (according to 3com site)
I will try to sell it at about US$5000 obo. (please dont flame me =
for the price, I did not get any idea with what is a reasonable price =
for that now, we bought it at around US$9000., the list price is $9999.) =
If you are interested or having any question about it please give me =
an e-mail.
Dear All,
I have problem with my Netserver Card and NMC Card.
Those device did not pass minimum requirement for
Y2K Compliant.
Is this following correct minimum requirement for Y2K version ?
1. QuadModem minimum code: 5.5.x atau 5.6.x
2. NMC minimum code: 5.0.x (mine: 4.2.2)
3. Netserver Card minimum code: 3.4.x (mine 3.1.7)
Are there patches for those card ?
-TOTAL CONTROL NETSERVER PRI CARD SET (ETHERNET)
-NETSERVER CARD
What another action should we do, if our USR device didn't
pass minimum requirement for Y2K version ?
Thanks,
Budiw
Subject:Re: (usr-tc) Ascend version 7 contains anti-spoofing code, 3com? From: Kurtiss Johnson <kurtiss_johnson@mw.3com.com> Date: 1998-11-02 14:19:34
Keep your eyes open for HiPer ARC v4.2, which has similar functionality in
a feature called
Source Address Assurance. It also operates either via global config or via
a per-user RADIUS
profile setting.
Kurtiss Johnson
Product Manager - Access Routers
3Com Corporation - Carrier Business Unit
847-222-2279
Aaron Nabil <nabil@spiritone.com> on 11/01/98 09:25:47 AM
Please respond to usr-tc@lists.xmission.com
cc: (Kurtiss Johnson/MW/US/3Com)
Note: Some recipients have been dropped due to syntax errors.
Please refer to the "$AdditionalHeaders" item for the complete headers.
Ascend now has both a global switch you can throw to check incoming
packets source addresses against the assigned address and a radius
RADIUS attribute to do it on a per-session basis.
How hard can this be? Why isn't it in 3com products?
--
Aaron Nabil
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:(usr-tc) Low Connections/Disconnects From: Chad J. LaFrenz <clafrenz@rof.net> Date: 1998-11-02 15:38:47
Hello All--
We have 5 Total Controls (NetServer and Quads) running these code versions:
T1: 3.5.0
SS Quads: 5.10.9
NetServer: 3.8.1
NMC: 5.4.95
Settings are default except for the following:
Carrier Loss Delay: 20
Transmit Level: 13
DTE Rate Mode: fixed
Transmit Flow Control: hardware
Hardware Flow Control: dataOnRtsHigh
Default DTE Data Rate: bps57k
Packet Bus Answer Only: enable
All Results codes enabled.
All the hubs have E&MTypeII T1's pulled into them.
Problem: Users complaining of low connection rates and disconnects.
What I have experienced: Normally I connect from my home with a Sportster
Voice Internal at 49K. When I have problems it is when I hear the normal
connection tones and then right before it would normally connect I hear a
double "bong" sound. It then connects and shows me my DTE rate and not the
actual connection rate (which means to me it is a non-listed connection
speed in my windows inf file). Checking the chassis and port that I am on
and it shows that I am connected at 24k-16.8K. I can disconnect hit another
port and I am back at 49K. This is all on the same hub and happens on all
our hubs. I am starting to watch specific ports to see if they consistently
connect at lower baud rates; however, I just started doing this and don't
have enough data to post any results at this point. On this point though,
has anyone created an MRTG script (or similar) that would log connection
speed results on specified hubs or ports? Thus, creating almost a database
of connection logs for a specific modem? Is this even possible? Could then
be used to see if a specific modem is always connecting at a lower rate when
average over time.
What other users have said: Disconnects happen frequently and randomly
although more frequent when they are downloading a file or email attachment.
Connect speeds vary from normal nice 40K+ to 9600 through 24K.
Performance Manager shows zero errors when viewing the 24 Hour Total Group
information. When watching the modems through P.M. the disconnect reasons
are DS0 Teardown or v.42 disconnect. I have verified that all the modems on
all the hubs are set to the same exact settings. When looking at connection
speeds through P.M. it ranges from 9600 to 50K through out all the hubs.
Guess I am at a loss and since this group has solved my last problem before
3Com even called me back I figured I would start here this time. 8^) Any
suggestions would be greatly appreciated.
Regards,
Chad J. LaFrenz
Senior System Administrator
RoFIntUG
V: 970-945-4920 F: 970-947-1923
Proudly serving the Aspen, Glenwood Springs, Rifle, and Vail Valleys.
http://www.rof.net
Our CLEC was reporting errors on spans we had attached to HiperDSP cards.
These errors didn't show up with a test set watching the span, but the
switch saw a constant stream of them. It seemed to be directly dependant
on the incoming call rate on the span. I'm guessing it was some margin problem
the T-BERD was immune to but the switch wasn't.
I was just about to try and get some fancy test gear to look at the
T1 stream itself, but the CLEC said the problem went away. The only
changes we made recently were swapping out some cards in our M31
muxes, and the HiperDSP upgrade.
So I downgraded one of the cards to 1.0.8. The error stream returned
(with a vengance) on that span.
So 3COM, whatever you did in 1.2.68, please don't undo it! It works!
--
Aaron Nabil
Subject:Re: (usr-tc) DS1 MIB support for HDMs From: David Bolen <db3l@ans.net> Date: 1998-11-02 16:40:27
mark@vielle.datasys.net (Mark R. Lindsey) writes:
> Has anyone read if the HDMs (HiPer DSPs) will support the DS1 MIB?
They do - RFC1406.
It's not the same MIB as that supported by the dual cards (which use
the original - since deprecated - RFC 1232) but serves the same
purpose, and it's time 3Com updated to a current version. RFC1406
superceded RFC1232 in '93 :-)
There is also a replacement for the USR-specific additions to 1232.
For the dual span cards, they use UDS1.MIB, whereas the HDM uses
RDS1.MIB. At the DS0 level, replace DS0.MIB (dual cards) with
RDS0.MIB.
I'll enclose an excerpt below from some internal docs I worked up when
we were adapting to the introduction of HDMs into our system that
might help correlate MIBs to components.
Note that you need to be running an HDM aware (16MB) version of NMC
code (5.5.x) to be able to query HDM tables - the 4MB version (5.4.x)
can identify the card but not support any of its tables.
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
- - - - - - - - - - - - - - - - - - - - - - - - -
Subject:(usr-tc) Lan to Lan routing, manual dial out works, on_demand does not route From: Martin Oberle <moberle@gmx.de> Date: 1998-11-02 17:41:43
Hi.
I have the problem that I can dial manually to a location but
dial on demand does not work proberly. Please see details below.
I use an netserver card with only one PRI connected. That gives me
30 ports. There are 32 modems on 8 Quadmodem cards in the chassis.
*****************************
This is the netserver:
Command> ver
U.S. Robotics
Total Control (tm) NETServer Card V.34/ISDN with Frame Relay V3.5.34
Build date: Jun 19 1997
Build time: 14:02:28
Network Interface Card: Ethernet & Frame Relay Combination (26)
ISDN Interface Card : MUNICH32 (4)
Packet Bus Circuit : Enhanced
Licensed for 60 ports.
*****************************
This is a location I want do dial to:
Command> sh loc goethe
Location: goethe Type: Manual
Destination: 195.x.x.129 Netmask: 255.255.255.224
Protocol: PPP Options: Quiet
Group: 0 Max Ports: 1
Idle Timeout: 2 minutes High Mark: 0 bytes
Mtu: 1500 Async Map: 00000000
Dial Script: Send Command Wait for Reply
------------------------------ ----------------------------
--
at&f1*v2=5e1v1q0&w\r
atdt09192352\r CONNECT
*****************************
When I dial manually to the location, everything works ok. Modem S5 dials
to the remote network. The connection
is established and I can ping every machine on the remote network.
Command> sh ro
Destination Gateway Flag Met Interface VPN
-------------------- ---------------- ---- --- --------- ---
0.0.0.0 /0 195.xx.xxx.1 NS 1 net0 0
195.xx.xxx.128 /27 195.xx.xxx.129 NLC 1 ptp5 0
195.xx.xxx.0 /25 195.xx.xxx.2 NL 1 net0 0
Command>
Command> if
net0: flags=8016<IP_UP,IPX_DOWN,BROADCAST>
inet 195.xx.xxx.2 netmask ffffff80 broadcast 195.xx.xxx.127 mtu 1500
ptp5: flags=8125<IP_UP,IPX_DOWN,POINT_TO_POINT,PRIVATE>
dest 195.xx.xxx.129 netmask ffffffe0 mtu 1500
Command>
*******************************
Now I set the location to on_demand. I can see the route and the interface
ptp129 in the configuration before an connection is estabished. For my
understanding this in normal. But is ptp129 a valid interface when I have
anly 32 Modems and 30 ISDN-cannels?
*********
Command> if
net0: flags=8016<IP_UP,IPX_DOWN,BROADCAST>
inet 195.xx.xxx.2 netmask ffffff80 broadcast 195.xx.xxx.127 mtu 1500
ptp129: flags=8165<IP_UP,IPX_DOWN,POINT_TO_POINT,PRIVATE,SUSPENDED>
dest 195.xx.xxx.129 netmask ffffffe0 mtu 1500
************
When I ping the remote network, Modem S5 dials to the remote network
and makes a connection. But as you can see below the routing table is
not correct. There is only a route to the remote router, not to the network.
The netserver thinks the remote network network is still on interface ptp129
an I can't ping machines on the remote network.
But when you look at the interface ptp5 it shows an other netmask than
it is in the routing table.
So what can be wrong? Is there any other configuration necessary?
(For example proxy_arp, enhanced_routing, netmasks or something
like this).
Or is this a known issue and a software update will fix this?
Thanks.
Command> sh ro
Destination Gateway Flag Met Interface VPN
-------------------- ---------------- ---- --- --------- ---
0.0.0.0 /0 195.xx.xxx.1 NS 1 net0 0
195.xx.xxx.129 /32 195.xx.xxx.129 HLC 1 ptp5 0
195.xx.xxx.128 /27 195.xx.xxx.129 NL 1 ptp129 0
195.xx.xxx.0 /25 195.xx.xxx.2 NL 1 net0 0
Command> if
net0: flags=8016<IP_UP,IPX_DOWN,BROADCAST>
inet 195.xx.xxx.2 netmask ffffff80 broadcast 195.xx.xxx.127 mtu 1500
ptp129: flags=8165<IP_UP,IPX_DOWN,POINT_TO_POINT,PRIVATE,SUSPENDED>
dest 195.xx.xxx.129 netmask ffffffe0 mtu 1500
ptp5: flags=8125<IP_UP,IPX_DOWN,POINT_TO_POINT,PRIVATE>
dest 195.xx.xxx.129 netmask ffffffe0 mtu 1500
Subject:(usr-tc) timeslot mapping oddity on HDSP From: Lon R. Stockton, Jr. <lon@moonstar.com> Date: 1998-11-02 20:05:54
Has anyone seen this before?
After several months of perfect operation, one of our HDSPs started
handing out busy signals while it had plenty of modems available.
Upon investigation, it turns out that timeslot #7 had been reassigned
to channel #24 (instead of channel 7). Timeslot #24 was still assigned
to channel #24. I'm using PRIs, so TS24 is my d-channel...and using
fixed assignment, so channel #24 is never used.
The question is...does anyone have any idea how this configuration
bit could have been changed? *Something* caused TS7 to be reassigned
to C24 from C7 after several months of trouble free operation...I'm
just wondering what could have happened.
Relevant notes:
*chassis only allows one workstation to connect for management;
At the time of the incident, the workstation was powered on,
not running TCM, and the operator was asleep. (:
*HDSPs are running 1.2.5 code.
*chassis contains 2xHDSP, HARC, NMC, and 2xPS_70w.
Subject:(usr-tc) de-encrypting USRadius.mdb From: Ken Hodges <ken@rabun.net> Date: 1998-11-02 22:24:02
Pardon this if this question is already covered by the list, but I have
been unable to find it if the answer is available in the database. (I'm new
to this mailing list)
I would like to move away from the Total Control Security and Accounting
Program (RADIUS) and give Livingston's RADIUS a try. The problem is: The
USRadius.mdb password list is encrypted, and with over 1000 customers this
would be a major task to re-program each one.
Does anyone know of a utility that would de-encrypt the passwords so that I
can use them with another RADIUS program ?
Thanks in advance,
Ken
_______________________________________
Ken Hodges, President and CEO
ACME BrainWorks Internet Services, Inc.
Clayton Computers, Inc.
Rabun County, Georgia
"Where Spring Spends the Summer"
http://www.acme-brain.com or http://www.rabun.net
ken@rabun.net 1-706-782-9239
When we upgraded our 2059 to V.90 I had the same problem
when testing with my own Sportster x2 modem. (They were
early editions I purchased in bulk at a Home Shopping
Liquidation center right after X2 came out.) I tried
adding pauses but it didn't work. When I upgraded the
Sportster firmware to V.90, the problem went away. We put
a BIG banner link on our home page pointing users to
the 3Com V.90 upgrade page. Many of our Sporster users
took advantage of it. I also upped the "Carrier loss
detect delay" from 7 to 14 which really helped the problem
of disconnects immediately following connecting.
We upgraded a few weeks ago. We only had one customer whose
modem just would not connect well. He could only connect
every 4th or 5th try at any decent speed. It was a used flex
winmodem type that he had found a V.90 upgrade for. I'm
convinced that modem was just not capable of V.90. After
working with him for a while he ended up dumping it and
buying a new Sportster. From the same Computer and phone
line he is now getting 49k to 53k connects.
It was kinda cool when we upgraded. We had busy signals
every night for a week while users played with their
new found speed. And the compliments poured in. We received
numerous calls the first few days, but we just preached
upgrade upgrade upgrade to the customers. We lost zero
customers and gained 20+ last week alone from referrals
from existing customers.
As for users connecting but not able to surf, make sure
they're not NT users that still have Software compression
enabled. They will show as running MS compression, with
plenty of CRC errors after authenticating and no bytes
transfered. If you ping them you will see a few bytes
output but nothing in return. Also the DNS IP addresses
on the hub disapeared when I upgraded the Firmware. You
might want to check that too. And the Netserver was acting
flakey after the upgrade until I power cycled the whole
rack.
I left one Quad running x2. We've received about a dozen
"You must have a bad modem" emails from people getting
used to V.90. So you might find going back to X2 will
generate even more calls from customers.
Steve
>We are having the same problems and we are finding a number of
>complaints coming from customers with USR Sportsters.
>
>We have 4 Total Control boxes (3 with net servers and 1 Hiper
>ARC) and we have postponed our plans for buying another Hiper ARC
>because of these problems. The Quad modems with Net Servers are
>less problematic than the Hiper DSP's with Hiper ARC.
>
>Some customers get very slow connections or connect fast but
>can't surf. Many claim they have to dial in many times before
>they get a good connection, but eventually they do.
>
>We are running 4.1.11 and 1.2.5, basically TCS 3.3 for all
>components. Are some older versions better? Is there some Beta
>code we should be trying?
>
>The problem is so bad we are considering going back to just X2.
>We are also looking into other equipment, which would kill us
>since we had become so comfortable with Total Control...
>
>______________________________________________
>
>Mario M. Bustamante, President
>AccessPro Communications Inc.
>Miami, Florida
>
>
>> -----Original Message-----
>> From: owner-usr-tc@lists.xmission.com
>> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of K Mitchell
>> Sent: Friday, October 30, 1998 3:40 PM
>> To: usr-tc@lists.xmission.com
>> Subject: (usr-tc) v.90 connect problems
>>
>>
>> Since upgrading my DSP's to 1.2.5 I've been having a
>> number of customers
>> with problems connecting, from plain no-connects to x2
>> or x2/v90 modems
>> connecting at a slower rate than before. Otherwise on my HiPer;
>> ARC 4.0.30
>> NMC 5.5.5
>>
>> Some of the no-connects were LT Winmodems and the
>> -v90=0 string helped, the
>> others I'm waiting to hear back on modem type.
>> Does DSP 1.2.6x or ARC 4.1.x seem to be helping these
>> problems for anybody
>> else that's been experiencing them?
>>
>>
>> Kirk
>>
>>
>>
>>
>> Kirk Mitchell-General Manager sysadmin@keyconn.net
>> Keystone Connect http://www.keyconn.net
>> Altoona, PA 814-941-5000 We Unlock the World
>>
Subject:Re: (usr-tc) HiperDSP v1.2.68 fixed mysterious T1 problem From: Mike Andrews <mandrews@termfrost.org> Date: 1998-11-03 00:26:06
Maybe this is related to the "stuck channel" problem...? 1.2.68 DID seem
to fix that for us, even though it didn't fix the v.90 connection problems
everyone's reporting. It's a start. :)
Mike Andrews (MA12) icq 6602506 -------------- mandrews@termfrost.org
VP 'n' Systems/Network Administrator --------------- mandrews@dcr.net
Digital Crescent, Frankfort, KY ----------- http://www.termfrost.org/
# view;touch;unzip;finger;mount;mv;mv;mv;yes;mv;yes;mv;yes;umount;sleep
On Mon, 2 Nov 1998, Aaron Nabil wrote:
>
> Our CLEC was reporting errors on spans we had attached to HiperDSP cards.
>
> These errors didn't show up with a test set watching the span, but the
> switch saw a constant stream of them. It seemed to be directly dependant
> on the incoming call rate on the span. I'm guessing it was some margin problem
> the T-BERD was immune to but the switch wasn't.
>
> I was just about to try and get some fancy test gear to look at the
> T1 stream itself, but the CLEC said the problem went away. The only
> changes we made recently were swapping out some cards in our M31
> muxes, and the HiperDSP upgrade.
>
> So I downgraded one of the cards to 1.0.8. The error stream returned
> (with a vengance) on that span.
>
> So 3COM, whatever you did in 1.2.68, please don't undo it! It works!
I have problems with USR ISDN modems on my Hipers. When they bond two
channels the second channel goes up and down 3 times before finally
connecting, then to add to the humiliation the bonding causes a 1 to 3
minute delay in passing data. I assume it is a RIP related issue, but then
again a Falleron router works fine. I have no answers and soon will go with
Lucent or Cisco as originally intended. I don't have hours to play tech
support tag so I'll go with old faithful which works from the get-go. Even
though I never dealt with Krish I hear he is the only reason people stay
with 3com at all. Kudos to him for his ability to deal with all this
nonsense. I cannot myself.
At 06:41 PM 10/31/98 -0500, you wrote:
>Oh good, it's not just me. :) This has been driving me absolutely crazy
>for weeks now!
>
>We're running ARC 4.1.78 and DSP 1.2.68. As far as I can tell:
>
>* It only affects v.90 calls -- x2 and v.34 work fine.
>
>* It only affects our DSP/ARC rack, not our Quad/NETserver racks.
>
>* It's not specific to LT Winmodems, as I first thought. We have several
>Sportster owners complaining. I haven't yet tried it with a Courier --
>that's next week's project. :)
>
>Anyone have ANY hints here? We've lost a customer or two over this now.
>
>
>Mike Andrews (MA12) icq 6602506 -------------- mandrews@termfrost.org
>VP 'n' Systems/Network Administrator --------------- mandrews@dcr.net
>Digital Crescent, Frankfort, KY ----------- http://www.termfrost.org/
># view;touch;unzip;finger;mount;mv;mv;mv;yes;mv;yes;mv;yes;umount;sleep
>
>On Sat, 31 Oct 1998, Mario M. Bustamante wrote:
>
>> We are having the same problems and we are finding a number of
>> complaints coming from customers with USR Sportsters.
>>
>> We have 4 Total Control boxes (3 with net servers and 1 Hiper
>> ARC) and we have postponed our plans for buying another Hiper ARC
>> because of these problems. The Quad modems with Net Servers are
>> less problematic than the Hiper DSP's with Hiper ARC.
>>
>> Some customers get very slow connections or connect fast but
>> can't surf. Many claim they have to dial in many times before
>> they get a good connection, but eventually they do.
>>
>> We are running 4.1.11 and 1.2.5, basically TCS 3.3 for all
>> components. Are some older versions better? Is there some Beta
>> code we should be trying?
>>
>> The problem is so bad we are considering going back to just X2.
>> We are also looking into other equipment, which would kill us
>> since we had become so comfortable with Total Control...
>>
>> ______________________________________________
>>
>> Mario M. Bustamante, President
>> AccessPro Communications Inc.
>> Miami, Florida
>>
>>
>> > -----Original Message-----
>> > From: owner-usr-tc@lists.xmission.com
>> > [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of K Mitchell
>> > Sent: Friday, October 30, 1998 3:40 PM
>> > To: usr-tc@lists.xmission.com
>> > Subject: (usr-tc) v.90 connect problems
>> >
>> >
>> > Since upgrading my DSP's to 1.2.5 I've been having a
>> > number of customers
>> > with problems connecting, from plain no-connects to x2
>> > or x2/v90 modems
>> > connecting at a slower rate than before. Otherwise on my HiPer;
>> > ARC 4.0.30
>> > NMC 5.5.5
>> >
>> > Some of the no-connects were LT Winmodems and the
>> > -v90=0 string helped, the
>> > others I'm waiting to hear back on modem type.
>> > Does DSP 1.2.6x or ARC 4.1.x seem to be helping these
>> > problems for anybody
>> > else that's been experiencing them?
>> >
>> >
>> > Kirk
>> >
>> >
>> >
>> >
>> > Kirk Mitchell-General Manager sysadmin@keyconn.net
>> > Keystone Connect http://www.keyconn.net
>> > Altoona, PA 814-941-5000 We Unlock the World
>> >
>> >
>> > -
>> > To unsubscribe to usr-tc, send an email to
>> > "majordomo@xmission.com"
>> > with "unsubscribe usr-tc" in the body of the message.
>> > For information on digests or retrieving files and
>> > old messages send
>> > "help" to the same address. Do not use quotes in
>> > your message.
>> >
>>
>>
>> -
>> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
>> with "unsubscribe usr-tc" in the body of the message.
>> For information on digests or retrieving files and old messages send
>> "help" to the same address. Do not use quotes in your message.
>>
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
Al Thiel, MIS
Spec.Net Internet Services of New York
http://www.spec.net
"I'm workin' on it!"
> man, it's hard to figure out what list to post to. it's
> off-topic no matter where you post. is it a cisco? is it a
> 3com? is there a good place for as5100 questions?
I guess splitting the difference is the way to go ...
> anyrate, i'm trying to bring one up, CT1 that's already working
> with a pm3, moved it over, seeing tons of garbage when trying to
> dial in with a terminal emulator, connecting at 9600 bps.
OK, I'll let the 3COM side deal with this one :-) As I understand
it, the only cisco-supported TC hookup for the 5100 chassis is
POTS lines, not T1, so I'll try to dodge this ...
however, if you say "tons of garbage" - does this mean some
good chars and some bad? Or are you just getting PPP data coming
in? If some good chars and some bad, then it just sounds like
your modems aren't negotiating EC - so what happens if you
configure them to do LAP-M like they should be doing.
> relevant parts of config are below, if anyone could send me
> something sanitied, i'd greatly appreciate it.
> afaik, async-bootp dns-server xxx.xxx.xxx.xxx will send a
> nameserver out like everyone else does?
Yep, it will send the DNS address(es) via PPP IPCP to a dialup
client such as Windows DUN.
> i remember reading somewhere that a command is needed to prevent
> users from "demanding" an ip address instead of letting the as51
> card specify it?
Um, this depends on the niceties of the AAA authorization config
& whether Radius is assigning addresses or what have you. No simple
answer.
> thanks for your time, any suggestions, configs, or clues are
> appreciated
I'll make comments as I go thru ...
> line 1 16
> session-timeout 30 output
> location DNI Stamford POP
> autoselect during-login
> autoselect ppp
> absolute-timeout 720
> login authentication ppplogin
> modem Dialin
> transport input all
> stopbits 1
> speed 115200
> flowcontrol hardware
Looks just fine to me.
> interface Group-Async1
> ip unnumbered Ethernet0
> no ip directed-broadcast
> ip tcp header-compression passive
You may get better peformance removing the above.
> encapsulation ppp
> async mode dedicated
whoah. No wonder you got "garbage" when dialing in from
a terminal - you were seeing PPP packets. Change this to
"async mode interactive".
> no snmp trap link-status
> peer default ip address pool dni-as1-1
> no fair-queue
> no cdp enable
> ppp authentication pap chap
> ppp multilink
do you really want to use multilink PPP over async? If so
you'll want to have 11.3 and configure a virtual-template.
> group-range 1 16
Can't tell from the above enough to tell whether you will or won't
allow a client to dial in w/ their own IP address (all the AAA stuff
will also be relevant.) (Not that I'm familiar enough with the
ins and outs of AAA to figure it out anyway, but I know enough to
know that AAA *is* relevant.)
Cheers,
Aaron
Subject:Re: (usr-tc) as5100 - am i crazy? From: System Administrator <jamie-tc@atconnex.net> Date: 1998-11-03 14:48:04
At 11:33 AM 10/31/98 +1100, Bob Purdon wrote:
>
>> * its a serious bitch to setup (Im no idiot, and it took me 2 weeks)
>
>Hmmm, we had one a while back and it took me around 30 minutes to get all
>3 AS51 cards humming. Then again, we already used 2511's for dialup
>access at that time, so we had the basic configuration.
It wasn't the cisco gear that gave me the headache (I learned what I needed
of IOS in the first couple of days) it was the amazing forgetting NVRAM on
the Quad's.... hehehe... seems that if you don't tell them as many times as
it usually takes to get a 4 year old to follow instructions, they won't
listen.
Now I know better, if I am changing the Quad settings, I change them a bunch
of times to make sure they stick.
I ended up learning a lot more Cisco along the way and now await the day
when Livingston can't make a router to serve me well and I have to pay $$$$
for a Cisco ;)
J
==========================================
James Arlen
SysAdmin
At Connex Internet
www.atconnex.net
519-836-8777
man, it's hard to figure out what list to post to. it's
off-topic no matter where you post. is it a cisco? is it a
3com? is there a good place for as5100 questions?
anyrate, i'm trying to bring one up, CT1 that's already working
with a pm3, moved it over, seeing tons of garbage when trying to
dial in with a terminal emulator, connecting at 9600 bps.
relevant parts of config are below, if anyone could send me
something sanitied, i'd greatly appreciate it.
afaik, async-bootp dns-server xxx.xxx.xxx.xxx will send a
nameserver out like everyone else does?
i remember reading somewhere that a command is needed to prevent
users from "demanding" an ip address instead of letting the as51
card specify it?
thanks for your time, any suggestions, configs, or clues are
appreciated
line 1 16
session-timeout 30 output
location DNI Stamford POP
autoselect during-login
autoselect ppp
absolute-timeout 720
login authentication ppplogin
modem Dialin
transport input all
stopbits 1
speed 115200
flowcontrol hardware
interface Group-Async1
ip unnumbered Ethernet0
no ip directed-broadcast
ip tcp header-compression passive
encapsulation ppp
async mode dedicated
no snmp trap link-status
peer default ip address pool dni-as1-1
no fair-queue
no cdp enable
ppp authentication pap chap
ppp multilink
group-range 1 16
|o|----------------------------------------------------------------------|o|
|o| bryan s. blank (203)-351-1178 voice |o|
|o| senior systems analyst (203)-351-1186 fax |o|
|o| discovernet, incorporated (203)-979-5126 emerg |o|
I have to admit that I posted the following nugget w/out knowing
what I was talking about ...
Aaron
---
> I have to take exception to this. My stack of AS5100s is all CT1
> connected. I take the lines and float them among AS5100s and AS5300s
> freely.
> > OK, I'll let the 3COM side deal with this one :-) As I understand
> > it, the only cisco-supported TC hookup for the 5100 chassis is
> > POTS lines, not T1, so I'll try to dodge this ...
bryan s. blank was heard to say:
> man, it's hard to figure out what list to post to. it's
> off-topic no matter where you post. is it a cisco? is it a
> 3com? is there a good place for as5100 questions?
heh, it's both... but neither really own up to it.
--Ricky
I have to take exception to this. My stack of AS5100s is all CT1
connected. I take the lines and float them among AS5100s and AS5300s
freely.
On Tue, 3 Nov 1998, Aaron Leonard wrote:
> From: Aaron Leonard <Aaron@Cisco.COM>
> To: bryan s. blank <bryan@supernet.net>
> Date: Tue, 03 Nov 1998 14:22:39 -0700 (PDT)
> Subject: Re: netserver quad / as5100
>
> > man, it's hard to figure out what list to post to. it's
> > off-topic no matter where you post. is it a cisco? is it a
> > 3com? is there a good place for as5100 questions?
>
> I guess splitting the difference is the way to go ...
>
> > anyrate, i'm trying to bring one up, CT1 that's already working
> > with a pm3, moved it over, seeing tons of garbage when trying to
> > dial in with a terminal emulator, connecting at 9600 bps.
>
> OK, I'll let the 3COM side deal with this one :-) As I understand
> it, the only cisco-supported TC hookup for the 5100 chassis is
> POTS lines, not T1, so I'll try to dodge this ...
>
> however, if you say "tons of garbage" - does this mean some
> good chars and some bad? Or are you just getting PPP data coming
> in? If some good chars and some bad, then it just sounds like
> your modems aren't negotiating EC - so what happens if you
> configure them to do LAP-M like they should be doing.
>
> > relevant parts of config are below, if anyone could send me
> > something sanitied, i'd greatly appreciate it.
>
> > afaik, async-bootp dns-server xxx.xxx.xxx.xxx will send a
> > nameserver out like everyone else does?
>
> Yep, it will send the DNS address(es) via PPP IPCP to a dialup
> client such as Windows DUN.
>
> > i remember reading somewhere that a command is needed to prevent
> > users from "demanding" an ip address instead of letting the as51
> > card specify it?
>
> Um, this depends on the niceties of the AAA authorization config
> & whether Radius is assigning addresses or what have you. No simple
> answer.
>
> > thanks for your time, any suggestions, configs, or clues are
> > appreciated
>
> I'll make comments as I go thru ...
>
> > line 1 16
> > session-timeout 30 output
> > location DNI Stamford POP
> > autoselect during-login
> > autoselect ppp
> > absolute-timeout 720
> > login authentication ppplogin
> > modem Dialin
> > transport input all
> > stopbits 1
> > speed 115200
> > flowcontrol hardware
>
> Looks just fine to me.
>
> > interface Group-Async1
> > ip unnumbered Ethernet0
> > no ip directed-broadcast
> > ip tcp header-compression passive
>
> You may get better peformance removing the above.
>
> > encapsulation ppp
> > async mode dedicated
>
> whoah. No wonder you got "garbage" when dialing in from
> a terminal - you were seeing PPP packets. Change this to
> "async mode interactive".
>
> > no snmp trap link-status
> > peer default ip address pool dni-as1-1
> > no fair-queue
> > no cdp enable
> > ppp authentication pap chap
> > ppp multilink
>
> do you really want to use multilink PPP over async? If so
> you'll want to have 11.3 and configure a virtual-template.
>
> > group-range 1 16
>
> Can't tell from the above enough to tell whether you will or won't
> allow a client to dial in w/ their own IP address (all the AAA stuff
> will also be relevant.) (Not that I'm familiar enough with the
> ins and outs of AAA to figure it out anyway, but I know enough to
> know that AAA *is* relevant.)
>
> Cheers,
>
> Aaron
>
--
latzko@noc.rutgers.edu n2mlq
Alex Latzko +1-732-445-5021 (voice) +1-732-445-2968 (fax)
Rutgers University Computing Services TD:NOG Backbone Operations
110 Frelinghuysen Road, Piscataway, NJ 08855-8089
|o| however, if you say "tons of garbage" - does this mean some
|o| good chars and some bad? Or are you just getting PPP data coming
|o| in? If some good chars and some bad, then it just sounds like
|o| your modems aren't negotiating EC - so what happens if you
|o| configure them to do LAP-M like they should be doing.
i know what ppp looks like, and this ain't it.
connects don't happen past 9600 baud, and now i don't get
garbage anymore, it just hangs ...
i blame 3com ;)
anyrate, are there any docs on this thing, as said, nobody wants
to claim it ... the stuff on CCO really doesn't touch on the
integration of the two products as far as i can tell, and 3com
doesn't seem to know what it is.
thanks again!
|o|----------------------------------------------------------------------|o|
|o| bryan s. blank (203)-351-1178 voice |o|
|o| senior systems analyst (203)-351-1186 fax |o|
|o| discovernet, incorporated (203)-979-5126 emerg |o|
Subject:Re: (usr-tc) Terminating ISDN on digital quads From: Tatai SV Krishnan <tkrishna@bubba.ae.usr.com> Date: 1998-11-03 22:38:55
On Wed, 4 Nov 1998, Peter D. Mayer wrote:
> Jack,
>
> There are 2 ways to terminate ISDN calls on a TC box. One is to have the
> NETServer act as an ISDN gateway, the other is to have the calls be answered
> directly by the Digital Quad modems. The first works decently for us, but
> I've heard from various folks on this list that the second method may be
> better overall. That is what I'm asking about. I appreciate the response,
> but I'm looking a little deeper than simply connecting with ISDN.
>
You are right - terminating the isdn calls on the modems is better -
Now when you set the pri you can set up the quad modems as quad modems or
quad I modems. For the most part the pri does use the nmc chassiawarness
and configures them as quad-I-modems. At this time if you set the
default gateway slot of the pri card as 0 then the isdn calls will be
terminated on the modems. In your setup either you have the pri card set
to a gateway slot other than 0 or the card has not taken the setup or you
have the modems set as quad-modems.
Check this you need not reboot the pri card.
krish
> Thanks,
> Peter D. Mayer
> NetWalk System Administrator
> dmayer@netwalk.com
>
> -----Original Message-----
> From: Jack Singer <jsinger@usacars.com>
> To: usr-tc@lists.xmission.com <usr-tc@lists.xmission.com>
> Date: Wednesday, November 04, 1998 10:04 AM
> Subject: Re: (usr-tc) Terminating ISDN on digital quads
>
>
> If you have PRI-T1's then all the customer has to do is connect to you. You
> do
> not have to do anything since a PRI is already digital. We have many
> customers
> that connect via ISDN by doing nothing other than connecting to our regular
> dial
> up line.
>
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
On Tue, 3 Nov 1998, bryan s. blank wrote:
> |o| however, if you say "tons of garbage" - does this mean some
> |o| good chars and some bad? Or are you just getting PPP data coming
> |o| in? If some good chars and some bad, then it just sounds like
> |o| your modems aren't negotiating EC - so what happens if you
> |o| configure them to do LAP-M like they should be doing.
>
> i know what ppp looks like, and this ain't it.
>
> connects don't happen past 9600 baud, and now i don't get
> garbage anymore, it just hangs ...
>
How is the modem setup - Disconnect the cisco attach a pc behind the
modem and then dial into it to see how you are connecting. May be you
are sending a init string from the cisco to the modem. You do not need a
init string but some do send it.
The modems should be set to flow control. You may be forcing the modem
to connect at 9600. So check you config on the cisco. I would suggest
do not send any init string and if you must then send at&f1.
> i blame 3com ;)
:-) Really? Why for your config?
\ T.S.V. Krishnan \
\ Network System Engineer \ ( : - : )
\ 3Com ............ \
----------------------------------------------/
tkrishna@bubba.ae.usr.com
----------------------------/ http://interproc.ae.usr.com ----/
The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
Any Sufficiently advanced bug is indistinguishable for a feature.
- Rick Kulawiec
>
> anyrate, are there any docs on this thing, as said, nobody wants
> to claim it ... the stuff on CCO really doesn't touch on the
> integration of the two products as far as i can tell, and 3com
> doesn't seem to know what it is.
>
> thanks again!
>
> |o|----------------------------------------------------------------------|o|
> |o| bryan s. blank (203)-351-1178 voice |o|
> |o| senior systems analyst (203)-351-1186 fax |o|
> |o| discovernet, incorporated (203)-979-5126 emerg |o|
>
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) Hub status goes red From: Jack Singer <jsinger@usacars.com> Date: 1998-11-04 09:08:49
We have quad total control boxes that are v.90 upgraded and the hub status
light on some of the Network Management Cards turn red. We reboot them and
they are fine. Can anyone tell us what is wrong or needs to be changed.
Jack Singer
jsinger@i-c.net
Subject:(usr-tc) Terminating ISDN on digital quads From: Peter D. Mayer <dmayer@netwalk.com> Date: 1998-11-04 09:36:33
I finally got around to trying the termination of ISDN calls on the digital
quad modems. I set the GW-Slot on the PRI card to 0, and I thought that was
the only thing necessary. No luck. ISDN calls just bounce off. Do I need
to reset the PRI card? The NETServer? The whole chassis? Do I need to set
something differently in the quads or the NETServer? I have the latest
version of all code (no ER code though). Hardware revision 4 of the PRI
card, 3 of the digital quads, and 7 of the NETServer. I appreciate any
advice that you guys who have it working could give me.
Thanks,
Peter D. Mayer
NetWalk System Administrator
dmayer@netwalk.com
Subject:Re: (usr-tc) Terminating ISDN on digital quads From: Jack Singer <jsinger@usacars.com> Date: 1998-11-04 10:03:43
If you have PRI-T1's then all the customer has to do is connect to you. You do
not have to do anything since a PRI is already digital. We have many customers
that connect via ISDN by doing nothing other than connecting to our regular dial
up line.
Jack Singer
jsinger@i-c.net
Peter D. Mayer wrote:
> I finally got around to trying the termination of ISDN calls on the digital
> quad modems. I set the GW-Slot on the PRI card to 0, and I thought that was
> the only thing necessary. No luck. ISDN calls just bounce off. Do I need
> to reset the PRI card? The NETServer? The whole chassis? Do I need to set
> something differently in the quads or the NETServer? I have the latest
> version of all code (no ER code though). Hardware revision 4 of the PRI
> card, 3 of the digital quads, and 7 of the NETServer. I appreciate any
> advice that you guys who have it working could give me.
>
> Thanks,
> Peter D. Mayer
> NetWalk System Administrator
> dmayer@netwalk.com
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
Subject:Re: (usr-tc) Terminating ISDN on digital quads From: Peter D. Mayer <dmayer@netwalk.com> Date: 1998-11-04 10:33:17
Jack,
There are 2 ways to terminate ISDN calls on a TC box. One is to have the
NETServer act as an ISDN gateway, the other is to have the calls be answered
directly by the Digital Quad modems. The first works decently for us, but
I've heard from various folks on this list that the second method may be
better overall. That is what I'm asking about. I appreciate the response,
but I'm looking a little deeper than simply connecting with ISDN.
Thanks,
Peter D. Mayer
NetWalk System Administrator
dmayer@netwalk.com
-----Original Message-----
If you have PRI-T1's then all the customer has to do is connect to you. You
do
not have to do anything since a PRI is already digital. We have many
customers
that connect via ISDN by doing nothing other than connecting to our regular
dial
up line.
Subject:Re: (usr-tc) Hub status goes red From: Ken Hodges <ken@rabun.net> Date: 1998-11-04 10:40:50
I have had the same experience....
When you reset the NMC, it may stay green for a few minutes or a few days
before going red again... doesn't seem to affect any performance of the
machne, though... 3Com has been unable to fix this... I even had them
replace the card with the same results.
Ken
At 09:08 AM 11/4/98 -0500, you wrote:
>We have quad total control boxes that are v.90 upgraded and the hub status
>light on some of the Network Management Cards turn red. We reboot them and
>they are fine. Can anyone tell us what is wrong or needs to be changed.
>
>Jack Singer
>jsinger@i-c.net
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
_______________________________________
Ken Hodges, President and CEO
ACME BrainWorks Internet Services, Inc.
Clayton Computers, Inc.
Rabun County, Georgia
"Where Spring Spends the Summer"
http://www.acme-brain.com or http://www.rabun.net
ken@rabun.net 1-706-782-9239
Subject:Re: (usr-tc) Hub status goes red From: Andrew Aken <ajaken@globaleyes.net> Date: 1998-11-04 10:52:04
We have the same problem and it is caused by a bad fan in the fan tray.
Ken Hodges wrote:
>
> I have had the same experience....
> When you reset the NMC, it may stay green for a few minutes or a few days
> before going red again... doesn't seem to affect any performance of the
> machne, though... 3Com has been unable to fix this... I even had them
> replace the card with the same results.
>
> Ken
>
> At 09:08 AM 11/4/98 -0500, you wrote:
> >We have quad total control boxes that are v.90 upgraded and the hub status
> >light on some of the Network Management Cards turn red. We reboot them and
> >they are fine. Can anyone tell us what is wrong or needs to be changed.
> >
> >Jack Singer
> >jsinger@i-c.net
> >
> >
> >-
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
> >
> _______________________________________
> Ken Hodges, President and CEO
> ACME BrainWorks Internet Services, Inc.
> Clayton Computers, Inc.
> Rabun County, Georgia
> "Where Spring Spends the Summer"
> http://www.acme-brain.com or http://www.rabun.net
> ken@rabun.net 1-706-782-9239
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
--
=======================================================
=========== Andrew Aken - President =========
====== GlobalEyes Communications, Inc. ======
=Southern Illinois' Fastest Connection to the Internet=
========== http://www.GlobalEyes.net ========
=======================================================
Whats the most current code we can run on a netserver 16? This one is a
plain netserver 16, not the plus unit. Is there a version 4 that will run
on one? If so, is it stable or does it have problems?
______________________________________________________
Thanks,
Greg Coffey 307-234-5443 Fax 307-234-5446
CoffeyNet v.90 56k Access for Casper & Douglas
142 S. Center St.
Casper, WY 82601 http://www.coffey.com
Subject:Re: (usr-tc) Re: netserver quad / as5100 From: Charles Hill <chill@ionet.net> Date: 1998-11-04 12:04:35
If you can reverse telnet to the modem and do an at&f1 for hardware flow
control defaults and then an at&b0w to lock the port rate to the speed of
the connected port. . . be sure the port is not set to 9600. Change your
init string to atz or include &b0. That prevents the garbage caused by a
port speed mismatch. -CH
On Wed, 4 Nov 1998, Tatai SV Krishnan wrote:
> On Tue, 3 Nov 1998, bryan s. blank wrote:
>
> > |o| however, if you say "tons of garbage" - does this mean some
> > |o| good chars and some bad? Or are you just getting PPP data coming
> > |o| in? If some good chars and some bad, then it just sounds like
> > |o| your modems aren't negotiating EC - so what happens if you
> > |o| configure them to do LAP-M like they should be doing.
> >
> > i know what ppp looks like, and this ain't it.
> >
> > connects don't happen past 9600 baud, and now i don't get
> > garbage anymore, it just hangs ...
> >
> How is the modem setup - Disconnect the cisco attach a pc behind the
> modem and then dial into it to see how you are connecting. May be you
> are sending a init string from the cisco to the modem. You do not need a
> init string but some do send it.
>
> The modems should be set to flow control. You may be forcing the modem
> to connect at 9600. So check you config on the cisco. I would suggest
> do not send any init string and if you must then send at&f1.
>
>
>
> > i blame 3com ;)
>
> :-) Really? Why for your config?
>
> -----------------------------------------
> \ T.S.V. Krishnan \
> \ Network System Engineer \ ( : - : )
> \ 3Com ............ \
> ----------------------------------------------/
> tkrishna@bubba.ae.usr.com
> ----------------------------/ http://interproc.ae.usr.com ----/
> The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
> -------------------------------------------------------------------------\
> Any Sufficiently advanced bug is indistinguishable for a feature.
> - Rick Kulawiec
> -------------------------------------------------------------------------/
>
>
> >
> > anyrate, are there any docs on this thing, as said, nobody wants
> > to claim it ... the stuff on CCO really doesn't touch on the
> > integration of the two products as far as i can tell, and 3com
> > doesn't seem to know what it is.
> >
> > thanks again!
> >
> > |o|----------------------------------------------------------------------|o|
> > |o| bryan s. blank (203)-351-1178 voice |o|
> > |o| senior systems analyst (203)-351-1186 fax |o|
> > |o| discovernet, incorporated (203)-979-5126 emerg |o|
> >
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
|o| If you can reverse telnet to the modem and do an at&f1 for hardware flow
|o| control defaults and then an at&b0w to lock the port rate to the speed of
|o| the connected port. . . be sure the port is not set to 9600. Change your
|o| init string to atz or include &b0. That prevents the garbage caused by a
|o| port speed mismatch. -CH
thanks so much, i think i've got it halfway working - just gotta
keep messing w3ith aaa ;)
thanks!
|o|----------------------------------------------------------------------|o|
|o| bryan s. blank (203)-351-1178 voice |o|
|o| senior systems analyst (203)-351-1186 fax |o|
|o| discovernet, incorporated (203)-979-5126 emerg |o|
Subject:Re: (usr-tc) Hub status goes red From: Ricky Beam <jfbeam@interpath.net> Date: 1998-11-04 16:01:38
Ken Hodges was heard to say:
>I have had the same experience....
>When you reset the NMC, it may stay green for a few minutes or a few days
>before going red again... doesn't seem to affect any performance of the
>machne, though... 3Com has been unable to fix this... I even had them
>replace the card with the same results.
That LED is red for a reason... TCM ain't gonna tell you jack. The only
thing that will tell you what is wrong is the RADIUS data the NMC will
generate when it detects a problem.
I'm guessing there is a problem with a fan somewhere. I've had several TCs
do this as well. All of them did have something wrong.
3Com is just a little too happy to replace cards instead of actually trying
to find out what's wrong.
--Ricky
Subject:(usr-tc) NETServer Trade-Up PLUS program From: Kurtiss Johnson <kurtiss_johnson@mw.3com.com> Date: 1998-11-04 16:28:38
Hi all,
I thought you'd like to see the details of the new NETServer Trade-Up
program with the new ARC/DSP bundle. This announcement went out to our
resellers last Friday, and should go out to the general public later this
week or early next week.
Kurtiss Johnson
Product Manager - Access Routers
3Com Carrier Business Unit
847-222-2279
********************************************
NETServer Trade Up Plus Promotion
Program Details!
Upgrade your NETServer based Total Control chassis to
HiPer ARC and gain additional revenue generating multiservice
access features, 24 additional ports to support more customers
and generate more revenue and experience better performance
over your existing NETServer card. Introducing an exciting
new promotion for existing Total Control NETServer customers.
This trade up program offers current Total Control NETServer
customers with the opportunity to migrate to the latest in
State-of-the-Art Multiservice Access Technology.
Now, for less than the price of a single HiPer DSP card, when
you purchase the NETServer Trade Up bundle (SKU 003465-0), it's
like receiving a HiPer ARC at no additional cost. This exceptional
offer provides the increased performance and additional features of
HiPer ARC technology with an additional 24 ports. Estimated street
price is $169 per port.
List Price
Trade Up Bundle SKU includes: Component Bundle
002106-0 HiPer Access Router Card $ 9,995
002092-0 HiPer DSP Card $ 11,500
002453-0 NMC Memory Upgrade $ 298
Total $ 21,793 $10,500 less rebate
New Service Features!
With the recent release of TCS v3.3, HiPer ARC is currently
at feature parity with NETServer. In addition, HiPer ARC now
delivers numerous significant enhancements including:
MPIP - allows Multi-Link PPP across multiple chassis
IPX - supports Novell NetWare
IP Multicast - enables multicast services
VPNs via L2TP - simulates private connections over a public network
Enhanced SNMP management support - provides better polling features
These feature enhancements enable service providers to offer
more revenue generating services to their customers.
End-User Rebate!
The Trade Up program is only valid for purchases of a HiPer ARC router
card (002106-0) or the new Trade Up SKU (003465-0) through December 31,
1998.
In order to receive the $3,200 end-user rebate, you must return your Total
Control NETServer Card prior to February 26th, 1999.
More Information!
Learn all about the NETServer Trade Up Plus promotion by visiting our web
site at http://www.3com.com/promotions/hipertrade/
The site includes:
- FAQ to answer consumers' burning questions.
- Enhanced Feature-by-feature Comparison Document
- Instructional "How To" Document that walks you through all of the steps
necessary to replace a NETServer with a HiPer ARC, including both
physical
installation as well as configuration transition.
- Lease option available for less that $8 per port per month.
Better Support!
Dedicated Toll-Free CSO Call Center for the sole use of HiPer ARC Trade Up
customers. Their brand-new ARCSwitcher Configuration Translation Utility
can
translate 90% of the NETServer configuration to the HiPer ARC (The
ARCSwitcher
requires NETServer release v3.7 or later. Upgrades to v3.8 must be
downloaded
by December 31, 1998). For information call the Help desk at 888-373-7367.
On-site Field Service Replacement Option available for those customers who
would like 3Com field services personnel to perform their card exchange.
Subject:Re: (usr-tc) Hub status goes red From: David Bolen <db3l@ans.net> Date: 1998-11-04 16:38:19
Ken Hodges <ken@rabun.net> writes:
> I have had the same experience....
> When you reset the NMC, it may stay green for a few minutes or a few days
> before going red again... doesn't seem to affect any performance of the
> machne, though... 3Com has been unable to fix this... I even had them
> replace the card with the same results.
The red LED indicates some sort of failure in the chassis. In
general, this will be one of a few things (I'm not positive this list
is exhaustive but I think it covers the cases I've run into):
* An NMC self-test failure
* An environmental sensor failure (temperature, power supply, or
in the integrated fan tray chassis, rotational speed of one of
the larger fans)
* A management communication failure with a card in the chassis.
* In the newer clocked chassis, multiple failures of the chassis
clock.
I suppose it's possible to be in error, but I've never had that myself
in the past. Normally it's something other than the NMC, but NMC self
test failures might require a replacement (for example, I've had
specific management bus UARTs fail on an NMC).
Normally, the management communication path will be obvious since a
component will be in a failed operational state (I believe TCM shows
such cards in yellow, if I recall correctly).
The other cases can all be queried via SNMP. I'm not sure just where
in TCM each query might be, but expect that each table should be
available via some route. I'll show dumps below from our own tool
just for information.
The NMC self test and chassis temperature is part of the NMC status
table (nmcStat) - I've marked (**) entries that might cause the led:
NMC:
CompSwVer:"5.5.0", DramInstalled:16384(0x4000),
EventId:547(0x223), NVRAMInstalled:8192(0x2000),
**NmcPktBusClk:available, **PktBusClkSrc:backplaneActive,
PowerUpTstFailBMap:0, **PowerUpTstFailures:0,
**Status:ok, **Temperature:22(0x16),
TestResult:0
You can run a non-disruptive self test of most NMC items as well as an
NMC command. The nmcStatTestResult object will have the same
bitmapped result as that stored in nmcStatPowerUpTstFailBMap from the
power up tests.
The power supply information is in the chassis power supply tree
(uchasPowerSupply) and gives both current status of the supplies as
well as a count of warnings/failures since the NMC last rebooted:
Power Supply Status Information:
Supply Status Failures Description
1 good 0 USR Chassis PowerSupply 70 Amp AC #1
2 good 0 USR Chassis PowerSupply 70 Amp AC #2
Output Status Nominal Offered Warnings
1 good 5.00 5.21 0
2 good -5.00 -5.01 0
3 good 12.00 12.38 0
4 good -12.00 -12.10 0
The environmental information can be gathered from the chassis
environmental table (uchasEnviron) and includes fan and temperature
sensors as well as counts of warnings/failures since the NMC last
rebooted. For the older chassis, the fan is for the fan behind the
power supply - for the integrated fan tray chassis it includes the
larger fans in the fan tray (the smaller ones aren't instrumented):
Environmental Sensor Information:
Sensor Status Warnings Failures
Fans good 2 1
Temperature good 0 0
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
Subject:Re: (usr-tc) Hub status goes red From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-04 16:43:45
Thus spake David Bolen
> * An environmental sensor failure (temperature, power supply, or
> in the integrated fan tray chassis, rotational speed of one of
> the larger fans)
>The other cases can all be queried via SNMP. I'm not sure just where
>in TCM each query might be, but expect that each table should be
>available via some route. I'll show dumps below from our own tool
>just for information.
I recently had a fan die in one of mine...I could *not* seem to find
where this was shown in TCM anywhere...I'd be interested in finding out
where (if at all) TCM can grab this info.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:(usr-tc) Dead Air... From: Terry Kennedy <terry@olypen.com> Date: 1998-11-04 16:52:29
I know this was covered just recently, but refresh my mind,
which by the way is not working well anymore, as to what I need
to set. I think it was stated that the lines wre freeing up before
modems reset and were ready to take calls. The solution was round robin
instead of fixed assignment. Was this done at the Phone co level or in the
DSP's at the card level- routing method? If it is indeed done at the
card level should I see modems from from 1-24 or will they still show
modem 1, then modem 2, and so on? These are DSP's running 1.2.5 with
channelized T1's, no PRI. Any help here or just the key search on interproc
sure would help.
If this is something the CO has to set, what do I need to ask for?
Thanks----
TErry Kennedy
Subject:Re: (usr-tc) Hub status goes red From: Ricky Beam <jfbeam@interpath.net> Date: 1998-11-04 17:14:32
Jeff Mcadams was heard to say:
>I recently had a fan die in one of mine...I could *not* seem to find
>where this was shown in TCM anywhere...I'd be interested in finding out
>where (if at all) TCM can grab this info.
TCM doesn't "show" you fan failures or temperature alarms.
--Ricky
Subject:Re: (usr-tc) Hub status goes red From: David Bolen <db3l@ans.net> Date: 1998-11-04 17:45:33
Ricky Beam <jfbeam@Interpath.net> writes:
> TCM doesn't "show" you fan failures or temperature alarms.
Hmm - well, the temperature you could determine by just checking it.
Spec is no higher than 40C, but the alarm triggers a few degrees above
that.
I've got an older TCM copy, and it seems the NMC status (including
temperature) is in the "NMC Identification" section of the configured
parameters. That would be equivalent to my nmcStat table from my last
note.
However, it does appear (at least as of the TCM "dat" files up to NMC
4 - I haven't loaded v5 files) that nothing in any of those files
references the uchasEnviron table, so I guess it can't display the
environmental stuff. Unfortunate. Definitely a TCM issue though and not
with the NMC or management MIBs directly, so if you've got a
standalone SNMP query tool you could use that.
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
Subject:Re: (usr-tc) NETServer Trade-Up PLUS program From: Brian <signal@shreve.net> Date: 1998-11-04 20:11:38
>
> List Price
> Trade Up Bundle SKU includes: Component Bundle
> 002106-0 HiPer Access Router Card $ 9,995
> 002092-0 HiPer DSP Card $ 11,500
> 002453-0 NMC Memory Upgrade $ 298
>
> Total $ 21,793 $10,500 less rebate
>
First, let me say, I am very happy with the way 3com pricing is going, its
been going down for quite some time, every bundle seems better than the
last, very competitive pricing.
However, I find the trade in program a bit misleading. What this boils
down to, correct me if I am wrong is:
out of pocket of around $11k
1 ARC
1 HDM
1 NMC upgrade
When we are used to more like:
out of pocket $10k
1 chassis
2 HDM's
1 ARC
1 NMC (already upgraded)
1 power supply
1 cables, software, etc.
My last quote for the above bundle, form Solunet, was around <$9500. I
don't see where the trade-in program makes sense.
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:(usr-tc) Line Quality Logging in RADIUS ? From: Carl Litt <carl@snel.execulink.com> Date: 1998-11-04 22:12:20
We are running both Netserver/Quad Modem and HiPerARC/DSP hubs
and are plagued by disconnects (well, it isn't _too_ bad, but tell
that to the users).
Is there any way to get the Performance Monitor stats logged back
to RADIUS? There's lots of space in those Vendor-Specific
attributes which I never see used. I'm looking for Reason for
Termination, Link Block Errors, Retrains Requested/Granted, and
whatever other stats are useful. Reason for Termination and the
RADIUS attribute Acct-Terminate-Cause are not the same thing... the
first has way better information than the second. Some of those
"Analog Stats" look helpful too.
It would be extremely useful to have this stuff logged somewhere
so that we can go back for any connection and have a good look
at what caused the disconnect.
On this topic, I was configuring a Quad/NS hub last night. I was
dialed in and sitting close enough to kick it when it hung up on
me after only 5 minutes. I checked the RADIUS logs and they said
the Cause was a Lost-Carrier. However, the modem I was dialed
into reported "Retransmit Limit". What's that about? The
modem I dialed in with was a 3Com/Mhz LAN+Modem PCMCIA card
(which never gets higher than 45,333).
Carl Litt
Network Administrator
Execulink Internet Services Corporation
Subject:RE: (usr-tc) Terminating ISDN on digital quads From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-04 23:43:34
I didn't see anyone mention this so I will. You may want to verify the
settings under ISDN Call Control Options on the modems. They should be
correct, but no since in assuming that. check it.
=======================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Tatai SV Krishnan
Sent: Tuesday, November 03, 1998 10:39 PM
Cc: usr-tc@lists.xmission.com
On Wed, 4 Nov 1998, Peter D. Mayer wrote:
> Jack,
>
> There are 2 ways to terminate ISDN calls on a TC box. One is to have the
> NETServer act as an ISDN gateway, the other is to have the calls be
answered
> directly by the Digital Quad modems. The first works decently for us, but
> I've heard from various folks on this list that the second method may be
> better overall. That is what I'm asking about. I appreciate the
response,
> but I'm looking a little deeper than simply connecting with ISDN.
>
You are right - terminating the isdn calls on the modems is better -
Now when you set the pri you can set up the quad modems as quad modems or
quad I modems. For the most part the pri does use the nmc chassiawarness
and configures them as quad-I-modems. At this time if you set the
default gateway slot of the pri card as 0 then the isdn calls will be
terminated on the modems. In your setup either you have the pri card set
to a gateway slot other than 0 or the card has not taken the setup or you
have the modems set as quad-modems.
Check this you need not reboot the pri card.
krish
> Thanks,
> Peter D. Mayer
> NetWalk System Administrator
> dmayer@netwalk.com
>
> -----Original Message-----
> From: Jack Singer <jsinger@usacars.com>
> To: usr-tc@lists.xmission.com <usr-tc@lists.xmission.com>
> Date: Wednesday, November 04, 1998 10:04 AM
> Subject: Re: (usr-tc) Terminating ISDN on digital quads
>
>
> If you have PRI-T1's then all the customer has to do is connect to you.
You
> do
> not have to do anything since a PRI is already digital. We have many
> customers
> that connect via ISDN by doing nothing other than connecting to our
regular
> dial
> up line.
>
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:Re: (usr-tc) Hub status goes red From: Bob Purdon <bobp@southcom.com.au> Date: 1998-11-04 23:55:07
> I have had the same experience....
> When you reset the NMC, it may stay green for a few minutes or a few days
> before going red again... doesn't seem to affect any performance of the
> machne, though... 3Com has been unable to fix this... I even had them
> replace the card with the same results.
Run the Alarm Manager and configure the NMC to send traps to it - that
way it'll tell you why the hub status LED is red. In our case, it
thought the power supply fan was broken (it's not).
Regards,
Bob Purdon,
Technical Manager,
Southern Internet Services.
Subject:Re: (usr-tc) Terminating ISDN on digital quads From: Peter D. Mayer <dmayer@netwalk.com> Date: 1998-11-05 01:59:34
Which settings should I verify for this? Everything in there looks fairly
normal. What do you have for the settings?
Thanks,
Peter D. Mayer
NetWalk System Administrator
dmayer@netwalk.com
-----Original Message-----
I didn't see anyone mention this so I will. You may want to verify the
settings under ISDN Call Control Options on the modems. They should be
correct, but no since in assuming that. check it.
=======================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Tatai SV Krishnan
Sent: Tuesday, November 03, 1998 10:39 PM
Cc: usr-tc@lists.xmission.com
On Wed, 4 Nov 1998, Peter D. Mayer wrote:
> Jack,
>
> There are 2 ways to terminate ISDN calls on a TC box. One is to have the
> NETServer act as an ISDN gateway, the other is to have the calls be
answered
> directly by the Digital Quad modems. The first works decently for us, but
> I've heard from various folks on this list that the second method may be
> better overall. That is what I'm asking about. I appreciate the
response,
> but I'm looking a little deeper than simply connecting with ISDN.
>
You are right - terminating the isdn calls on the modems is better -
Now when you set the pri you can set up the quad modems as quad modems or
quad I modems. For the most part the pri does use the nmc chassiawarness
and configures them as quad-I-modems. At this time if you set the
default gateway slot of the pri card as 0 then the isdn calls will be
terminated on the modems. In your setup either you have the pri card set
to a gateway slot other than 0 or the card has not taken the setup or you
have the modems set as quad-modems.
Check this you need not reboot the pri card.
krish
Subject:Re: (usr-tc) NETServer Trade-Up PLUS program From: RJ Hwang <rj_hwang@mw.3com.com> Date: 1998-11-05 03:03:53
-
Subject:Re: (usr-tc) NETServer Trade-Up PLUS program From: Brian Elfert <brian@citilink.com> Date: 1998-11-05 08:28:49
On Wed, 4 Nov 1998, Brian wrote:
> >
> > List Price
> > Trade Up Bundle SKU includes: Component Bundle
> > 002106-0 HiPer Access Router Card $ 9,995
> > 002092-0 HiPer DSP Card $ 11,500
> > 002453-0 NMC Memory Upgrade $ 298
> >
> > Total $ 21,793 $10,500 less rebate
> >
>
> First, let me say, I am very happy with the way 3com pricing is going, its
> been going down for quite some time, every bundle seems better than the
> last, very competitive pricing.
>
>
> However, I find the trade in program a bit misleading. What this boils
> down to, correct me if I am wrong is:
>
> out of pocket of around $11k
> 1 ARC
> 1 HDM
> 1 NMC upgrade
The street price on this bundle is somewhere around $6400. With the $3200
rebate, you're basically getting the Hiper ARC for free.
Brian
Subject:Re: (usr-tc) NETServer Trade-Up PLUS program From: Brian Elfert <brian@citilink.com> Date: 1998-11-05 08:38:17
On Thu, 5 Nov 1998, Jeff Mcadams wrote:
> Again, I'll throw out that I (and many others) *really* would like to
> see a low-density low-cost alternative to the HiPer equipment. I still
> haven't seen an even halfway credible response to why 3Com isn't doing
> this. The NETServer hardware is there, it should be trivial to port the
> Pilgrim code to it, and there it is.
Why would lower density equipment cost less?
HiPer bundles with 48 ports cost around $10,000. Do you really think that
the older bundles like the 2059 or the 1706 could be sold for less if
still available? I'll bet it actually costs more to manufacture a 2059 or
1706 bundle because there are a lot more circuit boards involved.
Brian
Subject:Re: (usr-tc) NETServer Trade-Up PLUS program From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-05 09:00:56
Thus spake Kurtiss Johnson
>With the recent release of TCS v3.3, HiPer ARC is currently
>at feature parity with NETServer.
Gotta throw out a nit-pick here. From the netserver
"set radius_options prompt delay x"
There's no equivalent feature command set for the HARC from what I can
tell....its a pretty minor feature, but its the only thing that stands
between us and using the HiPer Arc instead of the NETServers. At this
point I have to keep coniving ways to get new NETServers since the HARC
doesn't support this. :/
Again, I'll throw out that I (and many others) *really* would like to
see a low-density low-cost alternative to the HiPer equipment. I still
haven't seen an even halfway credible response to why 3Com isn't doing
this. The NETServer hardware is there, it should be trivial to port the
Pilgrim code to it, and there it is.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) NETServer Trade-Up PLUS program From: Curt Shambeau <curt@execpc.com> Date: 1998-11-05 09:37:07
> > List Price
> > Trade Up Bundle SKU includes: Component Bundle
> > 002106-0 HiPer Access Router Card $ 9,995
> > 002092-0 HiPer DSP Card $ 11,500
> > 002453-0 NMC Memory Upgrade $ 298
> >
> > Total $ 21,793 $10,500 less rebate
> >
>
> First, let me say, I am very happy with the way 3com pricing is going, its
> been going down for quite some time, every bundle seems better than the
> last, very competitive pricing.
>
>
> However, I find the trade in program a bit misleading. What this boils
> down to, correct me if I am wrong is:
>
> out of pocket of around $11k
> 1 ARC
> 1 HDM
> 1 NMC upgrade
>
>
> When we are used to more like:
>
> out of pocket $10k
> 1 chassis
> 2 HDM's
> 1 ARC
> 1 NMC (already upgraded)
> 1 power supply
> 1 cables, software, etc.
>
> My last quote for the above bundle, form Solunet, was around <$9500. I
> don't see where the trade-in program makes sense.
Those chassis bundles were great, weren't they?
However, your math is flawed because you didn't find the street price of
the new bundle, which should be under $6500. So, you buy the bundle at
$6500, subtract the netserver trade in ($3200), and end up with a
HDM and ARC for around $3300. That's $137.50/port for the HDM, and the
ARC is then FREE! Still sounds like a deal to me!
Even though the chassis deal was great, and gave you more, it was still at
$200/port or so.
| Curtis V. Shambeau | curt@execpc.com | http://www.execpc.com/~curt |
| Executive Vice President - Exec-PC, Inc. |
Subject:Re: (usr-tc) NETServer Trade-Up PLUS program From: Kurtiss Johnson <kurtiss_johnson@mw.3com.com> Date: 1998-11-05 09:48:07
Hi Brian,
I'm not trying to be misleading, but maybe the statement needs
clarification.
The price I posted is LIST price. So take the list ($10,500) and get to
your typical street price from resellers, at whatever discounts you can.
Let's say your street price is about $6600US (I think this won't upset any
of our resellers with insinuated precedence :-) ).
Now, send in the NETServer and the proof of purchase for the bundle, and
we'll cut you a check for $3200 back (the term "less rebate" means subtract
any eligible rebates from the purchase price). This brings your pricing to
about $3400 for the bundle. That's about $141 per port, and includes the
router card and the NMC memory upgrade. Compare that to the bundle you
mentioned, at about $197 per port (granted, you get all the surrounding
equipment as well, but many of you have "surplus" chassis stacked up in
corners already, right?)
Does this make a little more sense?
Kurtiss Johnson
Product Manager - Access Routers
3Com Corporation - Carrier Business Unit
847-222-2279
Brian <signal@shreve.net> on 11/04/98 08:11:38 PM
Please respond to usr-tc@lists.xmission.com
cc: (Kurtiss Johnson/MW/US/3Com)
Note: Some recipients have been dropped due to syntax errors.
Please refer to the "$AdditionalHeaders" item for the complete headers.
>
> List Price
> Trade Up Bundle SKU includes: Component Bundle
> 002106-0 HiPer Access Router Card $ 9,995
> 002092-0 HiPer DSP Card $ 11,500
> 002453-0 NMC Memory Upgrade $ 298
>
> Total $ 21,793 $10,500 less rebate
>
First, let me say, I am very happy with the way 3com pricing is going, its
been going down for quite some time, every bundle seems better than the
last, very competitive pricing.
However, I find the trade in program a bit misleading. What this boils
down to, correct me if I am wrong is:
out of pocket of around $11k
1 ARC
1 HDM
1 NMC upgrade
When we are used to more like:
out of pocket $10k
1 chassis
2 HDM's
1 ARC
1 NMC (already upgraded)
1 power supply
1 cables, software, etc.
My last quote for the above bundle, form Solunet, was around <$9500. I
don't see where the trade-in program makes sense.
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:Re: (usr-tc) NETServer Trade-Up PLUS program From: Christina Raeber <christina_raeber@mw.3com.com> Date: 1998-11-05 09:58:42
If you have any questions regarding the NETServer Trade Up Program you can
call:
888-373-7367 or email HiperARCUpgrade@3com.com
Curt Shambeau <curt@execpc.com> on 11/05/98 09:37:07 AM
Please respond to usr-tc@lists.xmission.com
cc: (Christina Raeber/MW/US/3Com)
> > List Price
> > Trade Up Bundle SKU includes: Component Bundle
> > 002106-0 HiPer Access Router Card $ 9,995
> > 002092-0 HiPer DSP Card $ 11,500
> > 002453-0 NMC Memory Upgrade $ 298
> >
> > Total $ 21,793 $10,500 less
rebate
> >
>
> First, let me say, I am very happy with the way 3com pricing is going,
its
> been going down for quite some time, every bundle seems better than the
> last, very competitive pricing.
>
>
> However, I find the trade in program a bit misleading. What this boils
> down to, correct me if I am wrong is:
>
> out of pocket of around $11k
> 1 ARC
> 1 HDM
> 1 NMC upgrade
>
>
> When we are used to more like:
>
> out of pocket $10k
> 1 chassis
> 2 HDM's
> 1 ARC
> 1 NMC (already upgraded)
> 1 power supply
> 1 cables, software, etc.
>
> My last quote for the above bundle, form Solunet, was around <$9500. I
> don't see where the trade-in program makes sense.
Those chassis bundles were great, weren't they?
However, your math is flawed because you didn't find the street price of
the new bundle, which should be under $6500. So, you buy the bundle at
$6500, subtract the netserver trade in ($3200), and end up with a
HDM and ARC for around $3300. That's $137.50/port for the HDM, and the
ARC is then FREE! Still sounds like a deal to me!
Even though the chassis deal was great, and gave you more, it was still at
$200/port or so.
| Curtis V. Shambeau | curt@execpc.com | http://www.execpc.com/~curt |
| Executive Vice President - Exec-PC, Inc. |
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:Re: (usr-tc) NETServer Trade-Up PLUS program From: Kurtiss Johnson <kurtiss_johnson@mw.3com.com> Date: 1998-11-05 10:14:43
OK, Jeff, I'll grant you that the "TCS v3.3" statement is a general
blanket, and that there may be a few things that aren't there yet (though I
thought the RAD_OPT functionality was not one of them...hmmm....).
I'm going to -kinda- object to your concerns for a lower density product,
though. The determination of "density" is not handled by the router card,
except that the NETServer doesn't scale UP as much as the HiPer ARC. The
"density" is determined by the modem cards, by whether you're using HiPer
DSP or Quads and Dual Trunk cards. Agree so far? So what prevents you
from using a HiPer ARC in combination with Quads, rather than the
NETServer? OK, so if there's feature that's missing (we really ARE trying
to get everything in ARC, but something might have fallen through the
cracks) then let me know about it!
One of the things that WON'T be done, however, is porting the Pilgrim
software to NETServer, for both technical and business issues.
Kurtiss Johnson
Product Manager - Access Routers
3Com Corporation - Carrier Business Unit
847-222-2279
Jeff Mcadams <jeffm@iglou.com> on 11/05/98 08:00:56 AM
Please respond to usr-tc@lists.xmission.com
cc: (Kurtiss Johnson/MW/US/3Com)
Note: Some recipients have been dropped due to syntax errors.
Please refer to the "$AdditionalHeaders" item for the complete headers.
Thus spake Kurtiss Johnson
>With the recent release of TCS v3.3, HiPer ARC is currently
>at feature parity with NETServer.
Gotta throw out a nit-pick here. From the netserver
"set radius_options prompt delay x"
There's no equivalent feature command set for the HARC from what I can
tell....its a pretty minor feature, but its the only thing that stands
between us and using the HiPer Arc instead of the NETServers. At this
point I have to keep coniving ways to get new NETServers since the HARC
doesn't support this. :/
Again, I'll throw out that I (and many others) *really* would like to
see a low-density low-cost alternative to the HiPer equipment. I still
haven't seen an even halfway credible response to why 3Com isn't doing
this. The NETServer hardware is there, it should be trivial to port the
Pilgrim code to it, and there it is.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:Re: (usr-tc) NETServer Trade-Up PLUS program From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-05 10:22:49
Thus spake Brian Elfert
>Why would lower density equipment cost less?
You think a 486 chip costs as much as a PPC603?
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:(usr-tc) Result code group From: Aaron Nabil <nabil@spiritone.com> Date: 1998-11-05 11:12:22
What should "Result code group" be set on a quad talking to a HiperArc? The
default is 0, should it be 1?
--
Aaron Nabil
Subject:Re: (usr-tc) NETServer Trade-Up PLUS program From: Kevin Benton <s1kevin@tims.net> Date: 1998-11-05 11:48:33
On Thu, 5 Nov 1998, Kurtiss Johnson wrote:
> One of the things that WON'T be done, however, is porting the Pilgrim
> software to NETServer, for both technical and business issues.
Care to expound on that? Technical issues? Are you talking the ability
to implement things in software because of a lack of appropriate hardware
or what? What is your team trying to implement in software that the ARC
can do which the NetServer is physically incapable of?
Kevin Benton
Network Engineer
SOTA Technologies
E-Mail: s1kevin@tims.net
Web: http://users.sota-oh.com/~s1kevin/
Unsolicited advertisements processing fee: $50 subject to change without notice
Subject:Re: (usr-tc) NETServer Trade-Up PLUS program From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-05 12:42:19
Thus spake Kurtiss Johnson
>OK, Jeff, I'll grant you that the "TCS v3.3" statement is a general
>blanket, and that there may be a few things that aren't there yet (though I
>thought the RAD_OPT functionality was not one of them...hmmm....).
Hey, if its there, and I just missed it, wonderful! :) I certainly
couldn't find it though. :)
>I'm going to -kinda- object to your concerns for a lower density product,
>though. The determination of "density" is not handled by the router card,
>except that the NETServer doesn't scale UP as much as the HiPer ARC.
Eh, yeah, we're getting into semantics here, but since you can slap a
NETServer in there for each two DSP's and conceivably have that work the
same way, the NETServer can, at least *somewhat*, determine density.
That, of course, is a rather oddball config, but one that we're
considering (or course if you can find that rad_opt functionality then
away we can go with ARC's).
>The
>"density" is determined by the modem cards, by whether you're using HiPer
>DSP or Quads and Dual Trunk cards. Agree so far?
Not completely, see above. :)
>So what prevents you
>from using a HiPer ARC in combination with Quads, rather than the
>NETServer?
Well...I'm not really looking for a low-density solution
explicitly...although it can be nice to have your hunt group spread
across multiple chassis to prevent single-points of failure, and that is
indeed addressed by your thoughts here...but if I'm only going to be
putting 48-96 ports in a chassis, and NETServers are (or could be at
least...*should* be I would think) cheaper, then its gonna cost me less
to use the NETServer hardware in that chassis, and I'm not sacrificing
much of anything (assuming they're running Pilgrim code, which shouldn't
be too terribly difficult a challenge) and saving a chunk of change. :)
>OK, so if there's feature that's missing (we really ARE trying
>to get everything in ARC, but something might have fallen through the
>cracks) then let me know about it!
That rad_opt thingy is the only thing that I noticed...but then we don't
use all the features of the NETServer code, so there might be more
missing, but they won't affect us. Again, the rad_opt thing is pretty
minor, and I can't really blame you for missing it, but unfortunately,
its a minor thing that our setup kind of depends on. :)
>One of the things that WON'T be done, however, is porting the Pilgrim
>software to NETServer, for both technical and business issues.
Gack...another strike for 3Com then, IMHO. I just can't seem to come up
with *any* really good reason not to do this.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:(usr-tc) Latest ER code for HiPer From: Andrew Aken <ajaken@globaleyes.net> Date: 1998-11-05 13:09:50
What's the latest ER code for the HiPer ARC and DSP's? We've had a rash
of problems with disconnects and "no answer" lately from customers with
both the cheapo modems as well as the USR v.90 modems.
--
=======================================================
=========== Andrew Aken - President =========
====== GlobalEyes Communications, Inc. ======
=Southern Illinois' Fastest Connection to the Internet=
========== http://www.GlobalEyes.net ========
=======================================================
Subject:Re: (usr-tc) NETServer Trade-Up PLUS program From: Brian <signal@shreve.net> Date: 1998-11-05 13:24:54
ok, I get this now :)
On Thu, 5 Nov 1998, Kurtiss Johnson wrote:
> Hi Brian,
>
> I'm not trying to be misleading, but maybe the statement needs
> clarification.
>
> The price I posted is LIST price. So take the list ($10,500) and get to
> your typical street price from resellers, at whatever discounts you can.
> Let's say your street price is about $6600US (I think this won't upset any
> of our resellers with insinuated precedence :-) ).
>
> Now, send in the NETServer and the proof of purchase for the bundle, and
> we'll cut you a check for $3200 back (the term "less rebate" means subtract
> any eligible rebates from the purchase price). This brings your pricing to
> about $3400 for the bundle. That's about $141 per port, and includes the
> router card and the NMC memory upgrade. Compare that to the bundle you
> mentioned, at about $197 per port (granted, you get all the surrounding
> equipment as well, but many of you have "surplus" chassis stacked up in
> corners already, right?)
>
> Does this make a little more sense?
>
> Kurtiss Johnson
> Product Manager - Access Routers
> 3Com Corporation - Carrier Business Unit
> 847-222-2279
>
>
>
>
>
> Brian <signal@shreve.net> on 11/04/98 08:11:38 PM
>
> Please respond to usr-tc@lists.xmission.com
>
> To: usr-tc@lists.xmission.com
> cc: (Kurtiss Johnson/MW/US/3Com)
> Subject: Re: (usr-tc) NETServer Trade-Up PLUS program
>
>
>
>
> Note: Some recipients have been dropped due to syntax errors.
> Please refer to the "$AdditionalHeaders" item for the complete headers.
>
>
>
> >
> > List Price
> > Trade Up Bundle SKU includes: Component Bundle
> > 002106-0 HiPer Access Router Card $ 9,995
> > 002092-0 HiPer DSP Card $ 11,500
> > 002453-0 NMC Memory Upgrade $ 298
> >
> > Total $ 21,793 $10,500 less rebate
> >
>
> First, let me say, I am very happy with the way 3com pricing is going, its
> been going down for quite some time, every bundle seems better than the
> last, very competitive pricing.
>
>
> However, I find the trade in program a bit misleading. What this boils
> down to, correct me if I am wrong is:
>
> out of pocket of around $11k
> 1 ARC
> 1 HDM
> 1 NMC upgrade
>
>
> When we are used to more like:
>
> out of pocket $10k
> 1 chassis
> 2 HDM's
> 1 ARC
> 1 NMC (already upgraded)
> 1 power supply
> 1 cables, software, etc.
>
> My last quote for the above bundle, form Solunet, was around <$9500. I
> don't see where the trade-in program makes sense.
>
> --------------------------------------------------------------------------
> Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
>
>
>
>
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:Re: (usr-tc) BAD date on HiperARC From: sagarwal@crash.ae.usr.com Date: 1998-11-05 14:09:26
Type in the following command at Hiper Prompt.
Hiper> set date <date>
The format is dd-mmm-yyyy
Sanjay
On Thu, 5 Nov 1998, pferraro wrote:
>
> Could someone please tell us how to correct the date on our
> HiperARC??? Currently it say 21 Oct 2079. THe GMT looks fine. We enable
> the NTP several weeks back... The system has been REBOOTED several times
> and we still have the 2079 date!!!
>
> I want to resolves this... If anyone has the answer, please let me know
>
> ==============================================================================
> Phillip Ferraro WorldNet Access, Inc
> pferraro@wna-linknet.com Onslow County's PREMIER InterNet Service
> Voice (910) 346-0835 824 Gumbranch Square, Suite R3
> FAX (910) 455-1933 Jacksonville, Nc 28540-6269
> ==============================================================================
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Here's what a hiperarc does when you do a "reset modem". The amazing
this is how the modem decodes ATS0=0 (try "list init") into all of these
commands!
I'm thinking a reasonable strategy for settings modem NVRAM is to
do a restore from default, a "reset modem_group all" on the hiperarc,
and then a save to nvram.
hiperarc hiperdsp
sets to default
DTE Interface Settings
Modem Escape Char (S2) 255 43
Call Control Options
Verbal/Numeric Result Codes numeric verbal
Result Code Groups 0 1
ARQ Result Codes arqResultsDisabled arqResultsEnabled
Rings for Auto Answer 0 1
ATZ Handling over Packet Bus atZPbIgnore normalAtz
hiperarc quad modem
sets to default
DTE Interface Settings
Modem Escape Char (S2) 255 43
Call Control Options
Result Codes (Qn) displayResult noResult
Verbal/Numeric Result Codes numeric verbal
Result Code Groups 0 1
ARQ Result Codes arqResultsDisabled arqResultsEnabled
Response to +++ ignoreEscCode enterCommmandMode
Packet Bus Answer Only (S47.5) enable disable
--
Aaron Nabil
Subject:RE: (usr-tc) Dead Air... From: Terry Kennedy <terry@olypen.com> Date: 1998-11-05 14:35:07
Thanks on that one. One more, please...
Have the 1.2.68 code. Is this as simple as the quads for upgrading.
ie.. tftp-reboot ? will all settings remain or will they get lost
like the quads use to do?
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Randy Doran
> Sent: Thursday, November 05, 1998 1:05 PM
> To: usr-tc@lists.xmission.com
> Subject: Re: (usr-tc) Dead Air...
>
>
> I have set the HiPerDSP cards to round-robin and it seems to have cleared
> up a lot of the operator intercept error messages we were getting
> prior. I
> don't think it matters whether it is done on the card or at the CO, the
> results should be the same. It is just easier to to it in the
> card because
> then you will have control over it.
>
> Randy Doran
> Circuit Engineer \ Dialup Network Administrator
>
> e.spire Communication's CyberGate Internet Services
>
> Phone: 954-429-8069 FAX: 954-429-8001
>
> At 04:52 PM 11/4/98 -0800, you wrote:
> >I know this was covered just recently, but refresh my mind,
> >which by the way is not working well anymore, as to what I need
> >to set. I think it was stated that the lines wre freeing up before
> >modems reset and were ready to take calls. The solution was round robin
> >instead of fixed assignment. Was this done at the Phone co level
> or in the
> >DSP's at the card level- routing method? If it is indeed done at the
> >card level should I see modems from from 1-24 or will they still show
> >modem 1, then modem 2, and so on? These are DSP's running 1.2.5 with
> >channelized T1's, no PRI. Any help here or just the key search
> on interproc
> >sure would help.
> >
> >If this is something the CO has to set, what do I need to ask for?
> >
> >Thanks----
> >
> >TErry Kennedy
> >
> >-
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) BAD date on HiperARC From: pferraro <pferraro@wna-linknet.com> Date: 1998-11-05 14:56:39
Could someone please tell us how to correct the date on our
HiperARC??? Currently it say 21 Oct 2079. THe GMT looks fine. We enable
the NTP several weeks back... The system has been REBOOTED several times
and we still have the 2079 date!!!
I want to resolves this... If anyone has the answer, please let me know
==============================================================================
Phillip Ferraro WorldNet Access, Inc
pferraro@wna-linknet.com Onslow County's PREMIER InterNet Service
Voice (910) 346-0835 824 Gumbranch Square, Suite R3
FAX (910) 455-1933 Jacksonville, Nc 28540-6269
==============================================================================
Subject:(usr-tc) SNMP on HiperARC's From: Rick Payne <rickp@ops.netcom.net.uk> Date: 1998-11-05 16:03:48
Using standard snmp tools, I can snmpwalk many of the tables in the
HiperARC's mibs, but I cannot get individual elements.
eg.
lucan:(rickp)~> snmpwalk hiper usrMPIPLinkBundleOwner
usRobotics.common.usrMPIP.usrMPIPBase.usrMPIPLinkTable.usrMPIPLinkEntry.usrMPIPLinkBundleOwner.0 : IpAddress: 10.0.0.1
usRobotics.common.usrMPIP.usrMPIPBase.usrMPIPLinkTable.usrMPIPLinkEntry.usrMPIPLinkBundleOwner.1 : IpAddress: 10.0.0.2
etc.
yet:
lucan:(rickp)~> snmpget hiper usrMPIPLinkBundleOwner.0
snmpget: Agent reported an error.
snmpget: SNMP: Variable does not exist or access is denied.
Am I missing something, or is it that the HiperARC's just don't like
"get" and prefer "getnext".
This occurs in 4.1.78.
Cheers,
Rick
Subject:Re: (usr-tc) Dead Air... From: Randy Doran <randydoran@usa.net> Date: 1998-11-05 16:04:36
I have set the HiPerDSP cards to round-robin and it seems to have cleared
up a lot of the operator intercept error messages we were getting prior. I
don't think it matters whether it is done on the card or at the CO, the
results should be the same. It is just easier to to it in the card because
then you will have control over it.
Randy Doran
Circuit Engineer \ Dialup Network Administrator
e.spire Communication's CyberGate Internet Services
Phone: 954-429-8069 FAX: 954-429-8001
At 04:52 PM 11/4/98 -0800, you wrote:
>I know this was covered just recently, but refresh my mind,
>which by the way is not working well anymore, as to what I need
>to set. I think it was stated that the lines wre freeing up before
>modems reset and were ready to take calls. The solution was round robin
>instead of fixed assignment. Was this done at the Phone co level or in the
>DSP's at the card level- routing method? If it is indeed done at the
>card level should I see modems from from 1-24 or will they still show
>modem 1, then modem 2, and so on? These are DSP's running 1.2.5 with
>channelized T1's, no PRI. Any help here or just the key search on interproc
>sure would help.
>
>If this is something the CO has to set, what do I need to ask for?
>
>Thanks----
>
>TErry Kennedy
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) BAD date on HiperARC From: sagarwal@crash.ae.usr.com Date: 1998-11-05 16:53:39
Brian,
Which NTP issue are you refering to?
Sanjay
On Thu, 5 Nov 1998, Brian K McIntire wrote:
> The information Sanjay gave you will set the date correctly but does not
> address your issues with NTP. Sanjay, are there any current issues with
> using NTP on these HiPer ARC's. I experienced the same problems he's
> reporting. Maybe it's my version of software. Don't know.
>
> =======================================================================
> = Brian K. McIntire bmcintire@commnet.com =
> = Systems Engineer III 317-558-5050 x113 =
> = CommNetPlus, Indianapolis, IN http://www.commnet.com =
> = =
> = Experts in Remote Access and all areas of NetWork Design, =
> = Security, and Implementation. Offering both installation and =
> = support, along with remote management and monitoring packages for =
> = dedicated clients (specializing in ISP's). =
> = Firm partnerships established with all the key players. =
> = Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
> = =
> = Serving North America and parts of Canada. Reselling to the world. =
> =======================================================================
>
>
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of
> sagarwal@crash.ae.usr.com
> Sent: Thursday, November 05, 1998 2:09 PM
> To: pferraro
> Cc: Brian; usr-tc@lists.xmission.com
> Subject: Re: (usr-tc) BAD date on HiperARC
>
>
> Type in the following command at Hiper Prompt.
> Hiper> set date <date>
>
> The format is dd-mmm-yyyy
>
> Sanjay
>
> On Thu, 5 Nov 1998, pferraro wrote:
>
> >
> > Could someone please tell us how to correct the date on our
> > HiperARC??? Currently it say 21 Oct 2079. THe GMT looks fine. We enable
> > the NTP several weeks back... The system has been REBOOTED several times
> > and we still have the 2079 date!!!
> >
> > I want to resolves this... If anyone has the answer, please let me know
> >
> >
> ============================================================================
> ==
> > Phillip Ferraro WorldNet Access, Inc
> > pferraro@wna-linknet.com Onslow County's PREMIER InterNet Service
> > Voice (910) 346-0835 824 Gumbranch Square, Suite R3
> > FAX (910) 455-1933 Jacksonville, Nc 28540-6269
> >
> ============================================================================
> ==
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) Latest ER code for HiPer From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-05 17:07:39
The latest release code is 4.1.11 on the HiPer ARC and 1.2.5 on the DSP
The latest ER code i have received from 3COM is 1.2.68 on the HiPer DSP
(fixes dead random dead air, ring no answer, and strange tone problems) and
4.1.78 (fixes web-TV) on the HiPer ARC
There are always additions and/or changes to code so I'm sure there is
something more current out there but these seem to work great for us.
=======================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Andrew Aken
Sent: Thursday, November 05, 1998 1:10 PM
What's the latest ER code for the HiPer ARC and DSP's? We've had a rash
of problems with disconnects and "no answer" lately from customers with
both the cheapo modems as well as the USR v.90 modems.
--
=======================================================
=========== Andrew Aken - President =========
====== GlobalEyes Communications, Inc. ======
=Southern Illinois' Fastest Connection to the Internet=
========== http://www.GlobalEyes.net ========
=======================================================
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:RE: (usr-tc) BAD date on HiperARC From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-05 17:07:40
The information Sanjay gave you will set the date correctly but does not
address your issues with NTP. Sanjay, are there any current issues with
using NTP on these HiPer ARC's. I experienced the same problems he's
reporting. Maybe it's my version of software. Don't know.
=======================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of
sagarwal@crash.ae.usr.com
Sent: Thursday, November 05, 1998 2:09 PM
Cc: Brian; usr-tc@lists.xmission.com
Type in the following command at Hiper Prompt.
Hiper> set date <date>
The format is dd-mmm-yyyy
Sanjay
On Thu, 5 Nov 1998, pferraro wrote:
>
> Could someone please tell us how to correct the date on our
> HiperARC??? Currently it say 21 Oct 2079. THe GMT looks fine. We enable
> the NTP several weeks back... The system has been REBOOTED several times
> and we still have the 2079 date!!!
>
> I want to resolves this... If anyone has the answer, please let me know
>
>
============================================================================
==
> Phillip Ferraro WorldNet Access, Inc
> pferraro@wna-linknet.com Onslow County's PREMIER InterNet Service
> Voice (910) 346-0835 824 Gumbranch Square, Suite R3
> FAX (910) 455-1933 Jacksonville, Nc 28540-6269
>
============================================================================
==
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:RE: (usr-tc) Dead Air... From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-05 18:02:17
Make sure you have 5.5.5 on your NMC (This is very important. Restore from
default does not seem to work properly on HiPer DSP's with earlier revisions
of NMC code.)
The settings should remain, however, never assume that. Once you have
downloaded the new code do the following:
Click on the HiPer DSP card itself and go to configure action/commands and
go to configure/action commands and perform a Hardware reset. When the card
comes back up....
Click on the modem utilization bar and go to configure/action commands and
select all modems and hit OK. restore from default and save to nvram
Now click on the whole card again and return to Configure/action commands
and choose Software settings for the command to execute. restore template 1
config from default and execute. Save template 1 config to NVRAM and
execute. Change the command to execute back to Hardware and perform one
last hardware reset.
Some of the above may not be completely necessary, but this seems to be the
desired approach whenever upgrading HiPer DSP's.
Hope this helps.
=======================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Terry Kennedy
Sent: Thursday, November 05, 1998 4:35 PM
Thanks on that one. One more, please...
Have the 1.2.68 code. Is this as simple as the quads for upgrading.
ie.. tftp-reboot ? will all settings remain or will they get lost
like the quads use to do?
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Randy Doran
> Sent: Thursday, November 05, 1998 1:05 PM
> To: usr-tc@lists.xmission.com
> Subject: Re: (usr-tc) Dead Air...
>
>
> I have set the HiPerDSP cards to round-robin and it seems to have cleared
> up a lot of the operator intercept error messages we were getting
> prior. I
> don't think it matters whether it is done on the card or at the CO, the
> results should be the same. It is just easier to to it in the
> card because
> then you will have control over it.
>
> Randy Doran
> Circuit Engineer \ Dialup Network Administrator
>
> e.spire Communication's CyberGate Internet Services
>
> Phone: 954-429-8069 FAX: 954-429-8001
>
> At 04:52 PM 11/4/98 -0800, you wrote:
> >I know this was covered just recently, but refresh my mind,
> >which by the way is not working well anymore, as to what I need
> >to set. I think it was stated that the lines wre freeing up before
> >modems reset and were ready to take calls. The solution was round robin
> >instead of fixed assignment. Was this done at the Phone co level
> or in the
> >DSP's at the card level- routing method? If it is indeed done at the
> >card level should I see modems from from 1-24 or will they still show
> >modem 1, then modem 2, and so on? These are DSP's running 1.2.5 with
> >channelized T1's, no PRI. Any help here or just the key search
> on interproc
> >sure would help.
> >
> >If this is something the CO has to set, what do I need to ask for?
> >
> >Thanks----
> >
> >TErry Kennedy
> >
> >-
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:RE: (usr-tc) BAD date on HiPerARC From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-05 18:37:03
please read the message you responded to. Your answer is there.
=======================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of
sagarwal@crash.ae.usr.com
Sent: Thursday, November 05, 1998 4:54 PM
Brian,
Which NTP issue are you refering to?
Sanjay
On Thu, 5 Nov 1998, Brian K McIntire wrote:
> The information Sanjay gave you will set the date correctly but does not
> address your issues with NTP. Sanjay, are there any current issues with
> using NTP on these HiPer ARC's. I experienced the same problems he's
> reporting. Maybe it's my version of software. Don't know.
>
> =======================================================================
> = Brian K. McIntire bmcintire@commnet.com =
> = Systems Engineer III 317-558-5050 x113 =
> = CommNetPlus, Indianapolis, IN http://www.commnet.com =
> = =
> = Experts in Remote Access and all areas of NetWork Design, =
> = Security, and Implementation. Offering both installation and =
> = support, along with remote management and monitoring packages for =
> = dedicated clients (specializing in ISP's). =
> = Firm partnerships established with all the key players. =
> = Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
> = =
> = Serving North America and parts of Canada. Reselling to the world. =
> =======================================================================
>
>
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of
> sagarwal@crash.ae.usr.com
> Sent: Thursday, November 05, 1998 2:09 PM
> To: pferraro
> Cc: Brian; usr-tc@lists.xmission.com
> Subject: Re: (usr-tc) BAD date on HiperARC
>
>
> Type in the following command at Hiper Prompt.
> Hiper> set date <date>
>
> The format is dd-mmm-yyyy
>
> Sanjay
>
> On Thu, 5 Nov 1998, pferraro wrote:
>
> >
> > Could someone please tell us how to correct the date on our
> > HiperARC??? Currently it say 21 Oct 2079. THe GMT looks fine. We
enable
> > the NTP several weeks back... The system has been REBOOTED several times
> > and we still have the 2079 date!!!
> >
> > I want to resolves this... If anyone has the answer, please let me
know
> >
> >
>
============================================================================
> ==
> > Phillip Ferraro WorldNet Access, Inc
> > pferraro@wna-linknet.com Onslow County's PREMIER InterNet Service
> > Voice (910) 346-0835 824 Gumbranch Square, Suite R3
> > FAX (910) 455-1933 Jacksonville, Nc 28540-6269
> >
>
============================================================================
> ==
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
On Fri, 30 Oct 1998, K Mitchell wrote:
> Since upgrading my DSP's to 1.2.5 I've been having a number of customers
> with problems connecting, from plain no-connects to x2 or x2/v90 modems
> connecting at a slower rate than before. Otherwise on my HiPer;
> ARC 4.0.30
> NMC 5.5.5
>
> Some of the no-connects were LT Winmodems and the -v90=0 string helped, the
> others I'm waiting to hear back on modem type.
> Does DSP 1.2.6x or ARC 4.1.x seem to be helping these problems for anybody
> else that's been experiencing them?
OK, quite a few people, me included, have reported this problem, and DSP
rev 1.2.68 doesn't fix it. (It does fix other problems.) ARC 4.1.11 or
4.1.78 doesn't fix it either.
I've seen no real fixes that have worked so far. (Adding commas to the
end of the number hasn't helped anyone here.)
If this is a real bug, CAN SOMEONE (preferably at 3com) PLEASE AT LEAST
SAY SO, so I have SOMETHING to tell my users? Or if nothing else, so I
know what code to downgrade to? This is really becoming a major problem
for us...
Mike Andrews (MA12) icq 6602506 -------------- mandrews@termfrost.org
VP 'n' Systems/Network Administrator --------------- mandrews@dcr.net
Digital Crescent, Frankfort, KY ----------- http://www.termfrost.org/
# view;touch;unzip;finger;mount;mv;mv;mv;yes;mv;yes;mv;yes;umount;sleep
At 07:57 PM 11/5/98 -0500, Mike Andrews wrote:
>On Fri, 30 Oct 1998, K Mitchell wrote:
>
>> Since upgrading my DSP's to 1.2.5 I've been having a number of customers
>> with problems connecting, from plain no-connects to x2 or x2/v90 modems
>> connecting at a slower rate than before. Otherwise on my HiPer;
>> ARC 4.0.30
>> NMC 5.5.5
>>
>> Some of the no-connects were LT Winmodems and the -v90=0 string helped, the
>> others I'm waiting to hear back on modem type.
>> Does DSP 1.2.6x or ARC 4.1.x seem to be helping these problems for anybody
>> else that's been experiencing them?
>
>OK, quite a few people, me included, have reported this problem, and DSP
>rev 1.2.68 doesn't fix it. (It does fix other problems.) ARC 4.1.11 or
>4.1.78 doesn't fix it either.
>
>I've seen no real fixes that have worked so far. (Adding commas to the
>end of the number hasn't helped anyone here.)
>
>If this is a real bug, CAN SOMEONE (preferably at 3com) PLEASE AT LEAST
>SAY SO, so I have SOMETHING to tell my users? Or if nothing else, so I
>know what code to downgrade to? This is really becoming a major problem
>for us...
I've gone back to the 1.0.8 DSP code, and that seems to be working much
better, albeit without v.90. I left one DSP at 1.2.5 for those few Flex
users that actually connected better on it.
I'm still waiting for a workable answer also.........
Kirk Mitchell-General Manager sysadmin@keyconn.net
Keystone Connect http://www.keyconn.net
Altoona, PA 814-941-5000 We Unlock the World
>OK, quite a few people, me included, have reported this problem, and DSP
>rev 1.2.68 doesn't fix it. (It does fix other problems.) ARC 4.1.11 or
>4.1.78 doesn't fix it either.
>I've seen no real fixes that have worked so far. (Adding commas to the
>end of the number hasn't helped anyone here.)
>If this is a real bug, CAN SOMEONE (preferably at 3com) PLEASE AT LEAST
>SAY SO, so I have SOMETHING to tell my users? Or if nothing else, so I
>know what code to downgrade to? This is really becoming a major problem
>for us...
It is a real bug (actually bugS, quite a few of them). Most are on the
client side, as that is the most difficult piece to code. Most have been
resolved in later releases from the client chipset vendors, though not all
of the end-producers have released that to their end-users. Incrementally,
they are nailing most of them pretty good. As a case in point, the latest
LT Winmodem code is working well for many people from what I hear. I
highly recommend continuing to recommend that your customers keep up to
date on their client code.
We also found some things on our side we could do better that will either
accomodate some of the earlier code in the field and/or increase
performance as a whole. And, yup, there are a few bug fixes in there too
(including everything in 1.2.68). We are testing some of those V.90
improvements now, and hope to get a build out to limited field testing
within the next week or so.
1.2.68 actually did have some fixes in there for these problems, but they
were not significant. which is why I did not bill that ER as such. I
thought I was pretty clear on that, sorry if there was any misconception
there. I am glad to hear that it fixed some of your operational problems,
as that was its' primary intent.
Keep your eyes on this list, as I will definitely make a special point of
posting the availability of any maintenance code. I do appreciate the
patience, I understand the frustration you are experiencing.
JP
At 10:59 PM 11/5/98 -0600, John Powell wrote:
>It is a real bug (actually bugS, quite a few of them). Most are on the
>client side, as that is the most difficult piece to code. Most have been
>resolved in later releases from the client chipset vendors, though not all
>of the end-producers have released that to their end-users. Incrementally,
>they are nailing most of them pretty good. As a case in point, the latest
>LT Winmodem code is working well for many people from what I hear. I
>highly recommend continuing to recommend that your customers keep up to
>date on their client code.
>
>We also found some things on our side we could do better that will either
>accomodate some of the earlier code in the field and/or increase
>performance as a whole. And, yup, there are a few bug fixes in there too
>(including everything in 1.2.68). We are testing some of those V.90
>improvements now, and hope to get a build out to limited field testing
>within the next week or so.
>
>1.2.68 actually did have some fixes in there for these problems, but they
>were not significant. which is why I did not bill that ER as such. I
>thought I was pretty clear on that, sorry if there was any misconception
>there. I am glad to hear that it fixed some of your operational problems,
>as that was its' primary intent.
>
>Keep your eyes on this list, as I will definitely make a special point of
>posting the availability of any maintenance code. I do appreciate the
>patience, I understand the frustration you are experiencing.
Thanks for the update. As a side note, I saw a lot of complaints on the
4.1.11 ARC code, so haven't upgraded from 4.0.30 yet. Has 4.1.78 resolved
those issues for those that experienced them, or am I better sticking with
4.0.30 for now?
Kirk Mitchell-General Manager sysadmin@keyconn.net
Keystone Connect http://www.keyconn.net
Altoona, PA 814-941-5000 We Unlock the World
Subject:(usr-tc) Moving from Netserver to HiperARC problems From: Randy Cosby <dcosby@infowest.com> Date: 1998-11-06 10:52:59
Last night we had had what appeared to be a netserver death, so we replaced
it with a spare hiperarc we had here. All seems to be configured right, but
now we only get the first 3 lines on the first T1 answering. I'm sure I
won't get help with this on the tech support line. Are there any extra
steps that need to be taken when making this conversion? Yes, I'm running
the 16 meg NMC.
thanks,
Randy Cosby <dcosby@infowest.com>
Vice President
InfoWest Global Internet Services, Inc.
(435)674-0165 http://www.infowest.com
Subject:(usr-tc) Preventing the answering of a 128K ISDN call From: Eric J. Lorenzo <lorenzo@ispchannel.com> Date: 1998-11-06 11:05:21
How do you block in both a NETServer and a HARC the answering of a 128K
ISDN calls? We don't want two modems being taken up by one user.
Thanks,
Eric
Subject:RE: (usr-tc) Latest ER code for HiPer From: Terry Kennedy <terry@olypen.com> Date: 1998-11-06 11:20:45
For what it's worth... We have the dead air problem and seems to have gotten
much worst lately. We were running 2 seperate racks with 4 DSP's each and
all seemed OK. When I put the 5th Dsp into each rack the problem worsened.
They were all running 1.2.5 through channelized T1, not PRI. I have since
upgraded to 1.2.68, no help. Since I have 3 racks running now with 5,5, and
2 DSp's in them, I am considering moving 1 DSP out of each of the first 2
racks
so I would have 4,4 and 4 Dsp's in each rack. I don't even know if this will
help. The dead air problem can't be client side, can it? I can beleive
connection
problems could be either, but not the dead air.
> -----Original Message-----
> The latest ER code i have received from 3COM is 1.2.68 on the HiPer DSP
> (fixes dead random dead air, ring no answer, and strange tone
> problems) and
Subject:RE: (usr-tc) Latest ER code for HiPer From: Eric Forcey <eric@psnw.com> Date: 1998-11-06 12:04:37
This is the same problem that we were seeing, but in our case it was on
T1 cards.
Are you showing that the modems are getting hit when the user calls in?
What we were showing was nothing hitting the chassis after about 5-6
calls came in. Ended up being fixed in an ER release for the CT1 card,
however the Tech that finally came up with the solution of the ER release
said that an interim fix (and he did this until he got the ER code to me)
was to change attenJitterOnTxmtr to attenJitterOnRcvr, accept the
settings, then go back and change the setting back to TX.
Again, don't know if it is a possible related problem or not since the
platforms are different.
On Fri, 6 Nov 1998, John Powell wrote:
> >The dead air problem can't be client side, can it? I can beleive
> >connection problems could be either, but not the dead air.
>
> If you are talking about dead-air when the call is answered (ie, you call
> it and there is no answer tone), you are correct, that would NOT be a
> client problem (unless it's speaker is broken <grin>).
>
> Could be telco, but I doubt it. Have you been able to trap it in action,
> seen the call come via mgmt or otherwise appearing on the HDM?
>
> JP
>
>
>
>
>
> "Terry Kennedy" <terry@olypen.com> on 11/06/98 01:20:45 PM
>
> Please respond to usr-tc@lists.xmission.com
>
> To: usr-tc@lists.xmission.com
> cc: (John Powell/MW/US/3Com)
> Subject: RE: (usr-tc) Latest ER code for HiPer
>
>
>
>
> For what it's worth... We have the dead air problem and seems to have
> gotten
> much worst lately. We were running 2 seperate racks with 4 DSP's each and
> all seemed OK. When I put the 5th Dsp into each rack the problem worsened.
> They were all running 1.2.5 through channelized T1, not PRI. I have since
> upgraded to 1.2.68, no help. Since I have 3 racks running now with 5,5, and
> 2 DSp's in them, I am considering moving 1 DSP out of each of the first 2
> racks
> so I would have 4,4 and 4 Dsp's in each rack. I don't even know if this
> will
> help. The dead air problem can't be client side, can it? I can beleive
> connection
> problems could be either, but not the dead air.
>
>
> > -----Original Message-----
>
> > The latest ER code i have received from 3COM is 1.2.68 on the HiPer DSP
> > (fixes dead random dead air, ring no answer, and strange tone
> > problems) and
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
>
>
>
>
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) Latest ER code for HiPer From: Terry Kennedy <terry@olypen.com> Date: 1998-11-06 12:19:31
> Try setting the dsp's for round robin.
Been there - done that
> Did you restore from default on the
> modems and the templates, save both to nvram and reboot?
Been there - done that
> What do you have for the number of dtmftones
Can't find the setting, for one template 1 sets up mftones, not
dtmftones, used to know that on the quads, used to set it to 4.
Anis incoming and dnis incoming are set to zero - is that the same thing?
> Could be telco, but I doubt it. Have you been able to trap it in action,
> been the call come via mgmt or otherwise appearing on the HDM?
>Are you showing that the modems are getting hit when the user calls in?
On both these I can never catch it. If i pick up a hand or modem it will
happen
sometimes and the customers keep complaining, but Everytime I look for it
it's not
there. Modems answer fine the next call.
Subject:RE: (usr-tc) Latest ER code for HiPer From: Terry Kennedy <terry@olypen.com> Date: 1998-11-06 12:32:06
> What do you have for the number of dtmftones
Can't find the setting, for one template 1 sets up mftones, not
dtmftones, used to know that on the quads, used to set it to 4.
Anis incoming and dnis incoming are set to zero - is that the same thing?
Please ignore that last idiocy of mine. Was looking in the wrong place.
T1 settings dtmf tones = 4. Should it be set to something else?
Subject:Re: (usr-tc) Latest ER code for HiPer From: Aaron Nabil <nabil@spiritone.com> Date: 1998-11-06 13:22:40
Eric Forcey writes...
>What we were showing was nothing hitting the chassis after about 5-6
>calls came in. Ended up being fixed in an ER release for the CT1 card,
>however the Tech that finally came up with the solution of the ER release
>said that an interim fix (and he did this until he got the ER code to me)
>was to change attenJitterOnTxmtr to attenJitterOnRcvr, accept the
>settings, then go back and change the setting back to TX.
This sounds like a good candidate for my mysterious alarm problem coming
from HiPerDSPs that got fixed with the latest release.
It's significant that the old help file lists transmitter attenuation
as both the default and the suggested value for internally clocked
(instead of loop timed) spans.
I can almost guarantee every span in the know universe is looped timed,
so why isn't "attenuate jitter on receiver" the default?
-a
Subject:RE: (usr-tc) Moving from Netserver to HiperARC problems From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-06 13:40:41
What version of NMC code?
What version of HiPer ARC?
when you type list int what do you see?
When you type list chassis what do you see?
When you click on the modems and go to Configure/program settings/Line
Interface Options what do
you see for a line interface source.
When you list IP pool what do you see?
please be as detailed as possible when contacting this list. Other wise you
are just wasting e-mails and time.
=======================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Randy Cosby
Sent: Friday, November 06, 1998 11:53 AM
Last night we had had what appeared to be a netserver death, so we replaced
it with a spare hiperarc we had here. All seems to be configured right, but
now we only get the first 3 lines on the first T1 answering. I'm sure I
won't get help with this on the tech support line. Are there any extra
steps that need to be taken when making this conversion? Yes, I'm running
the 16 meg NMC.
thanks,
Randy Cosby <dcosby@infowest.com>
Vice President
InfoWest Global Internet Services, Inc.
(435)674-0165 http://www.infowest.com
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:RE: (usr-tc) Latest ER code for HiPer From: John Powell <john_powell@mw.3com.com> Date: 1998-11-06 13:50:11
>The dead air problem can't be client side, can it? I can beleive
>connection problems could be either, but not the dead air.
If you are talking about dead-air when the call is answered (ie, you call
it and there is no answer tone), you are correct, that would NOT be a
client problem (unless it's speaker is broken <grin>).
Could be telco, but I doubt it. Have you been able to trap it in action,
seen the call come via mgmt or otherwise appearing on the HDM?
JP
"Terry Kennedy" <terry@olypen.com> on 11/06/98 01:20:45 PM
Please respond to usr-tc@lists.xmission.com
cc: (John Powell/MW/US/3Com)
For what it's worth... We have the dead air problem and seems to have
gotten
much worst lately. We were running 2 seperate racks with 4 DSP's each and
all seemed OK. When I put the 5th Dsp into each rack the problem worsened.
They were all running 1.2.5 through channelized T1, not PRI. I have since
upgraded to 1.2.68, no help. Since I have 3 racks running now with 5,5, and
2 DSp's in them, I am considering moving 1 DSP out of each of the first 2
racks
so I would have 4,4 and 4 Dsp's in each rack. I don't even know if this
will
help. The dead air problem can't be client side, can it? I can beleive
connection
problems could be either, but not the dead air.
> -----Original Message-----
> The latest ER code i have received from 3COM is 1.2.68 on the HiPer DSP
> (fixes dead random dead air, ring no answer, and strange tone
> problems) and
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:RE: (usr-tc) Latest ER code for HiPer From: Terry Kennedy <terry@olypen.com> Date: 1998-11-06 14:02:46
Yeah, figured that out, thanks. It is set to 4.
The dtmftone setting I'm talking about is under the t1 trunk
settings. Not
on the modems themselves.
Subject:RE: (usr-tc) Latest ER code for HiPer From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-06 14:54:11
try setting the dsp's for round robin. Did you restore from default on the
modems and the templates, save both to nvram and reboot?
What do you have for the number of dtmftones
=======================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Terry Kennedy
Sent: Friday, November 06, 1998 1:21 PM
For what it's worth... We have the dead air problem and seems to have gotten
much worst lately. We were running 2 seperate racks with 4 DSP's each and
all seemed OK. When I put the 5th Dsp into each rack the problem worsened.
They were all running 1.2.5 through channelized T1, not PRI. I have since
upgraded to 1.2.68, no help. Since I have 3 racks running now with 5,5, and
2 DSp's in them, I am considering moving 1 DSP out of each of the first 2
racks
so I would have 4,4 and 4 Dsp's in each rack. I don't even know if this will
help. The dead air problem can't be client side, can it? I can beleive
connection
problems could be either, but not the dead air.
> -----Original Message-----
> The latest ER code i have received from 3COM is 1.2.68 on the HiPer DSP
> (fixes dead random dead air, ring no answer, and strange tone
> problems) and
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:RE: (usr-tc) Latest ER code for HiPer From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-06 16:54:42
The dtmftone setting I'm talking about is under the t1 trunk settings. Not
on the modems themselves.
=======================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Terry Kennedy
Sent: Friday, November 06, 1998 2:20 PM
> Try setting the dsp's for round robin.
Been there - done that
> Did you restore from default on the
> modems and the templates, save both to nvram and reboot?
Been there - done that
> What do you have for the number of dtmftones
Can't find the setting, for one template 1 sets up mftones, not
dtmftones, used to know that on the quads, used to set it to 4.
Anis incoming and dnis incoming are set to zero - is that the same thing?
> Could be telco, but I doubt it. Have you been able to trap it in action,
> been the call come via mgmt or otherwise appearing on the HDM?
>Are you showing that the modems are getting hit when the user calls in?
On both these I can never catch it. If i pick up a hand or modem it will
happen
sometimes and the customers keep complaining, but Everytime I look for it
it's not
there. Modems answer fine the next call.
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:(usr-tc) (TCS 3.3) MPIP between Netserver and HyperARC From: Mark Lemmert <cto@athenet.net> Date: 1998-11-06 17:20:27
A word to the wise for anybody upgrading to TCS 3.3:
TCS 3.3 is supposed to have MPIP capability between
HyperARCs and Netservers right out of the box. This
is not the case.
If anybody is trying to get this to work here is what
you need to do:
1) The HyperARC must be the mpip server, 3com support
staff has been saying the netserver will work as well.
That is not accurate.
2) You will need to get a patch from support, 4.1.72 -7
to be put on the HyperARC. That fixes some problems
with the hyperarc supporting some additional mpip extensions
on the netserver.
3) Use the command 'set ppp rec pap' if you haven't already
to restrict the ppp authentication protocol on the HyperARC
to pap, otherwise it will try chap first and the mpip sessions
will fail as the netserver uses pap and apparently
you must have a common authentication protocol for mpip.
Mark Lemmert cto@athenet.net
Chief Technical Officer (920)954-9799
AthEnet Data Exchange http://www.athenet.net
On Thu, 5 Nov 1998, John Powell wrote:
> >If this is a real bug, CAN SOMEONE (preferably at 3com) PLEASE AT LEAST
> >SAY SO, so I have SOMETHING to tell my users? Or if nothing else, so I
> >know what code to downgrade to? This is really becoming a major problem
> >for us...
>
> It is a real bug (actually bugS, quite a few of them). Most are on the
> client side, as that is the most difficult piece to code. Most have been
> resolved in later releases from the client chipset vendors, though not all
> of the end-producers have released that to their end-users. Incrementally,
> they are nailing most of them pretty good. As a case in point, the latest
> LT Winmodem code is working well for many people from what I hear. I
> highly recommend continuing to recommend that your customers keep up to
> date on their client code.
OK, but a fair number of the people having problems have Sportsters.
(Some Winmodems, some not.) I wasn't aware there was any new Sportster
code out... if there is, where's it hidden? If there isn't, is this one
of the server-side problems?
> We also found some things on our side we could do better that will either
> accomodate some of the earlier code in the field and/or increase
> performance as a whole. And, yup, there are a few bug fixes in there too
> (including everything in 1.2.68). We are testing some of those V.90
> improvements now, and hope to get a build out to limited field testing
> within the next week or so.
OK...
> 1.2.68 actually did have some fixes in there for these problems, but they
> were not significant. which is why I did not bill that ER as such. I
> thought I was pretty clear on that, sorry if there was any misconception
> there. I am glad to hear that it fixed some of your operational problems,
> as that was its' primary intent.
Actually, you did make it clear. :) I wasn't expecting 1.2.68 to fix the
v.90 problems, I was expecting it to fix the dead air problems... and it
seems to have. I'm just trying to get a handle on how much of this is
client-side and how much of this is server-side -- especially when the
user has a 3com modem, which not surprisingly have been the most trouble
free (until now).
Subject:Re: (usr-tc) (TCS 3.3) MPIP between Netserver and HyperARC From: Tatai SV Krishnan <tkrishna@bubba.ae.usr.com> Date: 1998-11-06 20:30:03
On Fri, 6 Nov 1998, Mark Lemmert wrote:
> A word to the wise for anybody upgrading to TCS 3.3:
>
> TCS 3.3 is supposed to have MPIP capability between
> HyperARCs and Netservers right out of the box. This
> is not the case.
>
> If anybody is trying to get this to work here is what
> you need to do:
>
>
> 1) The HyperARC must be the mpip server, 3com support
> staff has been saying the netserver will work as well.
> That is not accurate.
The HiPer arc must be the Server and no one deny's this. It was always
said and documented that the hiper arc should be server.
>
>
> 2) You will need to get a patch from support, 4.1.72 -7
> to be put on the HyperARC. That fixes some problems
> with the hyperarc supporting some additional mpip extensions
> on the netserver.
>
>
This is not correct. This is two early to say. You received this code
perhaps 2 or three days ago. We have 3 instances of problems with MPIP,
two problems with same symptoms and one different. You and one other
customer had this problem. Hiper arc does work with MPIP with NETServer
on 4.1.11 code - the only problem is with volume of calls and with time
and code that the NETServer is using results in a list of
non-disconnected MPIP links between the server and the client. If this
code does fix the problem - then good we can say atleast in your case
this code does fix the NETServer/Hiper arc issues. I would wait to get
info from the other two customers before declaring this.
> 3) Use the command 'set ppp rec pap' if you haven't already
> to restrict the ppp authentication protocol on the HyperARC
> to pap, otherwise it will try chap first and the mpip sessions
> will fail as the netserver uses pap and apparently
> you must have a common authentication protocol for mpip.
>
>
Bundles and authentication. By default the NETServer starts pap first
and hiper arc starts chap first, the second link depending upon the first
link should use the same protocol to autheticate else the bundle owner
will discard and disconnect ( prop. MPIP ) that is what you are seeing.
You are right - you have to use the same auth protocol between the
NETServer and the HiPer arc.
krish
>
> Mark Lemmert cto@athenet.net
> Chief Technical Officer (920)954-9799
> AthEnet Data Exchange http://www.athenet.net
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) (TCS 3.3) MPIP between Netserver and HyperARC From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-06 20:56:44
Thus spake Mark Lemmert
>If anybody is trying to get this to work here is what
>you need to do:
>1) The HyperARC must be the mpip server, 3com support
> staff has been saying the netserver will work as well.
> That is not accurate.
Hrmm...they were pretty up front about that in the documentation...I'm
surprised tech support is telling you that as it is clearly wrong.
Another strike for tech support. :/
>3) Use the command 'set ppp rec pap' if you haven't already
> to restrict the ppp authentication protocol on the HyperARC
> to pap, otherwise it will try chap first and the mpip sessions
> will fail as the netserver uses pap and apparently
> you must have a common authentication protocol for mpip.
Eh? That's funky. There's nothing in there that I saw that would
require the same userid authentication. It will require that the userid
be the same across the links in the bundle, as that's a requirement for
MP to bundle the links together, but it shouldn't need the same
authentication protocol that I'm aware of.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:(usr-tc) HiPer MRTG From: K Mitchell <mitch@keyconn.net> Date: 1998-11-07 01:15:33
Can anyone point me at some help in setting up MRTG to monitor modem useage?
Thanks,
Kirk
Kirk Mitchell-General Manager sysadmin@keyconn.net
Keystone Connect http://www.keyconn.net
Altoona, PA 814-941-5000 We Unlock the World
http://www.ac.net/modemuse not snmp but i like it.
-Dave
At 01:15 AM 11/7/98 -0500, you wrote:
>Can anyone point me at some help in setting up MRTG to monitor modem useage?
>
>Thanks,
>Kirk
>
>
>
>Kirk Mitchell-General Manager sysadmin@keyconn.net
>Keystone Connect http://www.keyconn.net
>Altoona, PA 814-941-5000 We Unlock the World
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
Subject:Re: (usr-tc) (TCS 3.3) MPIP between Netserver and HyperARC From: Bob Purdon <bobp@southcom.com.au> Date: 1998-11-07 08:27:52
> Another strike for tech support. :/
...and here in Oz we're just about to start experiencing 3COM US tech
support is seems. Our maintenance contract requires us to make an
*international* call to the USA in order to get anything done :-(
Regards,
Bob Purdon,
Technical Manager,
Southern Internet Services.
With pmwho (available at livingston's site) and hawho (from
"http://home.dwave.net/arctools/") you can gather a list of users on ports
and create a nice web page every 5 minutes, very useful.
Take that file and do a grep "ESTABLISHED" | wc -l to get the number of
connected users. With a small shell script you can make the mrtg file
needed to graph this. The file looks like this, I think. (total connected
users is 121 in this example)
0
121
0
0
Check the mrtg docs to be sure about the file structure.
At 01:34 AM 11/7/98 -0500, you wrote:
>http://www.ac.net/modemuse not snmp but i like it.
>-Dave
>
>At 01:15 AM 11/7/98 -0500, you wrote:
>>Can anyone point me at some help in setting up MRTG to monitor modem useage?
>>
>>Thanks,
>>Kirk
>>
>>
>>
>>Kirk Mitchell-General Manager sysadmin@keyconn.net
>>Keystone Connect http://www.keyconn.net
>>Altoona, PA 814-941-5000 We Unlock the World
>>
>>
>>-
>> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
>> with "unsubscribe usr-tc" in the body of the message.
>> For information on digests or retrieving files and old messages send
>> "help" to the same address. Do not use quotes in your message.
>>
>>
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
Richard B. Stuplich DataWave, Not just faster, Better.
President, DataWave Technologies 17 T1 circuits and growing strong.
DataWave, Wausau's first local ISP to have a direct connection to the
midwest NAP, Frame Relay, ISDN, x2, k56Flex, v.90. The only area ISP to
have redundant T1 NAP connections, All modem compatibility and a staff
dedicated exclusively to providing Internet Service to this area.
This E-Mail Copyright 1998 Richard Stuplich, see http://dw.net/cr/
http://www.dcr.net/~mandrews/usrtoys, at the very bottom of the page, has
some NETserver and ARC configs for MRTG...
Mike Andrews (MA12) icq 6602506 -------------- mandrews@termfrost.org
VP 'n' Systems/Network Administrator --------------- mandrews@dcr.net
Digital Crescent, Frankfort, KY ----------- http://www.termfrost.org/
# view;touch;unzip;finger;mount;mv;mv;mv;yes;mv;yes;mv;yes;umount;sleep
On Sat, 7 Nov 1998, David OBrien wrote:
> http://www.ac.net/modemuse not snmp but i like it.
> -Dave
>
> At 01:15 AM 11/7/98 -0500, you wrote:
> >Can anyone point me at some help in setting up MRTG to monitor modem useage?
> >
> >Thanks,
> >Kirk
> >
> >
> >
> >Kirk Mitchell-General Manager sysadmin@keyconn.net
> >Keystone Connect http://www.keyconn.net
> >Altoona, PA 814-941-5000 We Unlock the World
> >
> >
> >-
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
> >
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) Latest ER code for HiPer From: Terry Kennedy <terry@olypen.com> Date: 1998-11-07 11:03:37
Jeff, I have them set to round robin on the DSP's and still
have the problem. I understand what you are saying but then
if that's so shouldn't the problem be gone.
>
> Guys...this *isn't* a bug. The ds0 is reset'ing, the modem is
> reset'ing...just the ds0 is faster at it is all. If you use
> fixed-assignment on the DSP's with first-available from the telco, this
> is the risk you take and what you're going to run into. Either don't
> use first-available with the telco, or don't use fixed-assignment on the
> DSP's, that's all there is to it.
> --
> Jeff McAdams Email: jeffm@iglou.com
Subject:RE: (usr-tc) Latest ER code for HiPer From: Terry Kennedy <terry@olypen.com> Date: 1998-11-07 11:03:37
Jeff, I have them set to round robin on the DSP's and still
have the problem. I understand what you are saying but then
if that's so shouldn't the problem be gone.
>
> Guys...this *isn't* a bug. The ds0 is reset'ing, the modem is
> reset'ing...just the ds0 is faster at it is all. If you use
> fixed-assignment on the DSP's with first-available from the telco, this
> is the risk you take and what you're going to run into. Either don't
> use first-available with the telco, or don't use fixed-assignment on the
> DSP's, that's all there is to it.
> --
> Jeff McAdams Email: jeffm@iglou.com
Subject:RE: (usr-tc) Latest ER code for HiPer From: Brian <signal@shreve.net> Date: 1998-11-07 11:44:10
On Fri, 6 Nov 1998, Brian K McIntire wrote:
> try setting the dsp's for round robin. Did you restore from default on the
> modems and the templates, save both to nvram and reboot?
> What do you have for the number of dtmftones
>
I have seen dead-air on answer on our racks from time to time. I run
fixed assignment right now on the HDM's and may try Round-Robin to see if
this clears things up. Has 3Com acknowledged that their is an issue with
fixed assignment on the HDM's?
> =======================================================================
> = Brian K. McIntire bmcintire@commnet.com =
> = Systems Engineer III 317-558-5050 x113 =
> = CommNetPlus, Indianapolis, IN http://www.commnet.com =
> = =
> = Experts in Remote Access and all areas of NetWork Design, =
> = Security, and Implementation. Offering both installation and =
> = support, along with remote management and monitoring packages for =
> = dedicated clients (specializing in ISP's). =
> = Firm partnerships established with all the key players. =
> = Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
> = =
> = Serving North America and parts of Canada. Reselling to the world. =
> =======================================================================
>
>
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Terry Kennedy
> Sent: Friday, November 06, 1998 1:21 PM
> To: usr-tc@lists.xmission.com
> Subject: RE: (usr-tc) Latest ER code for HiPer
>
>
> For what it's worth... We have the dead air problem and seems to have gotten
> much worst lately. We were running 2 seperate racks with 4 DSP's each and
> all seemed OK. When I put the 5th Dsp into each rack the problem worsened.
> They were all running 1.2.5 through channelized T1, not PRI. I have since
> upgraded to 1.2.68, no help. Since I have 3 racks running now with 5,5, and
> 2 DSp's in them, I am considering moving 1 DSP out of each of the first 2
> racks
> so I would have 4,4 and 4 Dsp's in each rack. I don't even know if this will
> help. The dead air problem can't be client side, can it? I can beleive
> connection
> problems could be either, but not the dead air.
>
>
> > -----Original Message-----
>
> > The latest ER code i have received from 3COM is 1.2.68 on the HiPer DSP
> > (fixes dead random dead air, ring no answer, and strange tone
> > problems) and
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:RE: (usr-tc) Latest ER code for HiPer From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-07 12:01:01
Are you certain all your trunks are configured for 4 from the telco. In
other words, are they sending and DNIS digits
=======================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Terry Kennedy
Sent: Friday, November 06, 1998 4:03 PM
Yeah, figured that out, thanks. It is set to 4.
The dtmftone setting I'm talking about is under the t1 trunk
settings. Not
on the modems themselves.
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:Re: (usr-tc) Latest ER code for HiPer From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-07 13:55:03
Thus spake Brian
>On Fri, 6 Nov 1998, Brian K McIntire wrote:
>> try setting the dsp's for round robin. Did you restore from default on the
>> modems and the templates, save both to nvram and reboot?
>> What do you have for the number of dtmftones
>I have seen dead-air on answer on our racks from time to time. I run
>fixed assignment right now on the HDM's and may try Round-Robin to see if
>this clears things up. Has 3Com acknowledged that their is an issue with
>fixed assignment on the HDM's?
Guys...this *isn't* a bug. The ds0 is reset'ing, the modem is
reset'ing...just the ds0 is faster at it is all. If you use
fixed-assignment on the DSP's with first-available from the telco, this
is the risk you take and what you're going to run into. Either don't
use first-available with the telco, or don't use fixed-assignment on the
DSP's, that's all there is to it.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) Latest ER code for HiPer From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-07 14:44:30
Thus spake Terry Kennedy
>Jeff, I have them set to round robin on the DSP's and still
>have the problem. I understand what you are saying but then
>if that's so shouldn't the problem be gone.
Not necessarily...it could happen if there's only the one ds0 free on
the span...if it drops, it will almost immediately be filled again with
the next call...at this point (particularly with chan-t1's don't know
how the DSP's deal with pri's) there's only the one modem available, and
the one ds0, so you're in the same situation ultimately. The only
*real* solutions that could be come up with would be to speed up modem
resets (or skip them altogether? is that possible?), slow down ds0
reset's...someone mentioned getting the telco to do this, or possibly
putting the ds0 in localoutofservice after a call hangs up for a brief
period of time giving the modem time to reset and then putting the ds0
back inservice.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) HiPer MRTG From: Andre Gustavo de Carvalho Albuquerque <gustavo@anita.visualnet.com.br> Date: 1998-11-07 15:06:09
Richard Stuplich wrote:
:) With pmwho (available at livingston's site) and hawho (from
:) "http://home.dwave.net/arctools/") you can gather a list of users on ports
:) and create a nice web page every 5 minutes, very useful.
:)
:) Take that file and do a grep "ESTABLISHED" | wc -l to get the number of
:) connected users. With a small shell script you can make the mrtg file
:) needed to graph this. The file looks like this, I think. (total connected
:) users is 121 in this example)
:)
:) 0
:) 121
:) 0
:) 0
:)
:) Check the mrtg docs to be sure about the file structure.
:)
What about:
Target[hiper]: 1.3.6.1.4.1.429.4.2.1.10.0&1.3.6.1.4.1.429.4.2.1.10.0:community@host
It would be much more simple, isn't it?
Regards, Gustavo
--
____________________________________________________________________
Andre Gustavo de C. Albuquerque gustavo@visualnet.com.br
PGP Public Key: http://www.visualnet.com.br/~gustavo/pgpkey.asc
>OK, but a fair number of the people having problems have Sportsters.
>(Some Winmodems, some not.) I wasn't aware there was any new Sportster
>code out... if there is, where's it hidden? If there isn't, is this one
>of the server-side problems?
I was responding to your comments on LT Winmodems, you did not mention
Sportsters. I know of no significant issues with interop with Sportsters
(Winmodem or controller based) and Total Control modems.
There will likely be improvements with the client side (all vendors) for
quite some time. There are quite a few tweeks to enhance performance and
coverage in the labs already. Heck, we are still improving V.34 4 years
later.
If you can provide details on problems with Sportsters, I will gladly take
this off-line and put you in contact with my peers on the PCD (home of the
Sportster) side of the house.
JP
PS. No, there are no recent releases of Sportster code related to V.90
fixes. Overall, they are working quite well.
Subject:RE: (usr-tc) Latest ER code for HiPer From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-08 04:05:35
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jeff Mcadams
Sent: Saturday, November 07, 1998 1:45 PM
Thus spake Terry Kennedy
>Jeff, I have them set to round robin on the DSP's and still
>have the problem. I understand what you are saying but then
>if that's so shouldn't the problem be gone.
Not necessarily...it could happen if there's only the one ds0 free on
the span...if it drops, it will almost immediately be filled again with
the next call...at this point (particularly with chan-t1's don't know
how the DSP's deal with pri's) there's only the one modem available, and
the one ds0, so you're in the same situation ultimately. The only
*real* solutions that could be come up with would be to speed up modem
resets (or skip them altogether? is that possible?), slow down ds0
reset's...someone mentioned getting the telco to do this
I haven't verified it but the person that mentioned it said the telco sets
their DS0 reset to 10ms by default and indicated the reset could be anything
all the way up to 2500ms. I suppose adjusting this may help
, or possibly
putting the ds0 in localoutofservice after a call hangs up for a brief
period of time giving the modem time to reset and then putting the ds0
back inservice.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:Re: (usr-tc) Latest ER code for HiPer From: Terry Kennedy <terry@olypen.com> Date: 1998-11-08 10:01:18
Is there a way to busy out the dso for
a coupleof secounds? I do this with my analog gear.
It seems that 3com would have put this in the code somewhere
-----Original Message-----
>Thus spake Terry Kennedy
>>Jeff, I have them set to round robin on the DSP's and still
>>have the problem. I understand what you are saying but then
>>if that's so shouldn't the problem be gone.
>
>Not necessarily...it could happen if there's only the one ds0 free on
>the span...if it drops, it will almost immediately be filled again with
>the next call...at this point (particularly with chan-t1's don't know
>how the DSP's deal with pri's) there's only the one modem available, and
>the one ds0, so you're in the same situation ultimately. The only
>*real* solutions that could be come up with would be to speed up modem
>resets (or skip them altogether? is that possible?), slow down ds0
>reset's...someone mentioned getting the telco to do this, or possibly
>putting the ds0 in localoutofservice after a call hangs up for a brief
>period of time giving the modem time to reset and then putting the ds0
>back inservice.
>--
>Jeff McAdams Email: jeffm@iglou.com
>Head Network Administrator Voice: (502) 966-3848
>IgLou Internet Services (800) 436-4456
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) Latest ER code for HiPer From: Brian <signal@shreve.net> Date: 1998-11-08 15:27:30
On Sat, 7 Nov 1998, Jeff Mcadams wrote:
> Thus spake Brian
> >On Fri, 6 Nov 1998, Brian K McIntire wrote:
> >> try setting the dsp's for round robin. Did you restore from default on the
> >> modems and the templates, save both to nvram and reboot?
> >> What do you have for the number of dtmftones
>
> >I have seen dead-air on answer on our racks from time to time. I run
> >fixed assignment right now on the HDM's and may try Round-Robin to see if
> >this clears things up. Has 3Com acknowledged that their is an issue with
> >fixed assignment on the HDM's?
>
> Guys...this *isn't* a bug. The ds0 is reset'ing, the modem is
> reset'ing...just the ds0 is faster at it is all. If you use
> fixed-assignment on the DSP's with first-available from the telco, this
> is the risk you take and what you're going to run into. Either don't
> use first-available with the telco, or don't use fixed-assignment on the
> DSP's, that's all there is to it.
makes sense.
> --
> Jeff McAdams Email: jeffm@iglou.com
> Head Network Administrator Voice: (502) 966-3848
> IgLou Internet Services (800) 436-4456
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:(usr-tc) De-encrypting From: Ken Hodges <ken@rabun.net> Date: 1998-11-08 16:15:15
Does anybody know of a way to de-encrypt the passwords in the TC$A database ?
Any help would be MOST appreciated.
Thanks,
Ken
_______________________________________
Ken Hodges, President and CEO
ACME BrainWorks Internet Services, Inc.
Clayton Computers, Inc.
Rabun County, Georgia
"Where Spring Spends the Summer"
http://www.acme-brain.com or http://www.rabun.net
ken@rabun.net 1-706-782-9239
Subject:(usr-tc) HiperARC for sale From: G. Owens <gowens@seark.net> Date: 1998-11-08 17:00:19
We currently have a HiperARC chassis with NMC card, Hiper access router
card, and 70A power supply for sale...... $3,700.00 Also anyone know a
source for used DSP cards?
Greg Owens
Magnolia InterNet Services
gowens@magnolia-net.com
Subject:(usr-tc) HiPerDSP - SNMP Get: No such name (help needed) From: Andy <beezer@xmission.com> Date: 1998-11-09 00:14:08
USR-TC archive didn't seem to have anything on this, so I'll try here.
In short, I preconfigured a HiPerDSP on a HiPerDSP/ARC chassis then
shipped it to a remote POP for installation on a Quad/NETServer chassis
(something I was told could be done with the right NMC flash update). I
can hardware-reset the card, but receive "SNMP Get: No such name" when I
try to control it's Programmed Settings, Performance Monitor, etc;
although I can ping it. Reseating the HiPerDSP and resetting the NMC
doesn't help.
Pete Ashdown that knows this stuff forwards and back is unavailable this
week and I know only enough to be dangerous. This leaves me clueless and
with our POP saturated. Any timely help would be much appreciated.
Some details on the remote POP are:
Slot H/W version S/W version
------ ------------------------------ -------------- --------------
1 DUAL T1 NIC/NAC 3.0.0 3.5.0
2-13 Quad V.34 NIC/NAC 2.0.0 5.10.9
14 High-Density NIC/NAC 0.49.0 1.0.7
15 {empty}
16 3COM ISDN NETServer NAC 0.10.0 3.8.1
17 3COM NMC w/clock 3.0 5.4.95
18 {active supply #1}
19 {active supply #2}
Using TCM/UNIX 05.05.01
As a side note, 5 of the Quad modems (s2c2, s5c2, s10c2, s11c4, s12c4) are
not answering calls. I've compared their settings with functioning
modems, reflashed, reseated and reset them (hard/soft), but still they do
not answer. The telco says the trunks are in least-used distributed hunt
order, and I get slow busy with all but those 5 filled, so I'm unclear
what is causing this secondary problem and if it was a result of flashing
up to 5.4.95 on the NMC.
Andy Dalrymple
XMission Technical Support
---
Subject:Re: (usr-tc) Latest ER code for HiPer From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-09 08:08:09
Thus spake Terry Kennedy
>Is there a way to busy out the dso for
>a coupleof secounds? I do this with my analog gear.
>It seems that 3com would have put this in the code somewhere
Depends on the signaling code you use. If you're on chan-t1, there
should be a soft busyout that you can use. On PRI, you can set the
channels local out of service...that is unless you're using NI-2 which
doesn't support service messages. With fixed-assignment on the modems,
I suspect you can autoresponse this behavior into the DSP's...assuming
the DSP's support auto-response.
Again, we don't have any HiPer stuff here at all (one ARC demo to play
with actually, but that's it) so I don't know the config on these things
at all.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
We have a TC chassis with 10 DSP cards, 1 NMC and 1 ARC card. We would like
to add a second ARC card. The manual implies that there is a way to load
balance automatically the DSP cards between the 2 ARC cards.
Note the reference to DSA and the last sentence from this excerpt from the
manual:
In order to properly configure more than one HiPer ARC on the Hub, statically
configure your installed modem cards by setting their card type ownership or
dynamically configure the cards using Dynamic Slot Assignment (DSA). A third
method employs DSA rebalancing which periodically reassigns slot ownership by
the Network Management Card (NMC).
(end manual excerpt)
We understand how to assign ownership manually however it seems that that
the cards should be able to do this themselves.
Can the ARC cards be set so that they share the load and automatically
compensate
for the removal/insertion of one of the ARC cards?
If this is possible how are the ippools set on each of the ARC cards?
Does each card have :
a. 1/2(120) the pool?
b. the whole(240) pool?
c. total(240) yet different pools?
Anyone have experience with this configuration?
I have some scripts that do a snmpwalk of the chassis to
check the state of the DS0. I have them for the hiper
chassis and also for the dual PRI and T1.
The zip includes 4 files
TCH-PRI.pl for dual PRI config
TCH-T1.pl for dual T1 config
hiperdsp.pl for hiperdsp config
mrtg.cfg a snippet of the config for mrtg. Copy and paste this into
your mrtg.cfg file. Of course you will need to put in the correct
community and ip addresses.
http://mrtg.cableone.net Active graphs
http://mrtg.cableone.net/tch-mrtg.zip
Eric T. Billeter Cable One
Internet Engineer 1314 North 3rd Street
ebilleter@cableone.net Phoenix, AZ 85004
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of K Mitchell
Sent: Friday, November 06, 1998 11:16 PM
Can anyone point me at some help in setting up MRTG to monitor modem useage?
Thanks,
Kirk
Kirk Mitchell-General Manager sysadmin@keyconn.net
Keystone Connect http://www.keyconn.net
Altoona, PA 814-941-5000 We Unlock the World
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:(no subject) From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-09 10:17:57
Does anyone know how to get a copy of the Cistron RADIUS. I've heard a lot
of people talking about it but did not see if anyone said where to find it.
=======================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
Subject:RE: (usr-tc) RADIUS Cistron From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-09 11:13:20
My thanks to you and Bogdan Pelinescu.
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Cedric MARSOT
Sent: Monday, November 09, 1998 10:05 AM
Sorry ... this is the good link:
ftp://ftp.cistron.nl/pub/people/miquels/radius/
Cedric MARSOT - INTERVALLE Tel : +33 (0)1 49 72 58 58
Chef Produits d'Acces Distants Fax : +33 (0)1 49 72 58 59
Email: mailto:Cedric.Marsot@intervalle.fr Pager : +33 (0)6 06 49 14 43
Web: http://www.intervalle.fr Mobile: +33 (0)6 60 21 76 79
Pager: mailto:0606491443@kobby.worldnet.net
MCT NT WKS/SRV, Win95, Tcpip, Exchange, IIS
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian K McIntire
> Sent: Monday, November 09, 1998 5:18 PM
> To: usr tc
> Subject:
>
>
> Does anyone know how to get a copy of the Cistron RADIUS.
> I've heard a lot
> of people talking about it but did not see if anyone said
> where to find it.
>
> ==============================================================
> =========
> = Brian K. McIntire
> bmcintire@commnet.com =
> = Systems Engineer III
> 317-558-5050 x113 =
> = CommNetPlus, Indianapolis, IN
http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:(usr-tc) NT user and bundling MP From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-09 11:47:08
OK...I know I've seen the answer to this somewhere, but can't find it
(even after looking through the archives).
Have a customer using NT dialing in with 2 channels...The problem being
that NT (at least with the Sportster modem he's using) dials both
channels at the same time, and then tries to bundle them together. Just
about every other piece of dial-up equipment/software I've seen with
dial the first channel, establish the bundle, then dial the second
channel. My understanding is that there is a way to make NT do this
(ie, not dial both channels at once), but I don't know NT enough to tell
my customer where to go to set this.
Anyone have any ideas on this?
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
On Monday, November 09, 1998 12:18 PM, Brian K McIntire
[SMTP:bmcintire@commnet.com] wrote:
> Does anyone know how to get a copy of the Cistron RADIUS. I've heard a
lot
> of people talking about it but did not see if anyone said where to find
it.
>
http://miquels.www.cistron.nl/radius/
Be Seeing You...
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Matthew Stainforth<>Technical Services Manager<>BrunNet Inc.<>(506)450-4562
A quart of wheat for a day's wages, and three quarts of barley for a day's
wages, and do not damage the oil and the wine!
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Subject:Re: (usr-tc) HiPerDSP - SNMP Get: No such name (help needed) From: Mike Andrews <mandrews@termfrost.org> Date: 1998-11-09 13:29:10
You have the wrong NMC code version.
5.4.95 is for 4 meg cards. If the middle number of the release is even,
it's a 4 meg release that can't manage HiPer cards. If the middle number
is odd, it's a 16 meg release that can.
You need release 5.5.5, and you will also need 16 meg of RAM and 8 meg of
flash on your NMC to run it. If your NMC only has 4 meg, you either need
to upgrade it, swap in another NMC that has more memory, or remove the
DSP.
The only thing I can suggest on the Quads is to busy out the DS0's on the
T1's and then unbusy them again. Sometimes this helps, at least on our
PRI-from-DMS100 setup...
Mike Andrews (MA12) icq 6602506 -------------- mandrews@termfrost.org
VP 'n' Systems/Network Administrator --------------- mandrews@dcr.net
Digital Crescent, Frankfort, KY ----------- http://www.termfrost.org/
# view;touch;unzip;finger;mount;mv;mv;mv;yes;mv;yes;mv;yes;umount;sleep
On Mon, 9 Nov 1998, Andy wrote:
>
> USR-TC archive didn't seem to have anything on this, so I'll try here.
>
>
> In short, I preconfigured a HiPerDSP on a HiPerDSP/ARC chassis then
> shipped it to a remote POP for installation on a Quad/NETServer chassis
> (something I was told could be done with the right NMC flash update). I
> can hardware-reset the card, but receive "SNMP Get: No such name" when I
> try to control it's Programmed Settings, Performance Monitor, etc;
> although I can ping it. Reseating the HiPerDSP and resetting the NMC
> doesn't help.
>
> Pete Ashdown that knows this stuff forwards and back is unavailable this
> week and I know only enough to be dangerous. This leaves me clueless and
> with our POP saturated. Any timely help would be much appreciated.
>
>
> Some details on the remote POP are:
>
> Slot H/W version S/W version
> ------ ------------------------------ -------------- --------------
> 1 DUAL T1 NIC/NAC 3.0.0 3.5.0
> 2-13 Quad V.34 NIC/NAC 2.0.0 5.10.9
> 14 High-Density NIC/NAC 0.49.0 1.0.7
> 15 {empty}
> 16 3COM ISDN NETServer NAC 0.10.0 3.8.1
> 17 3COM NMC w/clock 3.0 5.4.95
> 18 {active supply #1}
> 19 {active supply #2}
>
> Using TCM/UNIX 05.05.01
>
> As a side note, 5 of the Quad modems (s2c2, s5c2, s10c2, s11c4, s12c4) are
> not answering calls. I've compared their settings with functioning
> modems, reflashed, reseated and reset them (hard/soft), but still they do
> not answer. The telco says the trunks are in least-used distributed hunt
> order, and I get slow busy with all but those 5 filled, so I'm unclear
> what is causing this secondary problem and if it was a result of flashing
> up to 5.4.95 on the NMC.
>
>
> Andy Dalrymple
> XMission Technical Support
Subject:Re: (usr-tc) Latest ER code for HiPer From: Mike Andrews <mandrews@termfrost.org> Date: 1998-11-09 13:31:39
On Sat, 7 Nov 1998, Jeff Mcadams wrote:
> Thus spake Brian
> >On Fri, 6 Nov 1998, Brian K McIntire wrote:
> >> try setting the dsp's for round robin. Did you restore from default on the
> >> modems and the templates, save both to nvram and reboot?
> >> What do you have for the number of dtmftones
>
> >I have seen dead-air on answer on our racks from time to time. I run
> >fixed assignment right now on the HDM's and may try Round-Robin to see if
> >this clears things up. Has 3Com acknowledged that their is an issue with
> >fixed assignment on the HDM's?
>
> Guys...this *isn't* a bug. The ds0 is reset'ing, the modem is
> reset'ing...just the ds0 is faster at it is all. If you use
> fixed-assignment on the DSP's with first-available from the telco, this
> is the risk you take and what you're going to run into. Either don't
> use first-available with the telco, or don't use fixed-assignment on the
> DSP's, that's all there is to it.
...except that several people said that round-robin assignment was broken
on DSP's and only fixed-assignment worked correctly. This was probably
fixed in 1.2.68...
Mike Andrews (MA12) icq 6602506 -------------- mandrews@termfrost.org
VP 'n' Systems/Network Administrator --------------- mandrews@dcr.net
Digital Crescent, Frankfort, KY ----------- http://www.termfrost.org/
# view;touch;unzip;finger;mount;mv;mv;mv;yes;mv;yes;mv;yes;umount;sleep
Subject:(usr-tc) T1 parameters From: Richard Lorbieski <richard@mail.alpha1.net> Date: 1998-11-09 14:28:53
We plan to install a T1 (not PRI) in one of our pops. What are the
recomended settings for the DSP/Hiper ARC chassis (line coding, start
signal, etc...)? I lost the paper 3com sent me a few months ago.
Richard Lorbieski
> We plan to install a T1 (not PRI) in one of our pops. What are the
> recomended settings for the DSP/Hiper ARC chassis (line coding, start
> signal, etc...)? I lost the paper 3com sent me a few months ago.
ESF/B8ZS on the T1. Make sure it is Trunk Side, not Line side if you get
stuck with D4/AMI. We use wink start.
| Curtis V. Shambeau | curt@execpc.com | http://www.execpc.com/~curt |
| Executive Vice President - Exec-PC, Inc. |
Subject:(usr-tc) Netserver 16 and VSA - URGENT HELP NEEDED From: Bogdan Pelinescu <bpelin@itcnet.ro> Date: 1998-11-09 15:53:20
Can anyone tell me why a Netserver 16, running 3.3.0 software
doesn't send radius VSA ? Do I have to set something extra ?
I REALLY need VSA support.
I used Mike Wronski's radddebug to see the radius packets and I
was a little bit shocked by the result.
Below it's a sample of a start and stop packets:
Code :Accounting-Request(4) Identifier :223 Length :88
Acct-Session-Id( 44) Len: 10 Value: 0400287D
User-Name( 1) Len: 10 Value: xxxx
Client-Id( 4) Len: 6 Value: xx.xx.xx.xx
Client-Port-Id( 5) Len: 6 Value: 7
Acct-Status-Type( 40) Len: 6 Value: Start(1)
Acct-Authentic( 45) Len: 6 Value: RADIUS(1)
User-Service-Type( 6) Len: 6 Value: Framed-User(2)
Framed-Protocol( 7) Len: 6 Value: PPP(1)
Framed-Address( 8) Len: 6 Value: xx.xx.xx.xx
Acct-Delay-Time( 41) Len: 6 Value: 0
Code :Accounting-Request(4) Identifier :221 Length :100
Acct-Session-Id( 44) Len: 10 Value: 0400287C
User-Name( 1) Len: 10 Value: xxxx
Client-Id( 4) Len: 6 Value: xx.xx.xx.xx
Client-Port-Id( 5) Len: 6 Value: 7
Acct-Status-Type( 40) Len: 6 Value: Stop(2)
Acct-Session-Time( 46) Len: 6 Value: 73
Acct-Terminate-Cause( 49) Len: 6 Value: 0
Acct-Authentic( 45) Len: 6 Value: RADIUS(1)
User-Service-Type( 6) Len: 6 Value: Framed-User(2)
Framed-Protocol( 7) Len: 6 Value: PPP(1)
Framed-Address( 8) Len: 6 Value: xx.xx.xx.xx
Acct-Delay-Time( 41) Len: 6 Value: 0
Thanks,
Bogdan
Bogdan Pelinescu <bpelin@itcnet.ro> |
R&D Engineer | Tel: (401) 232 2770
Institute for Computers |
Networks & Communications Dept. | Fax: (401) 230 7845
Bucharest, Romania |
Subject:Re: (usr-tc) Latest ER code for HiPer From: Brian Uechi <brianu@lava.net> Date: 1998-11-09 16:45:56
On Sat, 7 Nov 1998, Jeff Mcadams wrote:
> Thus spake Terry Kennedy
> >Jeff, I have them set to round robin on the DSP's and still
> >have the problem. I understand what you are saying but then
> >if that's so shouldn't the problem be gone.
>
> Not necessarily...it could happen if there's only the one ds0 free on
> the span...if it drops, it will almost immediately be filled again with
> the next call...at this point (particularly with chan-t1's don't know
> how the DSP's deal with pri's) there's only the one modem available, and
> the one ds0, so you're in the same situation ultimately.
Since PRI only have 23 lines but the DSP has 24 modems you'd think
this wouldn't happen on PRI. But the the last time I looked only 23
modems are used in DSP PRI configurations. I think this problem is
not present in the quads/dual PRI because all 48 modems are used. A
partial solution (for PRI users only) is for 3Com to include all 24
DSP modems in the pool.
I don't remember where I have download it, but I can send you by email.
Cedric MARSOT - INTERVALLE Tel : +33 (0)1 49 72 58 58
Chef Produits d'Acces Distants Fax : +33 (0)1 49 72 58 59
Email: mailto:Cedric.Marsot@intervalle.fr Pager : +33 (0)6 06 49 14 43
Web: http://www.intervalle.fr Mobile: +33 (0)6 60 21 76 79
Pager: mailto:0606491443@kobby.worldnet.net
MCT NT WKS/SRV, Win95, Tcpip, Exchange, IIS
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian K McIntire
> Sent: Monday, November 09, 1998 5:18 PM
> To: usr tc
> Subject:
>
>
> Does anyone know how to get a copy of the Cistron RADIUS.
> I've heard a lot
> of people talking about it but did not see if anyone said
> where to find it.
>
> ==============================================================
> =========
> = Brian K. McIntire
> bmcintire@commnet.com =
> = Systems Engineer III
> 317-558-5050 x113 =
> = CommNetPlus, Indianapolis, IN
http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
I have just found it ...
Just check this ...
ftp://ftp.cistron.nl/pub/cistron/
Cedric MARSOT - INTERVALLE Tel : +33 (0)1 49 72 58 58
Chef Produits d'Acces Distants Fax : +33 (0)1 49 72 58 59
Email: mailto:Cedric.Marsot@intervalle.fr Pager : +33 (0)6 06 49 14 43
Web: http://www.intervalle.fr Mobile: +33 (0)6 60 21 76 79
Pager: mailto:0606491443@kobby.worldnet.net
MCT NT WKS/SRV, Win95, Tcpip, Exchange, IIS
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian K McIntire
> Sent: Monday, November 09, 1998 5:18 PM
> To: usr tc
> Subject:
>
>
> Does anyone know how to get a copy of the Cistron RADIUS.
> I've heard a lot
> of people talking about it but did not see if anyone said
> where to find it.
>
> ==============================================================
> =========
> = Brian K. McIntire
> bmcintire@commnet.com =
> = Systems Engineer III
> 317-558-5050 x113 =
> = CommNetPlus, Indianapolis, IN
http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Sorry ... this is the good link:
ftp://ftp.cistron.nl/pub/people/miquels/radius/
Cedric MARSOT - INTERVALLE Tel : +33 (0)1 49 72 58 58
Chef Produits d'Acces Distants Fax : +33 (0)1 49 72 58 59
Email: mailto:Cedric.Marsot@intervalle.fr Pager : +33 (0)6 06 49 14 43
Web: http://www.intervalle.fr Mobile: +33 (0)6 60 21 76 79
Pager: mailto:0606491443@kobby.worldnet.net
MCT NT WKS/SRV, Win95, Tcpip, Exchange, IIS
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian K McIntire
> Sent: Monday, November 09, 1998 5:18 PM
> To: usr tc
> Subject:
>
>
> Does anyone know how to get a copy of the Cistron RADIUS.
> I've heard a lot
> of people talking about it but did not see if anyone said
> where to find it.
>
> ==============================================================
> =========
> = Brian K. McIntire
> bmcintire@commnet.com =
> = Systems Engineer III
> 317-558-5050 x113 =
> = CommNetPlus, Indianapolis, IN
http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:(usr-tc) 70 amp nic From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-09 17:17:44
Anyone have one for sell. We have the power supply. Just need the back
Date sent: Mon, 9 Nov 1998 10:17:57 -0600
Send reply to: usr-tc@lists.xmission.com
> Does anyone know how to get a copy of the Cistron RADIUS. I've heard a lot
> of people talking about it but did not see if anyone said where to find it.
>
Try http://www.mdi.ca/sysadmin/cistron/
or http://miquels.www.cistron.nl/radius/
You'll not regret it :-)
Bogdan Pelinescu <bpelin@itcnet.ro> |
R&D Engineer | Tel: (401) 232 2770
Institute for Computers |
Networks & Communications Dept. | Fax: (401) 230 7845
Bucharest, Romania |
At 10:17 9-11-98 -0600, you wrote:
>Does anyone know how to get a copy of the Cistron RADIUS. I've heard a lot
>of people talking about it but did not see if anyone said where to find it.
>
Try an email to iljitsch@pine.nl
He has worked with the cistron radius (dont know if its available to others)
Subject:RE: (usr-tc) HiperARC From: Marshall Morgan <marshall@netdoor.com> Date: 1998-11-10 06:17:52
On Tuesday, November 10, 1998 5:11 AM, Phil Le Clercq
[SMTP:phil.le.clercq@cinergy.net] wrote:
> Hi all,
> I am after some basic config. ideas. Currently we are just a netserver and
> digital quad modem site.
> We are contracted to upgrade to HiperARC, the question is do all isdn calls
> terminate on the arc, like on the netserver or do they have to end up on a
> modem whether it be quad or DSP?
all ISDN calls are terminated on Quad I Modems or HiperDSP modems
Marshall Morgan
Internet Doorway, Inc (aka NETDOOR)
http://www.netdoor.com
601.969.1434 x28 | 800.952.1570 x28 | 601.969.3629 x28 | Fax 601.969.3838
Subject:Re: (usr-tc) Latest ER code for HiPer From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-10 08:11:32
Thus spake Brian Uechi
>On Sat, 7 Nov 1998, Jeff Mcadams wrote:
>> Not necessarily...it could happen if there's only the one ds0 free on
>> the span...if it drops, it will almost immediately be filled again with
>> the next call...at this point (particularly with chan-t1's don't know
>> how the DSP's deal with pri's) there's only the one modem available, and
>> the one ds0, so you're in the same situation ultimately.
>Since PRI only have 23 lines but the DSP has 24 modems you'd think
>this wouldn't happen on PRI. But the the last time I looked only 23
>modems are used in DSP PRI configurations. I think this problem is
>not present in the quads/dual PRI because all 48 modems are used. A
>partial solution (for PRI users only) is for 3Com to include all 24
>DSP modems in the pool.
I *thought* I had heard that behavior for PRI's on DSP's, but wasn't
sure. It is kinda goofy. Also, would be nice to have a first-available
settings for the modem assignment such as with the dual-pri cards and
quad's. This is what we use and we almost never run into problems with
this. Its *very* rare that it skips a resetting modem and gets into the
last two modems in the rack though, too. I'll go through and do a
visual check of our racks and if any have one or both of the last two
modems used, I go on a "wigged out modem hunt" on that chassis.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) HiperARC From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-10 08:13:46
Thus spake Phil Le Clercq
>We are contracted to upgrade to HiperARC, the question is do all isdn calls
>terminate on the arc, like on the netserver or do they have to end up on a
>modem whether it be quad or DSP?
You *should* be terminating your ISDN calls on quads anyway...in
general, it works much better.
The Arc's will not terminate ISDN calls themselves at all, they *must*
be handled by the modem cards (be it quad's or DSP's). There really is
little to no downside to this.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:(usr-tc) CONNECTION PROBLEM From: Jeff Binkley <jeff.binkley@asacomp.com> Date: 1998-11-10 08:56:00
We have a customer who has Brother NB-80C Geobook trying to connect with us
on a HiperArc running 4.1.78 code. The call continually fails in the LCP
negotiation stage and we never see the userid and password get sent. The
Geobook is a replacement because he had so many problems with the old one.
With the old one he was able to connect with us but since then we have done
a couple of HiPerArc release upgrades. Below is a copy of a MONITOR PPP
trace from the HiPerArc. I am concerned this may be a HiPerArc problem
but I want to be sure. We have sent the trace on to Brother too. They do
include a diskette from Earthlink and it does work with them. Does anyone
have a document for decoding these traces and figuring out what is going
on ?
We spoke to Brother and they were no help. Basically mumbling things like if
you are trying to connect to an NT server it won't work and we only support
dialing into Unix machines. They said send them to Earthlink. I don't think
we want to start eliminating groups of users from our collective ISps based
upon the equipment the end user has. That doesn't seem to be good business
practice.
Jeff Binkley
ASA Network Computing
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_REQ MRU 05 ea
ASYNC_MAP 00 00 00 00
AUTH_TYPE c0 23
MAGIC_NUM 43 de 5f 5d
PROTO_COMP
AC_COMP
MPP_MRRU 05 ea
MPP_ENDPTID 00
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REJ MRU 05 ea
MPP_MRRU 05 ea
MPP_ENDPTID 00
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_REQ MRU 05 ea
ASYNC_MAP 00 00 00 00
AUTH_TYPE c0 23
MAGIC_NUM 43 de 5f 5d
PROTO_COMP
AC_COMP
MPP_MRRU 05 ea
MPP_ENDPTID 00
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REQ ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
QUAL_PROTO c0 25 00 00 03 e8
PROTO_COMP
AC_COMP
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_REJ QUAL_PROTO c0 25 00 00 03 e8
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REJ MRU 05 ea
MPP_MRRU 05 ea
MPP_ENDPTID 00
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REQ ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_ACK ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_REQ MRU 05 ea
ASYNC_MAP 00 00 00 00
AUTH_TYPE c0 23
MAGIC_NUM 43 de 5f 5d
PROTO_COMP
AC_COMP
MPP_MRRU 05 ea
MPP_ENDPTID 00
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REQ ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_ACK ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REJ MRU 05 ea
MPP_MRRU 05 ea
MPP_ENDPTID 00
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_REQ MRU 05 ea
ASYNC_MAP 00 00 00 00
AUTH_TYPE c0 23
MAGIC_NUM 43 de 5f 5d
PROTO_COMP
AC_COMP
MPP_MRRU 05 ea
MPP_ENDPTID 00
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REQ ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_ACK ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REJ MRU 05 ea
MPP_MRRU 05 ea
MPP_ENDPTID 00
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_REQ MRU 05 ea
ASYNC_MAP 00 00 00 00
AUTH_TYPE c0 23
MAGIC_NUM 43 de 5f 5d
PROTO_COMP
AC_COMP
MPP_MRRU 05 ea
MPP_ENDPTID 00
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REQ ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_ACK ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REJ MRU 05 ea
MPP_MRRU 05 ea
MPP_ENDPTID 00
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_REQ MRU 05 ea
ASYNC_MAP 00 00 00 00
AUTH_TYPE c0 23
MAGIC_NUM 43 de 5f 5d
PROTO_COMP
AC_COMP
MPP_MRRU 05 ea
MPP_ENDPTID 00
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REQ ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_ACK ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
MPP_MRRU 05 ea
LCP CFG_REJ MRU 05 ea
MPP_MRRU 05 ea
MPP_ENDPTID 00
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_REQ MRU 05 ea
ASYNC_MAP 00 00 00 00
AUTH_TYPE c0 23
MAGIC_NUM 43 de 5f 5d
PROTO_COMP
AC_COMP
MPP_MRRU 05 ea
MPP_ENDPTID 00
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REQ ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_ACK ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REJ MRU 05 ea
MPP_MRRU 05 ea
MPP_ENDPTID 00
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_REQ MRU 05 ea
ASYNC_MAP 00 00 00 00
AUTH_TYPE c0 23
MAGIC_NUM 43 de 5f 5d
PROTO_COMP
AC_COMP
MPP_MRRU 05 ea
MPP_ENDPTID 00
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REQ ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_ACK ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REJ MRU 05 ea
MPP_MRRU 05 ea
MPP_ENDPTID 00
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_REQ MRU 05 ea
ASYNC_MAP 00 00 00 00
AUTH_TYPE c0 23
MAGIC_NUM 43 de 5f 5d
PROTO_COMP
AC_COMP
MPP_MRRU 05 ea
MPP_ENDPTID 00
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REQ ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_ACK ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REJ MRU 05 ea
MPP_MRRU 05 ea
MPP_ENDPTID 00
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_REQ MRU 05 ea
ASYNC_MAP 00 00 00 00
AUTH_TYPE c0 23
MAGIC_NUM 43 de 5f 5d
PROTO_COMP
AC_COMP
MPP_MRRU 05 ea
MPP_ENDPTID 00
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REQ ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Outgoing PPP Data on interface: slot:3/mod:1
LCP CFG_ACK ASYNC_MAP 00 0a 00 00
MAGIC_NUM 00 cd df 13
PROTO_COMP
AC_COMP
Incoming PPP Data on interface: slot:3/mod:1
LCP CFG_REJ MRU 05 ea
MPP_MRRU 05 ea
MPP_ENDPTID 00
Brother hangs up at this point...
This is a multi-part message in MIME format.
------=_NextPart_000_0131_01BE0C91.1FAEE760
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Does anyone know of an implementation for the Hyper Chassis that is =
authenticating through a remote server running Livingston Radius, to =
filter icmp or other keep alive signals on connections for dial-up =
users? There must be something simpler than the information that I have =
researched that I'm overlooking. Any help or push in the right =
direction would be greatly appreciated.
-Davey
------=_NextPart_000_0131_01BE0C91.1FAEE760
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN">
<HTML>
<HEAD>
<META content=3Dtext/html;charset=3Diso-8859-1 =
http-equiv=3DContent-Type>
<META content=3D'"MSHTML 4.72.3110.7"' name=3DGENERATOR>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT color=3D#000000 size=3D2>Does anyone know of an =
implementation for the=20
Hyper Chassis that is authenticating through a remote server running =
Livingston=20
Radius, to filter icmp or other keep alive signals on connections for =
dial-up=20
users? There must be something simpler than the information that I =
have=20
researched that I'm overlooking. Any help or push in the right =
direction=20
would be greatly appreciated.</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>-Davey</FONT></DIV></BODY></HTML>
------=_NextPart_000_0131_01BE0C91.1FAEE760--
Subject:(usr-tc) HiperARC From: Phil Le Clercq <phil.le.clercq@cinergy.net> Date: 1998-11-10 11:10:41
Hi all,
I am after some basic config. ideas. Currently we are just a netserver and
digital quad modem site.
We are contracted to upgrade to HiperARC, the question is do all isdn calls
terminate on the arc, like on the netserver or do they have to end up on a
modem whether it be quad or DSP?
I was going to pull the quads out of a chassis that will handle only ISDN,
but setting this up could scupper things up when we get the new gear....
Cheers for any help
Phil
Cinergy Communications
Subject:(usr-tc) Login Perfomance with TotalControl NetServer/Quadmodems and ISDN From: Ralph Helfenberger <rhelfenberger@comlight.ch> Date: 1998-11-10 11:18:23
Hi
We have TotalControl Equipment and Ascend Max 4000 Equipment. Customers
dialing with ISDN into the TotalControl have around 7 seconds until the
connection is established, customers dialing into Ascend about 3
seconds.
What is your experienc with customers dialing in with ISDN? What
connection times do you have? (connection time = time from the moment
the Terminal adaptor starts dialing until the Dial Up Networking window
disapears on the Windows desktop).
Configuration
PRI 3.02, Quad 5.9.9, NetServer 3.7.24
ISDN Calls are routed to the NetServer.
ISDN Protocol set to auto recognition.
Any input is very appreciated.
Ralph
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Davey Walbeck <davey@vii.com> Date: 1998-11-10 12:04:58
The easist solution for idle timeouts is to set the default user idletime on
the Hyper Chassis.
-Davey
-----Original Message-----
>Does anybody have a good solution in place for enforcing idle times and
>multiple logins?
>
>I am currently using PMMON and I get no end of complaints from people
>that pmmon disconnects them at inappropriate times among other things.
>
>If anybody has a different/better way of doing this I'de love to hear about
it.
>
>
>
>Mark Lemmert cto@athenet.net
>Chief Technical Officer (920)954-9799
>AthEnet Data Exchange http://www.athenet.net
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Davey Walbeck <davey@vii.com> Date: 1998-11-10 12:40:34
The problem is that we do not have just 3Com equipment. The PM3 and other
equipment that we have must be able to interact with radius, which is why we
have radius running on a separate server under Livingston Radius. Any other
ideas that wouldn't involve conpletely converting Radius?
-Davey
>Use 3com's radius server, it can handle idle timeouts, multiple logins,
>but beware, it has its drawbacks :-).
>
>Mark Lemmert wrote:
>>
>> Does anybody have a good solution in place for enforcing idle times and
>> multiple logins?
>>
>> I am currently using PMMON and I get no end of complaints from people
>> that pmmon disconnects them at inappropriate times among other things.
>>
>> If anybody has a different/better way of doing this I'de love to hear
about it.
>>
>> Mark Lemmert cto@athenet.net
>> Chief Technical Officer (920)954-9799
>> AthEnet Data Exchange http://www.athenet.net
>>
>> -
>> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
>> with "unsubscribe usr-tc" in the body of the message.
>> For information on digests or retrieving files and old messages send
>> "help" to the same address. Do not use quotes in your message.
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
I am currently running 3.8.1 on my Netservers and 4.1.72 -7 on
my HyperARC. Ever since I upgraded from 4.1.11 (arc) and 3.7.72 (Netserver)
I have seen major through put problems with any MPIP ISDN connection.
Typically I get around 100kbps throughput via ISDN and it is down
around 20-35kbps if the session is MPIP.
Has anybody else experienced this?
Mark Lemmert cto@athenet.net
Chief Technical Officer (920)954-9799
AthEnet Data Exchange http://www.athenet.net
Subject:(usr-tc) Solution for idle time limits and dual logins From: Mark Lemmert <cto@athenet.net> Date: 1998-11-10 12:54:52
Does anybody have a good solution in place for enforcing idle times and
multiple logins?
I am currently using PMMON and I get no end of complaints from people
that pmmon disconnects them at inappropriate times among other things.
If anybody has a different/better way of doing this I'de love to hear about it.
Mark Lemmert cto@athenet.net
Chief Technical Officer (920)954-9799
AthEnet Data Exchange http://www.athenet.net
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: MegaZone <megazone@megazone.org> Date: 1998-11-10 12:57:10
Once upon a time Davey Walbeck shaped the electrons to say...
>The problem is that we do not have just 3Com equipment. The PM3 and other
>equipment that we have must be able to interact with radius, which is why we
>have radius running on a separate server under Livingston Radius. Any other
>ideas that wouldn't involve conpletely converting Radius?
No - but try Cistron RADIUS. It started out based on Livingston 1.16 (as
did most servers) but it has grown quite a bit. It basically has all of the
features of Lucent 2.0.1 save for SecurID and Menu support. It talks to all
of the major NAS boxes, and handles VSAs - including 3Com VSAs in the latest
build.
It hs proven simultaneous login prevention support, hands proxy, RADIUS based
huntgroups, and lots of other keen stuff. Cistron even handles iPass now
and there are patches for SQL on the backend.
And it is free.
<URL:http://miquels.www.cistron.nl/radius/>
<URL:http://www.mdi.ca/sysadmin/cistron/>
I belive you'd be able to just drop your existing RADIUS users file in and
be off and running.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Richard Lorbieski <richard@mail.alpha1.net> Date: 1998-11-10 13:18:04
Use 3com's radius server, it can handle idle timeouts, multiple logins,
but beware, it has its drawbacks :-).
Mark Lemmert wrote:
>
> Does anybody have a good solution in place for enforcing idle times and
> multiple logins?
>
> I am currently using PMMON and I get no end of complaints from people
> that pmmon disconnects them at inappropriate times among other things.
>
> If anybody has a different/better way of doing this I'de love to hear about it.
>
> Mark Lemmert cto@athenet.net
> Chief Technical Officer (920)954-9799
> AthEnet Data Exchange http://www.athenet.net
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
On Tue, Nov 03, 1998 at 02:22:39PM -0700, Aaron Leonard wrote:
> > no fair-queue
> > no cdp enable
> > ppp authentication pap chap
> > ppp multilink
>
> do you really want to use multilink PPP over async? If so
> you'll want to have 11.3 and configure a virtual-template.
Can't speak for the 5100s, but I'll swear blind that 11.2.something
(greater that 9 I think) on a 5200 works fine with multilink PPP and
even multi-chassis, tested with both NT4 and Windahs95+DUN1.2
clients.
OK, when I say fine, I mean, it works... beyond three lines is seems
to be a bit of a loss...
-cw
Thus spake Mark Lemmert
>I am currently running 3.8.1 on my Netservers and 4.1.72 -7 on
>my HyperARC. Ever since I upgraded from 4.1.11 (arc) and 3.7.72 (Netserver)
>I have seen major through put problems with any MPIP ISDN connection.
>Typically I get around 100kbps throughput via ISDN and it is down
>around 20-35kbps if the session is MPIP.
Just to clarify what's going on there...MPIP isn't the culprit here.
MPIP is purely a coordination protocol. It lets the NETServers and
Arc's coordinate where a bundle is hosted...it isn't actually
responsible for getting the data from one NETServer to another to
actually be bundled. The protocol responsible for that is VTP.
To answer your other question...no...I haven't experienced any of this,
but I'm also using different versions of code.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Ricky Beam <jfbeam@interpath.net> Date: 1998-11-10 14:36:18
Richard Lorbieski was heard to say:
>Use 3com's radius server, it can handle idle timeouts, multiple logins,
>but beware, it has its drawbacks :-).
>
>Mark Lemmert wrote:
>>
>> Does anybody have a good solution in place for enforcing idle times and
>> multiple logins?
To be fair, any radius server can handle idle timeouts (well, as long as it
can handle sending a USR VSA.) As for multiple login tracking... radius is
a stateless protocol and as such, it's very difficult to maintain any stated
data. Most systems that do tracking "correctly" have a task outside the
radius server polling the NASes to keep state information correct.
Of all the RADIUS servers, I like the USR server best... it provides one of
the most powerful and configurable systems. Yes, it is abit slower than
all the others, but what it lets you do fair outways the speed. (And it's
generally "fast enough".) [I especially like being able to rewrite it's
database schema without having to touch it's source code.]
--Ricky
Subject:(no subject) From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-10 15:55:13
I have a zoom modem (Model # 2948) I can connect to one ISP running 1.2.5
and 4.0.30 but not to another ISP running 1.2.68 and 4.1.78 - 4. Any Ideas?
What is the latest code available?
I don't know if this is the right way to find the version but ati6 shows:
RCV56DPF L8570A Rev 47.20/47.20
I'm on site trying to find out what specifically may be causing the bulk of
our problems. I would appreciate input from anyone who may be able to put
in their 2 cents
=======================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
Subject:(usr-tc) HiperARC.. From: System Administrator <sysadmin@evcom.net> Date: 1998-11-10 16:16:24
Config'ing first HiperARC (btw, must say that CLI is MUCH better than
netserver), couple of questions:
1. Is there still a known problem with chassis awareness (w/ latest non-ER
code). Only one HARC, so should I do chassis slot assignments manually?
2. Am I to understand that the Idle-Timeout RADIUS attribute does not work
correctly with the HARC, and that I must use a VA (we have a highly customized
radius daemon, which supports loads of goodies, but not VA)?
Many thanks,
--
Jesse Sipprell
Technical Operations Director
Evolution Communications, Inc.
800-496-4736 (ext 106)
* Finger sysadmin@evcom.net for my PGP Public Key *
PS. Extra-special-thanks to the fine folks at interpath for providing
http://enterprise.interpath.net/~jfbeam/USR/hiperARC-setup
;=)
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: MegaZone <megazone@megazone.org> Date: 1998-11-10 16:23:23
Once upon a time K Mitchell shaped the electrons to say...
>limiting multiple logins has caused problems. It seems that S&A doesn't
>always get notification when a user logs out, therefore when a user logs
>back in, it gets interpreted as a 2nd concurrent login and is denied.
Which is one big reason why Cistron is better.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Subject:(usr-tc) Re: From: John Powell <john_powell@mw.3com.com> Date: 1998-11-10 17:05:48
ATI3 is the code rev usually. I6 only shows a chip rev or something
(doesn't seem to change with a flash).
Try S202=32 and see if that works. It will (on some Rockwell's) turn off
their flakey A/D detection.
I highly doubt this is a 1.2.68 vs 1.2.5 issue. Arc code would not be
related at all.
JP
"Brian K McIntire" <bmcintire@commnet.com> on 11/10/98 03:55:13 PM
Please respond to usr-tc@lists.xmission.com
cc: (John Powell/MW/US/3Com)
I have a zoom modem (Model # 2948) I can connect to one ISP running 1.2.5
and 4.0.30 but not to another ISP running 1.2.68 and 4.1.78 - 4. Any
Ideas?
What is the latest code available?
I don't know if this is the right way to find the version but ati6 shows:
RCV56DPF L8570A Rev 47.20/47.20
I'm on site trying to find out what specifically may be causing the bulk of
our problems. I would appreciate input from anyone who may be able to put
in their 2 cents
=======================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:RE: (usr-tc) Solution for idle time limits and dual logins From: Randy Cosby <dcosby@infowest.com> Date: 1998-11-10 17:35:36
I'll put in my two cents for Radiator (http://www.open.com.au/radiator). It
does checking two ways. First, it checks the database to see if a user is
logged on. The database is user-definable. We log it to mysql, and can
then output neat web pages in PHP to see who's online. If it does find the
user in the DB too many times, it does a double-check using a pmmon type
program. You can use different "checker" programs for different boxes.
I'd like it better if I always got a STOP, but this is livable.
Randy
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of MegaZone
> Sent: Tuesday, November 10, 1998 5:23 PM
> To: usr-tc@lists.xmission.com
> Subject: Re: (usr-tc) Solution for idle time limits and dual logins
>
>
> Once upon a time K Mitchell shaped the electrons to say...
> >limiting multiple logins has caused problems. It seems that S&A doesn't
> >always get notification when a user logs out, therefore when a user logs
> >back in, it gets interpreted as a 2nd concurrent login and is denied.
>
> Which is one big reason why Cistron is better.
>
> -MZ
> --
> <URL:mailto:megazone@megazone.org> Gweep, Discordian, Author,
> Engineer, me..
> Join ISP/C Internet Service Providers' Consortium
> <URL:http://www.ispc.org/>
> "A little nonsense now and then, is relished by the wisest men"
> 781-788-0130
> <URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail
> Discordia!
>
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: K Mitchell <mitch@keyconn.net> Date: 1998-11-10 17:51:08
At 01:18 PM 11/10/98 -0600, you wrote:
>Use 3com's radius server, it can handle idle timeouts, multiple logins,
>but beware, it has its drawbacks :-).
I've not had any problems with idle timeouts in 3Com's S&A server, but
limiting multiple logins has caused problems. It seems that S&A doesn't
always get notification when a user logs out, therefore when a user logs
back in, it gets interpreted as a 2nd concurrent login and is denied.
YMMV,
Kirk
Kirk Mitchell-General Manager sysadmin@keyconn.net
Keystone Connect http://www.keyconn.net
Altoona, PA 814-941-5000 We Unlock the World
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Rev. Jim <packrat@aus-etc.com> Date: 1998-11-10 19:31:47
>>Does anybody have a good solution in place for enforcing idle times and
>>multiple logins?
one thing that i noticed in all the replys is that there is no mention
about those that get around the idle timeout by keeping the mail reader
checking the mail every x minutes - i have had to place a Session-Timeout
of four hours to help - or am i just missing how to set a thresh-hold that
ignores so many bits per/time on the Idle-Timeout - as implemented now, it
is one useless command
"The avalanche has already started;
it is too late for the pebbles to vote." -- Kosh
packrat
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Mike Andrews <mandrews@termfrost.org> Date: 1998-11-10 20:55:56
On Tue, 10 Nov 1998, Rev. Jim wrote:
> >>Does anybody have a good solution in place for enforcing idle times and
> >>multiple logins?
>
>
> one thing that i noticed in all the replys is that there is no mention
> about those that get around the idle timeout by keeping the mail reader
> checking the mail every x minutes - i have had to place a Session-Timeout
> of four hours to help - or am i just missing how to set a thresh-hold that
> ignores so many bits per/time on the Idle-Timeout - as implemented now, it
> is one useless command
...and I think this is what the original post was REALLY asking. :)
There's no easy fix for this. You can write a program to check number of
octets sent/received on each port, and kick people off that are below a
threshold. TSMON (www.tsmon.com) supposedly does this. (We don't use it,
though.)
I'd prefer something a little more exact... like a filter you could use in
conjunction with the idle timeout. i.e. ignore POP3 traffic (110/tcp),
ICQ traffic (it sends a packet on port 4000/udp every 2 minutes!), all
ICMP, and a few others when considering what "idle time" is. This idea,
and the pros and cons of it, have been discussed to death several times on
several mailing lists (not this one)... I don't remember what the
conensus was, but, well, I don't think ANY vendor has implemented such a
beast yet. 3com definitely hasn't. I doubt Cisco, Livingston, or Ascend
have either.
I was going to work on one using tcpdump, but I figured out really fast
that it was going to eat a LOT of bpf devices and probably CPU time too,
and more important things came up...
Mike Andrews (MA12) icq 6602506 -------------- mandrews@termfrost.org
VP 'n' Systems/Network Administrator --------------- mandrews@dcr.net
Digital Crescent, Frankfort, KY ----------- http://www.termfrost.org/
# view;touch;unzip;finger;mount;mv;mv;mv;yes;mv;yes;mv;yes;umount;sleep
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: William Behrens <wbehrens@feist.com> Date: 1998-11-10 21:12:08
We are working on a program (runs on NT/SQL) that will poll the TC's and
kick multipule logins off. With user definable boot parameters like kicking
by idle time or newest login and validation before the actual kicking
proceedure takes place. We are losing almost 100 ports to multipule logins
and don't want to modify or delpoy a new radius version (12K users on BSDI).
Right now we are polling 22 TC boxes in 45 secs and kicking the multipule
logins 15 secs later. Granted its not lighting fast but will work much
better when we get it deployed on the local network with the TC boxes
instead of over a T1.
William Behrens
Director of Network Operations
ParaCom Technologies Inc.
wbehrens@paracom.com
-----Original Message-----
>Once upon a time K Mitchell shaped the electrons to say...
>>limiting multiple logins has caused problems. It seems that S&A doesn't
>>always get notification when a user logs out, therefore when a user logs
>>back in, it gets interpreted as a 2nd concurrent login and is denied.
>
>Which is one big reason why Cistron is better.
>
>-MZ
>--
><URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer,
me..
>Join ISP/C Internet Service Providers' Consortium
<URL:http://www.ispc.org/>
>"A little nonsense now and then, is relished by the wisest men"
781-788-0130
><URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail
Discordia!
>
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-10 21:24:51
Thus spake Rev. Jim
>>>Does anybody have a good solution in place for enforcing idle times and
>>>multiple logins?
>one thing that i noticed in all the replys is that there is no mention
>about those that get around the idle timeout by keeping the mail reader
>checking the mail every x minutes - i have had to place a Session-Timeout
>of four hours to help - or am i just missing how to set a thresh-hold that
>ignores so many bits per/time on the Idle-Timeout - as implemented now, it
>is one useless command
I've been asking for *years* for a filter or ability to define an
access-list or something to specify the types of traffic that reset the
idle-timer....and I don't exaggerate when I say years...I've asked this
of various vendors...and have just recently found the possibility that
one has implemented this. I've heard recently that Cisco might have
done this...but haven't played with it personally, so don't know for
sure if they have.
What about it 3Com? Can we *finally* have this feature? Please? The
code to look into packets is already there, you already filter based on
ports and protocols...how hard could it be to have the idle-timer check
a filter for it?
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:RE: (usr-tc) Problem with encrypted passwords From: Eric <elorenzo@mediacity.com> Date: 1998-11-10 21:25:52
Did you use the set ppp receive_authentication command pap? The HARC
might default to chap which I know does not work with our RADIUS server.
Eric
---
Eric J. Lorenzo
Field Service Engineer
v:650.237.1465
Lorenzo@ISPchannel.com
http://www.ISPchannel.com
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Theodore Cekan
> Sent: 10 November, 1998 21:01
> To: usr-tc@lists.xmission.com
> Subject: (usr-tc) Problem with encrypted passwords
>
>
> I have a new Hiper system that wont authenticate users using
> Radius. The
> password seems to be sent encrypted and my Radius server cant decrypt it
> correctly. What could cause this?
>
> Ted
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) v.90 connect problems From: Mario M. Bustamante <mario@accesspro.net> Date: 1998-11-10 21:33:35
I posted a message earlier describing connection problems with
Sportster modems. Up until recently, we assumed that our
customers were complaining because they were using the cheap V90
modems, and the problem was on their end.
As the complaints started to add up, I had out support people
fill out a form to gather information from problem users. The
process included asking the end user to run "winipcfg.exe" on
their Windows 95 or 98 systems and keep a log of good and bad
connections. Since we have mostly Net Server / Quad modem
equipment and only one Hiper ARC / DSP, we wanted to know what
equipment was causing the most problems and with what kind of
modem (on the customer end).
We have somewhat over 2,200 users, and we have always recommended
that they buy USR/3Com modems, so we were shocked at the number
of Sportster users having problems (and they were mad at us for
making them spend extra money to buy more problems). We were also
shocked to find that a lot of them were connecting to BellSouth
with a much better V90 connection (they use Ascend equipment).
I do not want to post results publicly, but after compiling the
results of almost 200 complaints, I can tell you that most of the
problems were related to the Hiper ARC / DSP, and that the
percentage of complaints from Sportster users was alarming.
I would love to get someone that knows Sportsters on a conference
call with someone that knows Hiper ARCs so we can see where the
problems lie. I know that there are a lot of bad phone lines and
a lot of people who set up their modems with the wrong settings,
but somebody has to help us give them tech support.
Any ideas?
Thanks,
_______________________________________________
Mario M. Bustamante, President
AccessPro Communications Inc.
Miami, Florida
Internet Service Providers, Web Hosting & Design
Microsoft Certified Web Presence Providers
Wide Area Networks, Security, Intranets.
http://www.AccessPro.net mario@accesspro.net
_______________________________________________
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of
> John Powell
> Sent: Saturday, November 07, 1998 10:15 PM
> To: usr-tc@lists.xmission.com
> Subject: Re: (usr-tc) v.90 connect problems
>
>
> >OK, but a fair number of the people having problems
> have Sportsters.
> >(Some Winmodems, some not.) I wasn't aware there was
> any new Sportster
> >code out... if there is, where's it hidden? If
> there isn't, is this one
> >of the server-side problems?
>
> I was responding to your comments on LT Winmodems, you
> did not mention
> Sportsters. I know of no significant issues with
> interop with Sportsters
> (Winmodem or controller based) and Total Control modems.
>
> There will likely be improvements with the client side
> (all vendors) for
> quite some time. There are quite a few tweeks to
> enhance performance and
> coverage in the labs already. Heck, we are still
> improving V.34 4 years
> later.
>
> If you can provide details on problems with
> Sportsters, I will gladly take
> this off-line and put you in contact with my peers on
> the PCD (home of the
> Sportster) side of the house.
>
> JP
>
> PS. No, there are no recent releases of Sportster
> code related to V.90
> fixes. Overall, they are working quite well.
>
>
>
> -
> To unsubscribe to usr-tc, send an email to
> "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and
> old messages send
> "help" to the same address. Do not use quotes in
> your message.
>
Subject:(usr-tc) Problem with encrypted passwords From: Theodore Cekan <ted@mho.net> Date: 1998-11-10 22:00:52
I have a new Hiper system that wont authenticate users using Radius. The
password seems to be sent encrypted and my Radius server cant decrypt it
correctly. What could cause this?
Ted
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Richard Stuplich <dick@dwave.net> Date: 1998-11-10 22:04:09
We use a program I wrote called portman. It uses a config file that looks
like this: (this is wider than a mail message, it will look a bit odd with
the comments)
# Global configuration section,
# Must be first section, may be mix of upper and lower case
# Section must end with an "$end" line.
# NOTE: Config pairs and "$end" tags can be seperated by any of the following:
# SPACE, TAB, COMMA, SEMICOLON or RETURN.
PortMaster squid.dwave.net # PM2 30 ports (Not a user variable)
#PortMaster octopus.dwave.net # PM2 29 ports (Not a user variable)
#PortMaster starfish.dwave.net # PM2 10 ports (Not a user variable)
PortMaster draco.dwave.net # PM3 48 ports (Not a user variable)
PortMaster abyss.dwave.net # PM3 48 ports (Not a user variable)
PortMaster digimon.dwave.net # PM3 48 ports (Not a user variable)
PortMaster bathory.dwave.net # PM3 48 ports (Not a user variable)
PortMaster monster.dwave.net # TCH 48 ports (Not a user variable) +1
PortMaster tc2.dwave.net # PM3 48 ports (Not a user variable)
TotalPorts 319 # The total number of ports we have to watch (Not a user
variable)
# Note: TotalPorts will see all ports that are returned by
# a pmwho on the device, so on PM3's you will get C0 if you use it
# or not. If it alwaus sits in "USERNAME" then it will always be
# counted as a port in use so you have to take that into account
# on TotalPorts. This is true for a TCH too.
# These are the default values to set for all users
FreeLow 30 # TotalPorts-FreeLow is considered Low load
FreeHig 7 # FreeLow-FreeHig is considered Med load
# above FreeHig is considered Hig
IdleLimitLow 20 # Minutes idle time if in Low load
IdleLimitMed 20 # Minutes idle time if in Med load
IdleLimitHig 15 # Minutes idle time if in Hig load
# Note: 0 = no limit, -1 means always match
MaxTimeLow 0 # Max time per connect Low
MaxTimeMed 480 # Max time per connect Med
MaxTimeHig 360 # Max time per connect Hig
Sessions 1 # Max number of concurent sessions
$end # End the global config section (MUST HAVE THIS)
# This is the users config area below, users configs fall through.
# Much like a 'case' in a 'switch' statement in C
# Use this area to deviate from the above globals for individual users,
# or groups of users by omiting the '$end' in a group.
# allow these to be idle forever but
# throw off as as soon as we go into high load
user beth
user stomper
user alk
user koskelin
user dwkris
user kozz
user bart
user bkniess
user brad
user vwluvbug
user chadwick
user chad
user shark
user dick
user jeff
user brennan
user kat
IdleLimitLow 0
MaxTimeHig -1
$end
# Group to allow 4 simul logins
user ruder
Sessions 4
$end
# Group to allow 3 simu logins
user wifc
user wdez
user wofm
Sessions 3
$end
# Group to allow 2 simul logins
user waow
user signpro
user wausaucc
user wfl
Sessions 2
$end
# Dedicated dial up accounts
user cclink
user lsmail
user mp
IdleLimitLow 0
IdleLimitMed 0
IdleLimitHig 0
MaxTimeLow 0
MaxTimeMed 0
MaxTimeHig 0
$end
# E-Mail only Accounts
user bcrb
user wexzo
# Telnet Only Accounts
user nzate
user smaxker
# DW Employee Forwards
user ds
user jp
user kl
user kk
user bk
user jb
user ch
user bal
user js
IdleLimitLow -1
IdleLimitMed -1
IdleLimitHig -1
MaxTimeLow -1
MaxTimeMed -1
MaxTimeHig -1
$end
This file is very cut down from our production portman.conf file but you
get the idea.
As you can see you can group multiple users together and set overrides for
the defaults at the top.
If a user is not in the lower part then they get the defaults set at the top.
It also uses 3 different settings for HIGH, MEDIUM and LOW load. We like
this so our free accounts can get thrown off when we hit high load. We
have been using portman for over a year and it not only does all this but
creates logs for terminations and sends email to multiple offenders.
We have this program working with PM2, PM3, USR Netserver and 3COM HiperArc
units. It is a unix program. If anyone is willing to help with the
finishing of the docs and other polishing I would love to get this ready
for free distribution.
All I need is some competent ISPs that use UNIX and want to use this. I
would help set it all up in exchange for some documentation help.
At 10:30 PM 11/10/98 -0500, you wrote:
>At 09:12 PM 11/10/98 -0600, you wrote:
>> We are working on a program (runs on NT/SQL) that will poll the TC's and
>>kick multipule logins off. With user definable boot parameters like kicking
>>by idle time or newest login and validation before the actual kicking
>>proceedure takes place. We are losing almost 100 ports to multipule logins
>>and don't want to modify or delpoy a new radius version (12K users on BSDI).
>>Right now we are polling 22 TC boxes in 45 secs and kicking the multipule
>>logins 15 secs later. Granted its not lighting fast but will work much
>>better when we get it deployed on the local network with the TC boxes
>>instead of over a T1.
>>
>>William Behrens
>>Director of Network Operations
>>ParaCom Technologies Inc.
>>wbehrens@paracom.com
>>
>--------------------------------------
>Hey William !
>If you can add to this the ability to also set a max session time, I'd be
>interested in purchasing this program from you !
>
>Keep me posted ?
>
>Ken
>_______________________________________
> Ken Hodges, President and CEO
> ACME BrainWorks Internet Services, Inc.
> Clayton Computers, Inc.
> Rabun County, Georgia
> "Where Spring Spends the Summer"
> http://www.acme-brain.com or http://www.rabun.net
> ken@rabun.net 1-706-782-9239
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
Richard B. Stuplich DataWave, Not just faster, Better.
President, DataWave Technologies 17 T1 circuits and growing strong.
DataWave, Wausau's first local ISP to have a direct connection to the
midwest NAP, Frame Relay, ISDN, x2, k56Flex, v.90. The only area ISP to
have redundant T1 NAP connections, All modem compatibility and a staff
dedicated exclusively to providing Internet Service to this area.
This E-Mail Copyright 1998 Richard Stuplich, see http://dw.net/cr/
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Brian Elfert <brian@citilink.com> Date: 1998-11-10 22:05:13
On Tue, 10 Nov 1998, Jeff Mcadams wrote:
> I've been asking for *years* for a filter or ability to define an
> access-list or something to specify the types of traffic that reset the
> idle-timer....and I don't exaggerate when I say years...I've asked this
> of various vendors...and have just recently found the possibility that
> one has implemented this. I've heard recently that Cisco might have
> done this...but haven't played with it personally, so don't know for
> sure if they have.
So, you filter out one thing. Anyone who really wants to get around idle
timeouts will just switch to something.
This would really only help if customers are accidently leaving an email
check running.
Brian
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Ken Hodges <ken@rabun.net> Date: 1998-11-10 22:30:54
At 09:12 PM 11/10/98 -0600, you wrote:
> We are working on a program (runs on NT/SQL) that will poll the TC's and
>kick multipule logins off. With user definable boot parameters like kicking
>by idle time or newest login and validation before the actual kicking
>proceedure takes place. We are losing almost 100 ports to multipule logins
>and don't want to modify or delpoy a new radius version (12K users on BSDI).
>Right now we are polling 22 TC boxes in 45 secs and kicking the multipule
>logins 15 secs later. Granted its not lighting fast but will work much
>better when we get it deployed on the local network with the TC boxes
>instead of over a T1.
>
>William Behrens
>Director of Network Operations
>ParaCom Technologies Inc.
>wbehrens@paracom.com
>
Hey William !
If you can add to this the ability to also set a max session time, I'd be
interested in purchasing this program from you !
Keep me posted ?
Ken
_______________________________________
Ken Hodges, President and CEO
ACME BrainWorks Internet Services, Inc.
Clayton Computers, Inc.
Rabun County, Georgia
"Where Spring Spends the Summer"
http://www.acme-brain.com or http://www.rabun.net
ken@rabun.net 1-706-782-9239
Subject:Re: (usr-tc) Problem with encrypted passwords From: Theodore Cekan <ted@mho.net> Date: 1998-11-10 22:43:38
Yes, it is set to pap.
-----Original Message-----
>Did you use the set ppp receive_authentication command pap? The HARC
>might default to chap which I know does not work with our RADIUS server.
>
>Eric
>---
>Eric J. Lorenzo
>Field Service Engineer
>v:650.237.1465
>Lorenzo@ISPchannel.com
>http://www.ISPchannel.com
Subject:Re: (usr-tc) Problem with encrypted passwords From: Theodore Cekan <ted@mho.net> Date: 1998-11-10 23:43:16
The secret is correct. I can trace Radius, and it compares the valid
password from the database with garbled string it thinks it just decrypted,
and send back a reject.
-----Original Message-----
>Theodore Cekan was heard to say:
>>I have a new Hiper system that wont authenticate users using Radius. The
>>password seems to be sent encrypted and my Radius server cant decrypt it
>>correctly. What could cause this?
>
>Invalid shared secret.
>
>--Ricky
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Ricky Beam <jfbeam@interpath.net> Date: 1998-11-11 00:45:59
Jeff Mcadams was heard to say:
>I've been asking for *years* for a filter or ability to define an
>access-list or something to specify the types of traffic that reset the
>idle-timer....and I don't exaggerate when I say years...I've asked this
>of various vendors...and have just recently found the possibility that
>one has implemented this. I've heard recently that Cisco might have
>done this...but haven't played with it personally, so don't know for
>sure if they have.
I know the fealing... FWIW, MorningStar PPP (software) has this functionality.
(I had to devise odd ways around it in school :-))
>What about it 3Com? Can we *finally* have this feature? Please? The
>code to look into packets is already there, you already filter based on
>ports and protocols...how hard could it be to have the idle-timer check
>a filter for it?
You've obviously never tried to get 3Com to do the logical thing, have you?
RADIUS server crypt() password support has taken over a year to get, and I
had to write it myself -- that's right ~7 lines of code that would have taken
an engineer minutes to create, took over a year and someone else giving them
the code to get it done. I, for one, was moving to RADIUS from a system
that already had crypt()ed passwords -- that is not reversible. I also want
truely encrypted passwords in the database. (FYI, the encrypt()/decrypt()
script functions only set/clear the sign bit (0x80) on each character. It
doesn't take a rocket scientist to see that or undo it -- 'less' will already
strip the sign bit for you!)
--Ricky
Subject:Re: (usr-tc) Problem with encrypted passwords From: Ricky Beam <jfbeam@interpath.net> Date: 1998-11-11 00:48:43
Theodore Cekan was heard to say:
>I have a new Hiper system that wont authenticate users using Radius. The
>password seems to be sent encrypted and my Radius server cant decrypt it
>correctly. What could cause this?
Invalid shared secret.
--Ricky
Ken Hodges was heard to say:
>Does anybody know of a way to de-encrypt the passwords in the TC$A database ?
Yes.
--Ricky
PS: This is the modified thing I used a year ago...
---
#!/usr/bin/perl
## INPUT: insert into USERS (USER_NAME,...) values ('foo',...);
##
## Eg: pg_dump -d [database] -t USERS -da | ./[this thing].pl
$output = "users.sql";
printf "Output: %s\n", $output;
open(OUTPUT, ">$output") || die "Foo! [$?/$|]\n";
while (<STDIN>) {
chop($_);
@INSERT = split(" ", $_, 6);
if (($INSERT[0] eq "insert") && ($INSERT[2] eq "USERS")) {
$INSERT[3] =~ s/(\()(.*)(\)(.*))/$2/;
$INSERT[5] =~ s/(\()(.*)(\)(.*))/$2/;
@KEYS = split(",", $INSERT[3]);
@VALUES = split(",", $INSERT[5], $#KEYS + 1);
for($i=0;$i<=$#KEYS;$i++) {
if($KEYS[$i] =~ /PASSWORD/io) {
### Remove the stupid encryption
@THING = unpack("c*", $VALUES[$i]);
for($c=1;$c<$#THING;$c++) { $THING[$c] &= ~0x80; }
$VALUES[$i] = pack("c*", @THING);
}
}
printf OUTPUT "insert into %s (%s) values (", $INSERT[2], $INSERT[3];
for($i=0;$i<$#KEYS;$i++) { printf OUTPUT "%s,", $VALUES[$i]; }
printf OUTPUT "%s);\n", $VALUES[$i];
}
}
close(OUTPUT);
Subject:Re: (usr-tc) Problem with encrypted passwords From: Ricky Beam <jfbeam@interpath.net> Date: 1998-11-11 02:55:04
Theodore Cekan was heard to say:
>The secret is correct. I can trace Radius, and it compares the valid
>password from the database with garbled string it thinks it just decrypted,
>and send back a reject.
In the trace, the inbound packet off the wire will have the password md5 hashed.
If the secrets are not identical, then when the RADIUS server undoes the hash,
it will still have "trash". It just takes a few seconds to make sure all the
secrets are set correctly everywhere -- the RADIUS server may have to be
restarted to load the secret if it's been changed.
--Ricky
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-11 07:47:22
Thus spake Brian Elfert
>On Tue, 10 Nov 1998, Jeff Mcadams wrote:
>> I've been asking for *years* for a filter or ability to define an
>> access-list or something to specify the types of traffic that reset the
>> idle-timer....and I don't exaggerate when I say years...I've asked this
>> of various vendors...and have just recently found the possibility that
>> one has implemented this. I've heard recently that Cisco might have
>> done this...but haven't played with it personally, so don't know for
>> sure if they have.
>So, you filter out one thing. Anyone who really wants to get around idle
>timeouts will just switch to something.
>This would really only help if customers are accidently leaving an email
>check running.
Actually, we used to do this...on a UNIX based pppd that supported it,
and it worked *quite* well, particularly since you could specify whether
the packets were incoming or outgoing as well on the filter rules. No,
it wasn't perfect, but it was a far sight better than what we've got on
access server equipment.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-11 07:50:54
Thus spake Ricky Beam
>Jeff Mcadams was heard to say:
>>I've been asking for *years* for a filter or ability to define an
>>access-list or something to specify the types of traffic that reset the
>>idle-timer....and I don't exaggerate when I say years...I've asked this
>>of various vendors...and have just recently found the possibility that
>>one has implemented this. I've heard recently that Cisco might have
>>done this...but haven't played with it personally, so don't know for
>>sure if they have.
>I know the fealing... FWIW, MorningStar PPP (software) has this functionality.
>(I had to devise odd ways around it in school :-))
Yup, we used to use MorningStar and used that....was *quite* nice.
That's where I got the idea for this feature. I remember asking
Livingston about the possibility of this back in the 3.1.4 ComOS days
(MZ might remember me hollering about it back then :) Also asked Xyplex
for it (and for an active routing protocol which then never came up with
at all...not even RIP)...that was *before* we had the Livingston stuff.
>RADIUS server crypt() password support has taken over a year to get, and I
>had to write it myself
Heh...my MPIP bug (ticket number 58316) is just now being really worked
on and its been 7 1/2 months or more since I first opened the ticket.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-11 07:55:02
Thus spake Ken Hodges
>If you can add to this the ability to also set a max session time, I'd be
>interested in purchasing this program from you !
Max session times are easily set in RADIUS...and work...at least from
our experience with them.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) Problem with encrypted passwords From: Tatai SV Krishnan <tkrishna@bubba.ae.usr.com> Date: 1998-11-11 08:57:31
set your ppp authentication prefrence to pap
set ppp auth pap
Your default setting on the Hiper arc has ppp authentication as follows
PPP AUTHENTICATION
DIAL_IN Users Authenticate: ANY
PPP Authentication Preference: DEFAULT
System Transmit Authentication Name: HiPer
This will try chap first and this should be your problem. type the above
command
set ppp auth pap
and your ppp set up will look like
PPP AUTHENTICATION
DIAL_IN Users Authenticate: ANY
PPP Authentication Preference: PAP
Sy
krish
\ T.S.V. Krishnan \
\ Network System Engineer \ ( : - : )
\ 3Com ............ \
----------------------------------------------/
tkrishna@bubba.ae.usr.com
----------------------------/ http://interproc.ae.usr.com ----/
The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
Any Sufficiently advanced bug is indistinguishable for a feature.
- Rick Kulawiec
On Tue, 10 Nov 1998, Theodore Cekan wrote:
> I have a new Hiper system that wont authenticate users using Radius. The
> password seems to be sent encrypted and my Radius server cant decrypt it
> correctly. What could cause this?
>
> Ted
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) Per-User Filtering using Bess or XStop From: Randy Cosby <dcosby@infowest.com> Date: 1998-11-11 11:45:13
We are looking at running a filtering system like Xstop or N2H2 Bess.
If you are running such a box, and are filtering on a per-user-basis, I've
got some questions.
I'm running both Netserver and HiperARC-based boxes. I've seen some posts
about using "VPN Neighbors", to route traffic on a per-user basis to the
box, but the docs say that is unsupported. Does it work? Any example
implementations you'd care to share? I'm using Radiator, so I can do just
about anything I need to on the Radius side.
Thanks,
Randy Cosby <dcosby@infowest.com>
Vice President
InfoWest Global Internet Services, Inc.
(435)674-0165 http://www.infowest.com
Subject:Re: (usr-tc) Problem with encrypted passwords From: Theodore Cekan <ted@mho.net> Date: 1998-11-11 12:10:31
Im not sure what the problem was. I reinstalled version 4.1.11 and now it
works. Go figure.
-----Original Message-----
>Theodore Cekan was heard to say:
>>The secret is correct. I can trace Radius, and it compares the valid
>>password from the database with garbled string it thinks it just
decrypted,
>>and send back a reject.
>
>In the trace, the inbound packet off the wire will have the password md5
hashed.
>If the secrets are not identical, then when the RADIUS server undoes the
hash,
>it will still have "trash". It just takes a few seconds to make sure all
the
>secrets are set correctly everywhere -- the RADIUS server may have to be
>restarted to load the secret if it's been changed.
>
>--Ricky
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
Mario M. Bustamante wrote:
.skipped.
> We have somewhat over 2,200 users, and we have always recommended
> that they buy USR/3Com modems, so we were shocked at the number
> of Sportster users having problems (and they were mad at us for
> making them spend extra money to buy more problems). We were also
> shocked to find that a lot of them were connecting to BellSouth
> with a much better V90 connection (they use Ascend equipment).
We use NETServer and Qmodems. Since our introduction of V.90
in September, at least 2 Sportsters users complained that
their couldn't connect at all or their connections were dropped
soon after being connected. In the end, they had to DISABLE
V.90 on their Sportsters and got stable and 50-ish K
X2 connections. One the other hand, there was one happy
Sportsters user who was constantly getting 53k with V.90.
But that was the ONLY happy Sportsters user!
Looks like there is work to be done on the V.90 codes?
fbsd
Subject:(usr-tc) HiperARC.. [resend, first one never hit the list?] From: System Administrator <sysadmin@evcom.net> Date: 1998-11-11 14:17:07
Config'ing first HiperARC (btw, must say that CLI is MUCH better than
netserver), couple of questions:
1. Is there still a known problem with chassis awareness (w/ latest non-ER
code). Only one HARC, so should I do chassis slot assignments manually?
2. Am I to understand that the Idle-Timeout RADIUS attribute does not work
correctly with the HARC, and that I must use a VA (we have a highly customized
radius daemon, which supports loads of goodies, but not VA)?
Many thanks,
--
Jesse Sipprell
Technical Operations Director
Evolution Communications, Inc.
800-496-4736 (ext 106)
* Finger sysadmin@evcom.net for my PGP Public Key *
PS. Extra-special-thanks to the fine folks at interpath for providing
http://enterprise.interpath.net/~jfbeam/USR/hiperARC-setup
;=)
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: MegaZone <megazone@megazone.org> Date: 1998-11-12 02:44:28
Once upon a time Jeff Mcadams shaped the electrons to say...
>What about it 3Com? Can we *finally* have this feature? Please? The
>code to look into packets is already there, you already filter based on
>ports and protocols...how hard could it be to have the idle-timer check
>a filter for it?
Lucent doesn't do it because they determined that this could be a performance
hit. It would be yet *another* comparison to be done on every packet, on
top of any real filtering that is being done. And:
1. Because of the risk of a performance hit (You know users would overdue
it - then blame the vendor.)
and
2. There is only a vocal minority asking for this, not a lot of users.
It was decided against for the time being.
I've only really seen the same handful of users asking for something like
this for a few years now - not many overall. Nice idea in general, but
it could be sticky in particulars.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: MegaZone <megazone@megazone.org> Date: 1998-11-12 02:46:10
Once upon a time William Behrens shaped the electrons to say...
> We are working on a program (runs on NT/SQL) that will poll the TC's and
>kick multipule logins off. With user definable boot parameters like kicking
>by idle time or newest login and validation before the actual kicking
It sounds like you're largely recreating TSMON...
What are the main differences between your program and TSMON?
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Once upon a time Theodore Cekan shaped the electrons to say...
>What are radius attributes 51 and 50?
When in doubt RTFRFC:
5.11. Acct-Multi-Session-Id
Description
This attribute is a unique Accounting ID to make it easy to link
together multiple related sessions in a log file. Each session
linked together would have a unique Acct-Session-Id but the same
Acct-Multi-Session-Id. It is strongly recommended that the Acct-
Multi-Session-Id be a printable ASCII string.
A summary of the Acct-Session-Id attribute format is shown below.
The fields are transmitted from left to right.
0 1 2
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type | Length | String ...
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Rigney Informational [Page 20]
RFC 2139 RADIUS Accounting April 1997
Type
50 for Acct-Multi-Session-Id.
Length
>= 3
String
The String field SHOULD be a string of printable ASCII characters.
5.12. Acct-Link-Count
Description
This attribute gives the count of links which are known to have
been in a given multilink session at the time the accounting
record is generated. The NAS MAY include the Acct-Link-Count
attribute in any Accounting-Request which might have multiple
links.
A summary of the Acct-Link-Count attribute format is show below. The
fields are transmitted from left to right.
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type | Length | Value
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Value (cont) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Type
51 for Acct-Link-Count.
Length
6
Value
The Value field is four octets, and contains the number of links
seen so far in this Multilink Session.
Rigney Informational [Page 21]
RFC 2139 RADIUS Accounting April 1997
It may be used to make it easier for an accounting server to know
when it has all the records for a given Multilink session. When
the number of Accounting-Requests received with Acct-Status-Type =
Stop and the same Acct-Multi-Session-Id and unique Acct-Session-
Id's equals the largest value of Acct-Link-Count seen in those
Accounting-Requests, all Stop Accounting-Requests for that
Multilink Session have been received.
An example showing 8 Accounting-Requests should make things
clearer. For clarity only the relevant attributes are shown, but
additional attributes containing accounting information will also
be present in the Accounting-Request.
Multi-Session-Id Session-Id Status-Type Link-Count
"10" "10" Start 1
"10" "11" Start 2
"10" "11" Stop 2
"10" "12" Start 3
"10" "13" Start 4
"10" "12" Stop 4
"10" "13" Stop 4
"10" "10" Stop 4
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: MegaZone <megazone@megazone.org> Date: 1998-11-12 02:54:26
Once upon a time Randy Cosby shaped the electrons to say...
>I'll put in my two cents for Radiator (http://www.open.com.au/radiator). It
>does checking two ways. First, it checks the database to see if a user is
>logged on. The database is user-definable. We log it to mysql, and can
>then output neat web pages in PHP to see who's online. If it does find the
>user in the DB too many times, it does a double-check using a pmmon type
>program. You can use different "checker" programs for different boxes.
Same as Cistron really. It keeps a DB - opens records for a START, closes
them for a STOP. It does SNMP checking for units from Lucent, Cisco,
and the 3Com HiPer, finger for units like Ascend, and a telnet command
line parser for units like the 3Com NetServer.
Radiator and Cistron have many of the same features - only Cistron is free.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Subject:Re: (usr-tc) sr0netserv xxx- assigned IP pool out of addresses! From: Ricky Beam <jfbeam@interpath.net> Date: 1998-11-12 06:24:15
Ralph Helfenberger was heard to say:
>We have a problem, that after some time the Netserver runs out of IP
>adresses.There are enough addresses assigned. It looks like the the IP
>adresses are not released after a connection.
>
>Configuration:
> Netserver 3.8.1
> RADIUS
>The setup is a bit special because users, who are not in the RADIUS
>database are given to an NT Server as PPTP users. They will be
>authenticated there. This means that we have PPP users and PPTP users on
>the same chassis.
>
>Has anybody seen similar problems?
Yes, I have... there appears to be two bugs. First, there is some sort of
error with those netservers with "standard" packetbus hardware where it
never realizes the session is closed and thus leases the IP. Or so that's
what I've heard. I never got a straight answer about it.
The second is an error with handling multi-link PPP sessions. I have an ER
release that doesn't seem to be affected by it (3.8.95.) It would appear the
netserver is allocating too many addresses for MLPPP sessions and then
forgetting it handed them out.
NetS> ver
U.S. Robotics
Total Control (tm) NETServer Card V.34/ISDN with Frame Relay V3.8.95
Build date: Sep 10 1998
Build time: 12:43:51
Network Interface Card: Ethernet & Frame Relay Combination (26)
ISDN Interface Card : MUNICH32 (4)
Packet Bus Circuit : Enhanced
NetS> show uptime
System has been up for 35900 seconds (0 days 9 hrs 58 min 20 sec)
NetS> show ippool
...
pool 192.168.38.193 62 1
...
NetS> show ses
...
S45 Ptemp02 192.168.38.195 Netwrk In ESTABLISHED 9:53 5
S46 < MPlink: S45 > 192.168.38.195 Netwrk In ESTABLISHED 9:52 -
...
I switched to 3.8.95 only because of leaking addresses... there may be other
problems I haven't seen yet. One problem all the 3.8 code has had is errors
in handling dialout. 3.8.95 seems to always say "NO DIAL TONE" -- but at least
I'm seeing the modem's messages. I'm not sure what 3.8.95 was supposed to
fix for me -- Mike Wronski mailed me the code shortly after 3.8.1 was moved
to "released".
--Ricky
Subject:Re: (usr-tc) sr0netserv xxx- assigned IP pool out of addresses! From: Tatai SV Krishnan <tkrishna@bubba.ae.usr.com> Date: 1998-11-12 07:43:24
In your setup who has the IP pools? The reason I am asking this question
is because of your setup. You have both the PPP and PPTP user setup. In
PPTP all the Pools etc are maintained by NT. IN PPP you can have the
NETServer or the RAdius server assing the IP address.
Depending on your case - if you are using a private pool and if radius is
asking ip address for PPTP users - it may never be released. Need to
know more about your setup in terms of PPTP users and PPP users.
krish
\ T.S.V. Krishnan \
\ Network System Engineer \ ( : - : )
\ 3Com ............ \
----------------------------------------------/
tkrishna@bubba.ae.usr.com
----------------------------/ http://interproc.ae.usr.com ----/
The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
Any Sufficiently advanced bug is indistinguishable for a feature.
- Rick Kulawiec
On Thu, 12 Nov 1998, Ralph Helfenberger wrote:
> We have a problem, that after some time the Netserver runs out of IP
> adresses.There are enough addresses assigned. It looks like the the IP
> adresses are not released after a connection.
>
> Configuration:
> Netserver 3.8.1
> RADIUS
> The setup is a bit special because users, who are not in the RADIUS
> database are given to an NT Server as PPTP users. They will be
> authenticated there. This means that we have PPP users and PPTP users on
> the same chassis.
>
>
> Has anybody seen similar problems?
>
> Ralph
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) sr0netserv xxx- assigned IP pool out of addresses! From: Tatai SV Krishnan <tkrishna@bubba.ae.usr.com> Date: 1998-11-12 07:47:19
On Thu, 12 Nov 1998, Ricky Beam wrote:
> Ralph Helfenberger was heard to say:
> >We have a problem, that after some time the Netserver runs out of IP
> >adresses.There are enough addresses assigned. It looks like the the IP
> >adresses are not released after a connection.
> >
> >Configuration:
> > Netserver 3.8.1
> > RADIUS
> >The setup is a bit special because users, who are not in the RADIUS
> >database are given to an NT Server as PPTP users. They will be
> >authenticated there. This means that we have PPP users and PPTP users on
> >the same chassis.
> >
> >Has anybody seen similar problems?
>
> Yes, I have... there appears to be two bugs. First, there is some sort of
> error with those netservers with "standard" packetbus hardware where it
> never realizes the session is closed and thus leases the IP. Or so that's
> what I've heard. I never got a straight answer about it.
Hmm... Hardware - causing a Software problem? Enhanced packet bus and
standard packet bus have nothing do with IP pools. You have other
problems with standard packet bus but loosing ip address is not one of them.
In the above setup one of the possible reasons the customer is loosing IP
address may be because of his way of assining IP addresses - meaning he
may be giving out ip address for PPTP users which may be lost and never
be released.
>
> The second is an error with handling multi-link PPP sessions. I have an ER
> release that doesn't seem to be affected by it (3.8.95.) It would appear the
> netserver is allocating too many addresses for MLPPP sessions and then
> forgetting it handed them out.
>
Totally different issue. A lot of customer are using 3.8.1 and this is
the first instance I have heard about loosing IP address.
krish
> NetS> ver
> U.S. Robotics
> Total Control (tm) NETServer Card V.34/ISDN with Frame Relay V3.8.95
> Build date: Sep 10 1998
> Build time: 12:43:51
>
> Network Interface Card: Ethernet & Frame Relay Combination (26)
> ISDN Interface Card : MUNICH32 (4)
> Packet Bus Circuit : Enhanced
>
> NetS> show uptime
> System has been up for 35900 seconds (0 days 9 hrs 58 min 20 sec)
> NetS> show ippool
> ...
> pool 192.168.38.193 62 1
> ...
> NetS> show ses
> ...
> S45 Ptemp02 192.168.38.195 Netwrk In ESTABLISHED 9:53 5
> S46 < MPlink: S45 > 192.168.38.195 Netwrk In ESTABLISHED 9:52 -
> ...
>
> I switched to 3.8.95 only because of leaking addresses... there may be other
> problems I haven't seen yet. One problem all the 3.8 code has had is errors
> in handling dialout. 3.8.95 seems to always say "NO DIAL TONE" -- but at least
> I'm seeing the modem's messages. I'm not sure what 3.8.95 was supposed to
> fix for me -- Mike Wronski mailed me the code shortly after 3.8.1 was moved
> to "released".
>
> --Ricky
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-12 08:18:26
Thus spake MegaZone
>Once upon a time Jeff Mcadams shaped the electrons to say...
>>What about it 3Com? Can we *finally* have this feature? Please? The
>>code to look into packets is already there, you already filter based on
>>ports and protocols...how hard could it be to have the idle-timer check
>>a filter for it?
>Lucent doesn't do it because they determined that this could be a performance
>hit. It would be yet *another* comparison to be done on every packet, on
>top of any real filtering that is being done.
How much of a performance hit would there be? You only actually have to
look into the packet and get the values once, yes, its another
comparison, but that shouldn't be that hard of a hit at all. Especially
if your filtering code is up to snuff. Like, for example, we don't
assign or define any filters on our NETServers (nor did we on our PM's
when we used them), *please* don't tell me that the code still goes
through the process of looking into the packet and checking filters if
they're non-existant in that case.
>1. Because of the risk of a performance hit (You know users would overdue
>it - then blame the vendor.)
Its called a feature, *any* feature has the possibility of doing this.
Cisco has had policy routing in their routers for quite some time
*knowing* that its a *big* performance hit because it had to be process
switched...they just recently got it to be fast-switched...and even then
not completely...but the feature was there anyway because some people
could really use it.
>2. There is only a vocal minority asking for this, not a lot of users.
>It was decided against for the time being.
OK, let's here it folks...how many people think this would be very
useful? Maybe you hadn't though of it previously, but it could be
useful to you. Again, the feature being requested is a filter to define
what packets reset the idle timer on a connection, so you could say for
example
deny tcp port 109
deny tcp port 110
deny icmp
or something along those lines, and even if your user were checking mail
every 2 minutes or pinging a system out on the 'net, your system could
still disconnect them with an idle timeout.
Let's hear from you if you think this is a good idea since NAS vendors
obviously can't be trusted to figure out a good idea on their own.
>I've only really seen the same handful of users asking for something like
>this for a few years now - not many overall. Nice idea in general, but
>it could be sticky in particulars.
Oh, how hard can it be?
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:(usr-tc) Connection problems into DSP From: Eric <elorenzo@ispchannel.com> Date: 1998-11-12 08:19:50
We just replaced two NETServers in the field with HARCs loaded with TCS
3.3 software. The HiPer DSPs they're owners of are still running TCS 3.1.
We're getting reports of users getting disconnected during the
authorization process. They say they have to attempt a connection 3 or 4
times before they can get logged in.
Could this be an issue with different TCS levels on the cards or does
something just need rebooting?
Eric
Subject:(usr-tc) Follow up after ISPcon San Jose. From: Tim Patterson <tim@harborside.com> Date: 1998-11-12 08:45:13
3com reps at the ISPcon in San Jose promised a call back regarding the
failed TC tech support situation at 3com. So far I have not heard
anything, did anyone at the show get a call?
Tim Patterson
Harborside Internet
541-469-8844 800-680-8855
FAX 541-469-9163
Subject:RE: (usr-tc) Solution for idle time limits and dual logins From: Behrens, William <wbehrens@paracom.com> Date: 1998-11-12 08:59:30
>
> It sounds like you're largely recreating TSMON...
>
> What are the main differences between your program and TSMON?
>
> -MZ
> --
> <URL:mailto:megazone@megazone.org> Gweep, Discordian, Author,
> Engineer, me..
> Join ISP/C Internet Service Providers' Consortium
> <URL:http://www.ispc.org/>
> "A little nonsense now and then, is relished by the wisest
> men" 781-788-0130
> <URL:http://www.megazone.org/> <URL:http://www.gweep.net/>
> Hail Discordia!
>
Well its a long story. First the person who wrote PMMON and variants
used to work at our company (when it was Future Net). That code was taken
with him (source included) when he left (his name is not important and
doesn't need to be mentioned here in this forum). This application runs on
Unix and uses port 23 telnet sessions to retrieve information from the NAS.
We are writing this application using the protocol that PMconsole /
Netserver Manager uses to communicate with the NAS. This has considerable
advantages as it decreases the polling time and puts less load on the NAS.
We are also using SQL server as the database to store the users and
information that is associated with them. We then have access to this data
for other purposes. Customer service managers can run reports on
availability vs. time of day etc. without programming (link to data from
access or some other evil Microsoft product). We can keep long term stats,
baselining the NAS usage to pre-order equipment and circuits. Once the data
is collected there is a whole host of things we can do with it.
I know that TSMON is very powerful. The developer is a well know
person here in Wichita and is very good at what he does. There are a lot of
people who would like a NT option though. This program is not intended to be
marketed or sold. We are just replacing what we (as a company) believe
should be ours to begin with, where we have control of the source code and
can do with it what we will. The current developer has signed a Proprietary
Rights Agreement so we shouldn't have future ownership issues. We have never
had any intentions of legal action against the developer of PMMON/TSMON
(even though he now makes money with it). We (ParaCom) just feel that its
silly to purchase something you feel you already own (its a moral issue with
the executives of the company).
William Behrens
Director of Network Operations
ParaCom Technologies Inc. (formerly Feist Internet / Future Net)
www.feist.com
www.paracom.com
wbehrens@paracom.com
Subject:Re: (usr-tc) sr0netserv xxx- assigned IP pool out of addresses! From: Ricky Beam <jfbeam@interpath.net> Date: 1998-11-12 10:05:02
Tatai SV Krishnan was heard to say:
>> Yes, I have... there appears to be two bugs. First, there is some sort of
>> error with those netservers with "standard" packetbus hardware where it
>> never realizes the session is closed and thus leases the IP. Or so that's
>> what I've heard. I never got a straight answer about it.
>Hmm... Hardware - causing a Software problem? Enhanced packet bus and
>standard packet bus have nothing do with IP pools. You have other
>problems with standard packet bus but loosing ip address is not one of them.
>In the above setup one of the possible reasons the customer is loosing IP
>address may be because of his way of assining IP addresses - meaning he
>may be giving out ip address for PPTP users which may be lost and never
>be released.
As I said, I never got a straight answer, but the ER code I was given for
the "standard" netservers stopped it from losing every address it had --
named or otherwise. And it's not the hardware perse, but more the code
being written for newer hardware not working correctly on the older hardware.
As I was lead to believe, 3.7.24 on "standard" netservers would sometimes
not close the packet bus session correctly which means it never released
an address assigned to that session. It also means it would eventually
run out of packet bus sessions. I'm not saying that's what was happening;
it's just what I was lead to believe -- the helpdesk filter applies.
>> The second is an error with handling multi-link PPP sessions. I have an ER
>> release that doesn't seem to be affected by it (3.8.95.) It would appear the
>> netserver is allocating too many addresses for MLPPP sessions and then
>> forgetting it handed them out.
>
>Totally different issue. A lot of customer are using 3.8.1 and this is
>the first instance I have heard about loosing IP address.
I never used 3.8.1 code in production. It's buggy. I pointed out the bugs
TWICE while it was BETA. It makes me think no one is bothering to pay any
attention at all to the beta mail lists setup for feedback. It's either
that or 3Com really doesn't give a damn about the netservers. (Or both.)
And I and several others have complained about netservers gradually losing
IP addresses before. I spent over an hour teaching the tech support toady
that the last number in 'show ippool' is the number of addresses from that
pool the netserver thinks is in use. [I hate tech support toady's. But,
it's not their fault nobody ever teaches them about these things -- altho',
I'd much rather hear "I don't know" than be feed a line of BS.]
--Ricky
Subject:Re: (usr-tc) sr0netserv xxx- assigned IP pool out of addresses! From: Tatai SV Krishnan <tkrishna@bubba.ae.usr.com> Date: 1998-11-12 10:08:27
On Thu, 12 Nov 1998, Ralph Helfenberger wrote:
> As far as I can see the IP Pools are maintained in the Netserver. But
> this is where the problem could come from. I use the 3Com Security and
> Accounting server. If a user connects wich is not in the RADIUS database
> then this user will be authenicated as the default user in RADIUS. Now
> the default User has the protocol setup as PPTP. He will be forwarded to
> the configured NT Machine... The question is whether an IP Adress will
> be used for this connection (although it shouldn't, because NT is giving
> this user an adress) and if yes, what happens when the user disconnects?
>
The problem here is that the Radius server is asking for an IP address
during setup for the a PPTP user. This is wrong, It should not ask for
one. Ofcourse this is a bug
The general setup for pptp should be
user Password = password,
user-service-type = framed-user
framed-protocol = pptp
login-host = ip address of nt pptp
but with the 3com radius it is asking for an IP address, if you do not
assign it will ask the netserver to assign which is wrong.
The workaround:
Either give the default user an ip address such as 10.10.10.10 - the ip
address is never used so it does not make a difference
or set a private pool and specify the user to that pool
krish
> Ralph
>
>
>
> Tatai SV Krishnan wrote:
> >
> > In your setup who has the IP pools? The reason I am asking this question
> > is because of your setup. You have both the PPP and PPTP user setup. In
> > PPTP all the Pools etc are maintained by NT. IN PPP you can have the
> > NETServer or the RAdius server assing the IP address.
> >
> > Depending on your case - if you are using a private pool and if radius is
> > asking ip address for PPTP users - it may never be released. Need to
> > know more about your setup in terms of PPTP users and PPP users.
> >
> > krish
> >
> > -----------------------------------------
> > \ T.S.V. Krishnan \
> > \ Network System Engineer \ ( : - : )
> > \ 3Com ............ \
> > ----------------------------------------------/
> > tkrishna@bubba.ae.usr.com
> > ----------------------------/ http://interproc.ae.usr.com ----/
> > The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
> > -------------------------------------------------------------------------\
> > Any Sufficiently advanced bug is indistinguishable for a feature.
> > - Rick Kulawiec
> > -------------------------------------------------------------------------/
> >
> > On Thu, 12 Nov 1998, Ralph Helfenberger wrote:
> >
> > > We have a problem, that after some time the Netserver runs out of IP
> > > adresses.There are enough addresses assigned. It looks like the the IP
> > > adresses are not released after a connection.
> > >
> > > Configuration:
> > > Netserver 3.8.1
> > > RADIUS
> > > The setup is a bit special because users, who are not in the RADIUS
> > > database are given to an NT Server as PPTP users. They will be
> > > authenticated there. This means that we have PPP users and PPTP users on
> > > the same chassis.
> > >
> > >
> > > Has anybody seen similar problems?
> > >
> > > Ralph
> > >
> > > -
> > > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > > with "unsubscribe usr-tc" in the body of the message.
> > > For information on digests or retrieving files and old messages send
> > > "help" to the same address. Do not use quotes in your message.
> > >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Lon R. Stockton, Jr. <lon@moonstar.com> Date: 1998-11-12 11:00:40
On Thu, 12 Nov 1998, Jeff Mcadams wrote:
> or something along those lines, and even if your user were checking mail
> every 2 minutes or pinging a system out on the 'net, your system could
> still disconnect them with an idle timeout.
I love the idea...if my competitors did such things, I could really
eat them alive in the marketing arena. And I'm sure that the customers
I couldn't steal from them at that point would gladly switch to using
RealAudio streams to keep their connection up, sapping both the port
and the bandwidth.
So count me as a yes vote. I wouldn't use the feature myself of course,
but I could hope my competition would be short-sighted enough to do so.
Subject:(usr-tc) sr0netserv xxx- assigned IP pool out of addresses! From: Ralph Helfenberger <rhelfenberger@comlight.ch> Date: 1998-11-12 11:45:08
We have a problem, that after some time the Netserver runs out of IP
adresses.There are enough addresses assigned. It looks like the the IP
adresses are not released after a connection.
Configuration:
Netserver 3.8.1
RADIUS
The setup is a bit special because users, who are not in the RADIUS
database are given to an NT Server as PPTP users. They will be
authenticated there. This means that we have PPP users and PPTP users on
the same chassis.
Has anybody seen similar problems?
Ralph
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-12 12:02:06
Thus spake Lon R. Stockton, Jr.
>On Thu, 12 Nov 1998, Jeff Mcadams wrote:
>> or something along those lines, and even if your user were checking mail
>> every 2 minutes or pinging a system out on the 'net, your system could
>> still disconnect them with an idle timeout.
>I love the idea...if my competitors did such things, I could really
>eat them alive in the marketing arena. And I'm sure that the customers
>I couldn't steal from them at that point would gladly switch to using
>RealAudio streams to keep their connection up, sapping both the port
>and the bandwidth.
If you pick up my abusive users that are pulling stuff like that...more
power to you...they can eat up your ports without giving any extra
revenue stream.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Richard Stuplich <dick@dwave.net> Date: 1998-11-12 12:19:53
At 01:02 PM 11/12/98 -0500, you wrote:
>
>On Thu, 12 Nov 1998, Jeff Mcadams wrote:
>
>> If you pick up my abusive users that are pulling stuff like that...more
>> power to you...they can eat up your ports without giving any extra
>> revenue stream.
>
>Ah, but they do provide extra revenue, since I don't participate in the
>lie commonly called 'unlimited service'. After 150 hours, they pay
>extra. Not only do I get to sidestep the idle_timer issue, I can sidestep
>the max_session_length and max_sessions issues as well.
>
We offer unlimited access and mean it. If you think an ISP can't offer
unlimited, and mean, it you are sorely mistaken. Unfortunately most ISPs
oversell bandwidth and offer non-unlimited unlimited. This reflects poorly
on all ISPs. I do admire an ISP that will not offer unlimited if they are
not willing to deliver it, as you. To say that an ISP can't offer
unlimited and mean it is wrong.
Richard B. Stuplich DataWave, Not just faster, Better.
President, DataWave Technologies 17 T1 circuits and growing strong.
DataWave, Wausau's first local ISP to have a direct connection to the
midwest NAP, Frame Relay, ISDN, x2, k56Flex, v.90. The only area ISP to
have redundant T1 NAP connections, All modem compatibility and a staff
dedicated exclusively to providing Internet Service to this area.
This E-Mail Copyright 1998 Richard Stuplich, see http://dw.net/cr/
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Brian Elfert <brian@citilink.com> Date: 1998-11-12 12:46:21
On Thu, 12 Nov 1998, Lon R. Stockton, Jr. wrote:
>
> On Thu, 12 Nov 1998, Jeff Mcadams wrote:
>
> > or something along those lines, and even if your user were checking mail
> > every 2 minutes or pinging a system out on the 'net, your system could
> > still disconnect them with an idle timeout.
>
> I love the idea...if my competitors did such things, I could really
> eat them alive in the marketing arena. And I'm sure that the customers
> I couldn't steal from them at that point would gladly switch to using
> RealAudio streams to keep their connection up, sapping both the port
> and the bandwidth.
So, you want all the customers that stay online all day, but do nothing
while online except check mail every few minutes?
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Lon R. Stockton, Jr. <lon@moonstar.com> Date: 1998-11-12 13:02:24
On Thu, 12 Nov 1998, Jeff Mcadams wrote:
> If you pick up my abusive users that are pulling stuff like that...more
> power to you...they can eat up your ports without giving any extra
> revenue stream.
Ah, but they do provide extra revenue, since I don't participate in the
lie commonly called 'unlimited service'. After 150 hours, they pay
extra. Not only do I get to sidestep the idle_timer issue, I can sidestep
the max_session_length and max_sessions issues as well.
I know of hundreds of people that have called or email USR/3COM
about the sportster v.90 issue.
If you are denying the problem, then someone at your company is
throwing away all the messages before anyone can see them.
Or you are hiding the truth.
Which is it???
At 09:15 PM 11/7/98 -0600, you wrote:
>>OK, but a fair number of the people having problems have Sportsters.
>>(Some Winmodems, some not.) I wasn't aware there was any new Sportster
>>code out... if there is, where's it hidden? If there isn't, is this one
>>of the server-side problems?
>
>I was responding to your comments on LT Winmodems, you did not mention
>Sportsters. I know of no significant issues with interop with Sportsters
>(Winmodem or controller based) and Total Control modems.
>
>There will likely be improvements with the client side (all vendors) for
>quite some time. There are quite a few tweeks to enhance performance and
>coverage in the labs already. Heck, we are still improving V.34 4 years
>later.
>
>If you can provide details on problems with Sportsters, I will gladly take
>this off-line and put you in contact with my peers on the PCD (home of the
>Sportster) side of the house.
>
>JP
>
>PS. No, there are no recent releases of Sportster code related to V.90
>fixes. Overall, they are working quite well.
>
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Carl Jagerski
Network Administrator, Forward Communications
carll@forcomm.net
412-378-4490 Fax: 412-375-0156
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Brian Elfert <brian@citilink.com> Date: 1998-11-12 13:35:27
On Thu, 12 Nov 1998, Richard Stuplich wrote:
> We offer unlimited access and mean it. If you think an ISP can't offer
> unlimited, and mean, it you are sorely mistaken. Unfortunately most ISPs
You are NOT offering truly unlimited service Here is a quote from your
web page: "Unlinited active usage per month included" (mispellings
included).
Unlimited is defined as "without limits". Requiring someone to be active
is a limit. Just because you may not have a session limit or an idle
timeout doesn't mean you offer a truly unlimited service.
> Richard B. Stuplich DataWave, Not just faster, Better.
> President, DataWave Technologies 17 T1 circuits and growing strong.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Your whole web page is a bunch of marketing crapola. You're trying to
decieve people into thinking you have 17 T1s to the Internet, and you
don't. Your own web page has a press release announcing the second T1 to
the Internet in Sept '98. I'll bet that 15 of those T1s are really for
dialin or some purpose besides connecting to the Internet.
Your web page also says that you will have enough bandwidth to deliver
1.5Mbps to each customer who buys a T1 from you. You better not have more
than 1 T1 customer then, because you aren't providing enough bandwidth so
that each T1 customer could use a full 1.5Mbps. If you really have a T1's
worth of bandwidth sitting around for each T1 customer, how could you sell
for any less than your Internet backbones?
Until you are buying bandwidth at the DS3 level, you're not going to get
1.5Mbps of bandwidth to an Internet backbone any cheaper than any of your
T1 customers could buy it direct from a backbone for.
Brian
Subject:(usr-tc) Unlimited can mean unlimited. From: Richard Stuplich <dick@dwave.net> Date: 1998-11-12 14:21:40
>> We offer unlimited access and mean it. If you think an ISP can't offer
>> unlimited, and mean, it you are sorely mistaken. Unfortunately most ISPs
>
>You are NOT offering truly unlimited service Here is a quote from your
>web page: "Unlinited active usage per month included" (mispellings
>included).
>
>Unlimited is defined as "without limits". Requiring someone to be active
>is a limit. Just because you may not have a session limit or an idle
>timeout doesn't mean you offer a truly unlimited service.
When we say U-N-L-I-M-I-T-E-D we mean it. You can misinterpret a "Cover
your ass" statement any way you like. Doesn't change the facts that some
ISPs offer unlimited and mean it. Like us.
Do you have a contract with your users? In that contract do you state that
you make no warranties for the usefulness of your product? In that case
you must not provide a very good service, right?
There is a difference between converging your ass and policy.
I'd post our usage report showing 26+ days of usage for $19.95 unlimited
accounts that we allow access month after month if it wouldn't violate our
non-disclosure contract with our customer.
This is not the place to fight about this. Reply to me privately and I'll
prove we mean it.
Richard B. Stuplich DataWave, Not just faster, Better.
President, DataWave Technologies 17 T1 circuits and growing strong.
DataWave, Wausau's first local ISP to have a direct connection to the
midwest NAP, Frame Relay, ISDN, x2, k56Flex, v.90. The only area ISP to
have redundant T1 NAP connections, All modem compatibility and a staff
dedicated exclusively to providing Internet Service to this area.
This E-Mail Copyright 1998 Richard Stuplich, see http://dw.net/cr/
Subject:(usr-tc) Dynamic IPX on HiPer ARC 4.1.78 -4 and with RADIUS 6.0.8 From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-12 14:46:54
I configured a HiPer ARC running 4.1.11 with an IPX network going out Eth:2.
I configured a user on the HiPer ARC with the IPX network # fffffffc. It
worked like a charm. Then the customer reported trouble with web-TV. We
upgraded to 4.1.78 - 4. Once upgraded the network # fffffffc was not
working. The only way we could get ipx to work at all was to assign a
specific ipx # or to configure an ipx pool. I have two questions: Is this
a bug or did I miss something? Are there plans to include IPX pools in
RADIUS since the HiPer ARC seems to support that attribute.
I already spoke to someone at 3COM. They told me they would get back to me
yesterday, but didn't. Now it's Thursday and they don't work today.
Any comments?
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-12 15:06:04
Thus spake Lon R. Stockton, Jr.
>On Thu, 12 Nov 1998, Jeff Mcadams wrote:
>> If you pick up my abusive users that are pulling stuff like that...more
>> power to you...they can eat up your ports without giving any extra
>> revenue stream.
>Ah, but they do provide extra revenue, since I don't participate in the
>lie commonly called 'unlimited service'.
Neither do we...if you find any place on our pages that says "unlimited
service", please let me know as it definitely shouldn't be there. We
*do* have "flat-rate" but that still allows us to put idle-timers,
max-session timers, and filters on dial-ups. Since you don't have a
true flat-rate, perhaps its me that would eat your lunch in marketing.
We do have abusive users...and we have measures in place to deal with
them...this however, would be one such measure that would *help* (not
absoluately cure) us deal with them.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) Dynamic IPX on HiPer ARC 4.1.78 -4 and with RADIUS 6.0.8 From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-12 15:08:27
Thus spake Brian K McIntire
>I already spoke to someone at 3COM. They told me they would get back to me
>yesterday, but didn't. Now it's Thursday and they don't work today.
>Any comments?
Heh...yeah...welcome to the user's side. :D
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) Unlimited can mean unlimited. From: Mike Andrews <mandrews@termfrost.org> Date: 1998-11-12 15:38:41
Can I ask a favor? Can we dispense with the "unlimited" religious war?
It's been beat to death on the inet-access list several dozen times and
we really don't need to read it here... ;-)
Mike Andrews (MA12) icq 6602506 -------------- mandrews@termfrost.org
VP 'n' Systems/Network Administrator --------------- mandrews@dcr.net
Digital Crescent, Frankfort, KY ----------- http://www.termfrost.org/
# view;touch;unzip;finger;mount;mv;mv;mv;yes;mv;yes;mv;yes;umount;sleep
Subject:RE: (usr-tc) Dynamic IPX on HiPer ARC 4.1.78 -4 and with RADIUS 6.0.8 From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-12 15:42:25
:( Thanks Jeff.
Thus spake Brian K McIntire
>I already spoke to someone at 3COM. They told me they would get back to me
>yesterday, but didn't. Now it's Thursday and they don't work today.
>Any comments?
Heh...yeah...welcome to the user's side. :D
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:RE: (usr-tc) Dynamic IPX on HiPer ARC 4.1.78 -4 and with RADIUS 6.0.8 From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-12 15:42:25
:( Thanks Jeff.
Thus spake Brian K McIntire
>I already spoke to someone at 3COM. They told me they would get back to me
>yesterday, but didn't. Now it's Thursday and they don't work today.
>Any comments?
Heh...yeah...welcome to the user's side. :D
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:Re: (usr-tc) Unlimited can mean unlimited. From: Frank Basso <frank@got.net> Date: 1998-11-12 16:03:49
Take it offline boys.
-----Original Message-----
>>> We offer unlimited access and mean it. If you think an ISP can't offer
>>> unlimited, and mean, it you are sorely mistaken. Unfortunately most
ISPs
>>
>>You are NOT offering truly unlimited service Here is a quote from your
>>web page: "Unlinited active usage per month included" (mispellings
>>included).
>>
>>Unlimited is defined as "without limits". Requiring someone to be active
>>is a limit. Just because you may not have a session limit or an idle
>>timeout doesn't mean you offer a truly unlimited service.
>
>When we say U-N-L-I-M-I-T-E-D we mean it. You can misinterpret a "Cover
>your ass" statement any way you like. Doesn't change the facts that some
>ISPs offer unlimited and mean it. Like us.
>
>Do you have a contract with your users? In that contract do you state that
>you make no warranties for the usefulness of your product? In that case
>you must not provide a very good service, right?
>
>There is a difference between converging your ass and policy.
>
>I'd post our usage report showing 26+ days of usage for $19.95 unlimited
>accounts that we allow access month after month if it wouldn't violate our
>non-disclosure contract with our customer.
>
>This is not the place to fight about this. Reply to me privately and I'll
>prove we mean it.
>
> Richard B. Stuplich DataWave, Not just faster, Better.
> President, DataWave Technologies 17 T1 circuits and growing strong.
>---------------------------------------------------------------------------
-
>DataWave, Wausau's first local ISP to have a direct connection to the
>midwest NAP, Frame Relay, ISDN, x2, k56Flex, v.90. The only area ISP to
>have redundant T1 NAP connections, All modem compatibility and a staff
>dedicated exclusively to providing Internet Service to this area.
>---------------------------------------------------------------------------
-
>This E-Mail Copyright 1998 Richard Stuplich, see http://dw.net/cr/
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) sr0netserv xxx- assigned IP pool out of addresses! From: Ralph Helfenberger <rhelfenberger@comlight.ch> Date: 1998-11-12 16:08:46
As far as I can see the IP Pools are maintained in the Netserver. But
this is where the problem could come from. I use the 3Com Security and
Accounting server. If a user connects wich is not in the RADIUS database
then this user will be authenicated as the default user in RADIUS. Now
the default User has the protocol setup as PPTP. He will be forwarded to
the configured NT Machine... The question is whether an IP Adress will
be used for this connection (although it shouldn't, because NT is giving
this user an adress) and if yes, what happens when the user disconnects?
Ralph
Tatai SV Krishnan wrote:
>
> In your setup who has the IP pools? The reason I am asking this question
> is because of your setup. You have both the PPP and PPTP user setup. In
> PPTP all the Pools etc are maintained by NT. IN PPP you can have the
> NETServer or the RAdius server assing the IP address.
>
> Depending on your case - if you are using a private pool and if radius is
> asking ip address for PPTP users - it may never be released. Need to
> know more about your setup in terms of PPTP users and PPP users.
>
> krish
>
> -----------------------------------------
> \ T.S.V. Krishnan \
> \ Network System Engineer \ ( : - : )
> \ 3Com ............ \
> ----------------------------------------------/
> tkrishna@bubba.ae.usr.com
> ----------------------------/ http://interproc.ae.usr.com ----/
> The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
> -------------------------------------------------------------------------\
> Any Sufficiently advanced bug is indistinguishable for a feature.
> - Rick Kulawiec
> -------------------------------------------------------------------------/
>
> On Thu, 12 Nov 1998, Ralph Helfenberger wrote:
>
> > We have a problem, that after some time the Netserver runs out of IP
> > adresses.There are enough addresses assigned. It looks like the the IP
> > adresses are not released after a connection.
> >
> > Configuration:
> > Netserver 3.8.1
> > RADIUS
> > The setup is a bit special because users, who are not in the RADIUS
> > database are given to an NT Server as PPTP users. They will be
> > authenticated there. This means that we have PPP users and PPTP users on
> > the same chassis.
> >
> >
> > Has anybody seen similar problems?
> >
> > Ralph
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
Subject:Re: (usr-tc) filtering service (fwd) From: Angela A. Burford <aratis@erie.net> Date: 1998-11-12 17:37:23
This message is in MIME format. The first part should be readable text,
while the remaining parts are likely unreadable without MIME-aware tools.
Send mail to mime@docserver.cac.washington.edu for more info.
--------------C9164EDBAC9612E3D30DD268
Content-Type: TEXT/PLAIN; CHARSET=us-ascii
Content-ID: <Pine.LNX.3.95.981112171319.6081C@moose.erie.net>
Hi,
I have some filtering features that work with my livingston pm3's and i
would like to implement them in my usr chassis as well. The feature is
the following:
When my users can access a webpage and customize the setup of their
filters. That information is used to build a filter
for that specific user. The user can change the settings of his filter
anytime he connects. When he calls in, radius passes the name of
the filter for that user to the pm3 and the filter itself is loaded into
the portmaster through choicenet, so that it will be used in that
connection. So a different filter for each user is stored in one of my
central radius serer (a bsdi box running customized livingston radius),
and loaded into the right pm3 each time the user establishes a connection.
I'd like to make this work with my usr's too (i have chassis with
netservers, quad modems nmc and dual pri cards and other with hiperDSP's
hiperARC and nmc cards). I understand that I can
pass the names of the filters to be used at the time the user connects,
but is it possible to load the actual filter that is stored in my
radius server into the hiperARC / netserver so that it will applied to the
connection? If so, how can that be done?
I'd appreciate any information you can give me about the topic.
Thank you very much in advance
Angela
---------- Forwarded message ----------
Reply-To: usr-tc@lists.xmission.com
Packet filters. I've written many. Read up on the packet filters for your
netserver.
Jason Cropper
Craig Thompson wrote:
> Has anyone set up their TC to allow some customers to be sent through a
> filtering service?
>
> A company we are looking at for this says that the PortMaster has a
> config option that says "ChoiceNet Server."
>
> I was wondering if there is a workaround for doing this on the TC?
>
> Please help if you can.
>
> Craig Thompson
> ----------------------------------------------------------------------
> WingNET Internet Services,
> P.O. Box 3000 // Cleveland, TN 37320-3000
> 423-559-LINK (v) 423-559-5444 (f)
> http://www.wingnet.net
> ----------------------------------------------------------------------
>
> Always be sincere, even if you don't mean it.
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
--------------C9164EDBAC9612E3D30DD268--
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Ricky Beam <jfbeam@interpath.net> Date: 1998-11-12 18:04:17
Jeff Mcadams was heard to say:
>Thus spake Lon R. Stockton, Jr.
>>On Thu, 12 Nov 1998, Jeff Mcadams wrote:
>>> or something along those lines, and even if your user were checking mail
>>> every 2 minutes or pinging a system out on the 'net, your system could
>>> still disconnect them with an idle timeout.
>
>>I love the idea...if my competitors did such things, I could really
>>eat them alive in the marketing arena. And I'm sure that the customers
>>I couldn't steal from them at that point would gladly switch to using
>>RealAudio streams to keep their connection up, sapping both the port
>>and the bandwidth.
>
>If you pick up my abusive users that are pulling stuff like that...more
>power to you...they can eat up your ports without giving any extra
>revenue stream.
While I happen to agree that an idleness filter would be a Godd Thing(tm),
I'm afraid it's just a waste of time to code, process, and maintain. No
matter what you assign as "idle traffic", there are 64k or ports to choose
from and hundreds of programs already written to get around the idle timer.
I would help with the casual web browsing user. You could filter out a fair
amount of junk that really should not be considered active traffic (ICMP
in particular.) But it's trivial to get around idle timers. Use a session
timer and maybe a "logged out time" threshold -- but that really is a mess.
I would vote in favor of the feature (on HARC) even if .1% of the user base
would use it.
--Ricky
Subject:RE: (usr-tc) Solution for idle time limits and dual logins From: Mike Andrews <mandrews@termfrost.org> Date: 1998-11-12 18:04:39
On Thu, 12 Nov 1998, Behrens, William wrote:
> We are writing this application using the protocol that PMconsole /
> Netserver Manager uses to communicate with the NAS. This has considerable
> advantages as it decreases the polling time and puts less load on the NAS.
Where is this protocol documented? Anywhere? (Or did you reverse
engineer it?) If this can do the equivalent of "pmwho" faster, I'd be
interested in hacking up some Perl code to do this for my own utilities...
Mike Andrews (MA12) icq 6602506 -------------- mandrews@termfrost.org
VP 'n' Systems/Network Administrator --------------- mandrews@dcr.net
Digital Crescent, Frankfort, KY ----------- http://www.termfrost.org/
# view;touch;unzip;finger;mount;mv;mv;mv;yes;mv;yes;mv;yes;umount;sleep
Subject:Re: (usr-tc) Dynamic IPX on HiPer ARC 4.1.78 -4 and with RADIUS 6.0.8 From: Ricky Beam <jfbeam@interpath.net> Date: 1998-11-12 18:12:39
Brian K McIntire was heard to say:
>Are there plans to include IPX pools in
>RADIUS since the HiPer ARC seems to support that attribute.
I thought SA6 already had that in that mess of a database. If not, it's
trival enough to add -- getting the java widget to manage it is another
story.
--Ricky
Angela A. Burford was heard to say:
>I'd like to make this work with my usr's too (i have chassis with
>netservers, quad modems nmc and dual pri cards and other with hiperDSP's
>hiperARC and nmc cards). I understand that I can
>pass the names of the filters to be used at the time the user connects,
>but is it possible to load the actual filter that is stored in my
>radius server into the hiperARC / netserver so that it will applied to the
>connection? If so, how can that be done?
Well, you can certainly tftp the filter into the HARC, but the netserver is
gonna be a real bitch to do this. Umm, would it be too much to ask to
replace the netservers with ARCs??? (I know the ARC has a bunch of bugs,
but at least 3Com _can_ fix those. The netserver is hopeless.)
--Ricky
Subject:Re: (usr-tc) filtering service (fwd) From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-12 19:49:51
Thus spake Angela A. Burford
>the portmaster through choicenet, so that it will be used in that
>I'd like to make this work with my usr's too (i have chassis with
>netservers, quad modems nmc and dual pri cards and other with hiperDSP's
>hiperARC and nmc cards). I understand that I can
>pass the names of the filters to be used at the time the user connects,
>but is it possible to load the actual filter that is stored in my
>radius server into the hiperARC / netserver so that it will applied to the
>connection? If so, how can that be done?
Can't reasonably be done on NETServers...the HARC's have the ability to
handle "dynamic filters" given a RADIUS server that supports it. Not
having HARC's to really play with extensively, I don't know how to do
this. :)
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: andy <smitha@mach3ww.com> Date: 1998-11-12 20:06:58
While we are talking about timeouts...
How can I set:
#1 an idletimeout and session timeout in the netserver card (3.7.24)
#2 an idletimeout and session timeout in the hiperarc card (4.0.30)
#3 do they work?
-
andrew
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-12 22:21:40
Thus spake andy
>While we are talking about timeouts...
>How can I set:
>#1 an idletimeout and session timeout in the netserver card (3.7.24)
Idle-Timeout = 720
Session-Timeout = 42516
These are in seconds, using Merit RADIUS...your dictionary may be
slightly different.
>#2 an idletimeout and session timeout in the hiperarc card (4.0.30)
I would assume similar, but don't know.
>#3 do they work?
On the NETServers, yes.
Keep in mind (and I'm sure its hard to forget with all the discussion
going on) that its quite easy to defeat idle-timeouts as they stand...a
single packet of *any* type will reset the idle timer.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:RE: (usr-tc) Dynamic IPX on HiPer ARC 4.1.78 -4 and with RADIUS 6.0.8 From: Charles Sprickman <spork@inch.com> Date: 1998-11-12 22:48:28
Does anyone have a complete list of 3com addresses that are receptive to
complaints? Things get better, things get worse, but a carefully worded
"status report" to 3com by anyone that feels some issues are still
unresolved can really only help the situation. Squeaky wheel and all
that...
Charles
On Thu, 12 Nov 1998, Brian K McIntire wrote:
> :( Thanks Jeff.
>
>
> Thus spake Brian K McIntire
> >I already spoke to someone at 3COM. They told me they would get back to me
> >yesterday, but didn't. Now it's Thursday and they don't work today.
> >Any comments?
>
> Heh...yeah...welcome to the user's side. :D
> --
> Jeff McAdams Email: jeffm@iglou.com
> Head Network Administrator Voice: (502) 966-3848
> IgLou Internet Services (800) 436-4456
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
=-----------------= =
| Charles Sprickman Internet Channel |
| INCH System Administration Team (212)243-5200 |
| spork@inch.com access@inch.com |
= =----------------=
Subject:RE: (usr-tc) Solution for idle time limits and dual logins From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-13 00:21:16
First of all, Jeff, go to sleep. :)
Second.....
Just in case you meant how do you set the idle times directly on the
NetServer or Hiper ARC:
The NetServer does not support session timeout without RADIUS
for idletimeout set all idletime 15 (Where 15 is in minutes)
For the HiPer ARC just apply the values you want to the default user. All
users created after that will have the correct settings.
This is very sloppy in my opinion. Better to use the RADIUS.
=======================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= =
= Experts in Remote Access and all areas of NetWork Design, =
= Security, and Implementation. Offering both installation and =
= support, along with remote management and monitoring packages for =
= dedicated clients (specializing in ISP's). =
= Firm partnerships established with all the key players. =
= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
= =
= Serving North America and parts of Canada. Reselling to the world. =
=======================================================================
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jeff Mcadams
Sent: Thursday, November 12, 1998 9:22 PM
Thus spake andy
>While we are talking about timeouts...
>How can I set:
>#1 an idletimeout and session timeout in the netserver card (3.7.24)
Idle-Timeout = 720
Session-Timeout = 42516
These are in seconds, using Merit RADIUS...your dictionary may be
slightly different.
>#2 an idletimeout and session timeout in the hiperarc card (4.0.30)
I would assume similar, but don't know.
>#3 do they work?
On the NETServers, yes.
Keep in mind (and I'm sure its hard to forget with all the discussion
going on) that its quite easy to defeat idle-timeouts as they stand...a
single packet of *any* type will reset the idle timer.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:(usr-tc) How much sleep do I need? From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-13 08:15:06
Thus spake Brian K McIntire
>First of all, Jeff, go to sleep. :)
Go to sleep? I just got up this morning :)
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Once upon a time Brian K McIntire shaped the electrons to say...
>========================================================================
>= Brian K. McIntire bmcintire@commnet.com =
>= Systems Engineer III 317-558-5050 x113 =
>= CommNetPlus, Indianapolis, IN http://www.commnet.com =
>= Your Remote Access and Security Experts! =
>= =
>= Our experienced technicians can design and install your high speed =
>= network for you. From Operating Systems and servers to routers and =
>= firewalls nobody does it better than CommNetPlus. Our Technical =
>= Support staff is available to you 24 hours a day to meet your needs.=
>= Offering remote management and monitoring packages for dedicated =
>= clients. (Specializing in ISP's). "Let us do the work for you!" =
>= =
>= Firm partnerships est. with current and emerging leaders of today's =
>= technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
>= Serving North America and parts of Canada. Reselling to the world. =
>========================================================================
A 17 line .sig... Could you perhaps STOP THAT?! You do NOT need to place
an ad this size on every post.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Subject:Re: (usr-tc) Dynamic IPX on HiPer ARC 4.1.78 -4 and with RADIUS 6.0.8 From: Tatai SV Krishnan <tkrishna@bubba.ae.usr.com> Date: 1998-11-13 09:16:16
Depends on what the IPX system setup was. All you say here is that you
set the ipx for unnumberd initially and then after your upgrade you could
not use unnumbred. Well is there some equipment on the network that is
not using split-horizon? Did anyone poison the route?
Get me some debug or open a ticket and we will have it working.
krish
\ T.S.V. Krishnan \
\ Network System Engineer \ ( : - : )
\ 3Com ............ \
----------------------------------------------/
tkrishna@bubba.ae.usr.com
----------------------------/ http://interproc.ae.usr.com ----/
The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
Any Sufficiently advanced bug is indistinguishable for a feature.
- Rick Kulawiec
On Thu, 12 Nov 1998, Brian K McIntire wrote:
>
> I configured a HiPer ARC running 4.1.11 with an IPX network going out Eth:2.
> I configured a user on the HiPer ARC with the IPX network # fffffffc. It
> worked like a charm. Then the customer reported trouble with web-TV. We
> upgraded to 4.1.78 - 4. Once upgraded the network # fffffffc was not
> working. The only way we could get ipx to work at all was to assign a
> specific ipx # or to configure an ipx pool. I have two questions: Is this
> a bug or did I miss something? Are there plans to include IPX pools in
> RADIUS since the HiPer ARC seems to support that attribute.
> I already spoke to someone at 3COM. They told me they would get back to me
> yesterday, but didn't. Now it's Thursday and they don't work today.
> Any comments?
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) filtering service (fwd) From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-13 09:55:34
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jeff Mcadams
|Sent: Thursday, November 12, 1998 6:50 PM
|To: usr-tc@lists.xmission.com
|Subject: Re: (usr-tc) filtering service (fwd)
|
|
|Thus spake Angela A. Burford
|>the portmaster through choicenet, so that it will be used in that
|
|>I'd like to make this work with my usr's too (i have chassis with
|>netservers, quad modems nmc and dual pri cards and other with hiperDSP's
|>hiperARC and nmc cards). I understand that I can
|>pass the names of the filters to be used at the time the user connects,
|>but is it possible to load the actual filter that is stored in my
|>radius server into the hiperARC / netserver so that it will applied to the
|>connection? If so, how can that be done?
|
|Can't reasonably be done on NETServers...the HARC's have the ability to
|handle "dynamic filters" given a RADIUS server that supports it. Not
|having HARC's to really play with extensively, I don't know how to do
|this. :)
|--
The VSA's for building filters in a RADIUS access accept packet work for both
HARC and NETSERVER.
Thus TFTP is not required. If you wanted to change the filter rules mid-call,
your out of luck with your netserver.
-M
Subject:RE: (usr-tc) Solution for idle time limits and dual logins From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-13 10:00:43
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian K McIntire
|Sent: Friday, November 13, 1998 12:21 AM
|To: usr-tc@lists.xmission.com
|Subject: RE: (usr-tc) Solution for idle time limits and dual logins
|
|The NetServer does not support session timeout without RADIUS
|for idletimeout set all idletime 15 (Where 15 is in minutes)
|
|For the HiPer ARC just apply the values you want to the default user. All
|users created after that will have the correct settings.
|
|This is very sloppy in my opinion. Better to use the RADIUS.
Sloppy? Default user applies to all dial-in users. Not just locally created ones.
This is not sloppy. It's quite convenient. If you set the same Idle timeout or
session timeout for all/most of your users. Just set that on the default user and
remove the attributes from your default user in the RADIUS users file. If a
particular user or set of users gets different values then include them in
RADIUS. RADIUS attributes always take precedence over default user settings.
-M
|=======================================================================
|= Brian K. McIntire bmcintire@commnet.com =
|= Systems Engineer III 317-558-5050 x113 =
|= CommNetPlus, Indianapolis, IN http://www.commnet.com =
|= =
|= Experts in Remote Access and all areas of NetWork Design, =
|= Security, and Implementation. Offering both installation and =
|= support, along with remote management and monitoring packages for =
|= dedicated clients (specializing in ISP's). =
|= Firm partnerships established with all the key players. =
|= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
|= =
|= Serving North America and parts of Canada. Reselling to the world. =
|=======================================================================
|
Isn't this signature a little long?? :) (4/5 lines max in netiquette)
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Brian <signal@shreve.net> Date: 1998-11-13 10:06:44
On Tue, 10 Nov 1998, Mark Lemmert wrote:
> Does anybody have a good solution in place for enforcing idle times and
> multiple logins?
>
> I am currently using PMMON and I get no end of complaints from people
> that pmmon disconnects them at inappropriate times among other things.
>
> If anybody has a different/better way of doing this I'de love to hear about it.
PMMON.
PMMON doesn't disconnect at inappropriate times, users lie. It only
disconnects after a set amount of inactivity. Unless you're talking about
a hard limit you have set (which we do, and set at 24hours).
Brian
>
>
>
> Mark Lemmert cto@athenet.net
> Chief Technical Officer (920)954-9799
> AthEnet Data Exchange http://www.athenet.net
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:RE: (usr-tc) DUN1.3 and MPP From: Terry Kennedy <terry@olypen.com> Date: 1998-11-13 10:07:53
Ours doesn't work unless it hits the same rack.
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian
> Sent: Friday, November 13, 1998 8:28 AM
> To: USRobotics TC Mailing List
> Subject: (usr-tc) DUN1.3 and MPP
>
>
> I have a user trying to connect using DUN1.3 and Win98. He is trying to
> do MPP using 2 USR 56k external sportsters.
>
> Should this work? Both modems connect but then the 2nd one disconnects.
> I have him setup for dual channel, so i am not sure what is going on
> exactly and wasn't sure if their are some known issues here.
>
> Brian
>
>
> --------------------------------------------------------------------------
> Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service
> Provider
> Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) DUN1.3 and MPP From: Eric Forcey <eric@psnw.com> Date: 1998-11-13 10:20:25
We have/had this same problem as well.
According to support @ 3com, they were aware of issues with DUN 1.3 and
MPP saying that it wouldn't work. Haven't heard if there has been a work
around yet
-Eric
On Fri, 13 Nov 1998, Terry Kennedy wrote:
> Ours doesn't work unless it hits the same rack.
>
> > -----Original Message-----
> > From: owner-usr-tc@lists.xmission.com
> > [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian
> > Sent: Friday, November 13, 1998 8:28 AM
> > To: USRobotics TC Mailing List
> > Subject: (usr-tc) DUN1.3 and MPP
> >
> >
> > I have a user trying to connect using DUN1.3 and Win98. He is trying to
> > do MPP using 2 USR 56k external sportsters.
> >
> > Should this work? Both modems connect but then the 2nd one disconnects.
> > I have him setup for dual channel, so i am not sure what is going on
> > exactly and wasn't sure if their are some known issues here.
> >
> > Brian
> >
> >
> > --------------------------------------------------------------------------
> > Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service
> > Provider
> > Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> > signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> > (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) DUN1.3 and MPP From: Brian <signal@shreve.net> Date: 1998-11-13 10:27:55
I have a user trying to connect using DUN1.3 and Win98. He is trying to
do MPP using 2 USR 56k external sportsters.
Should this work? Both modems connect but then the 2nd one disconnects.
I have him setup for dual channel, so i am not sure what is going on
exactly and wasn't sure if their are some known issues here.
Brian
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:Re: (usr-tc) filtering service (fwd) From: Brian <signal@shreve.net> Date: 1998-11-13 10:37:43
On Thu, 12 Nov 1998, Jeff Mcadams wrote:
> Thus spake Angela A. Burford
> >the portmaster through choicenet, so that it will be used in that
>
> >I'd like to make this work with my usr's too (i have chassis with
> >netservers, quad modems nmc and dual pri cards and other with hiperDSP's
> >hiperARC and nmc cards). I understand that I can
> >pass the names of the filters to be used at the time the user connects,
> >but is it possible to load the actual filter that is stored in my
> >radius server into the hiperARC / netserver so that it will applied to the
> >connection? If so, how can that be done?
>
> Can't reasonably be done on NETServers...the HARC's have the ability to
> handle "dynamic filters" given a RADIUS server that supports it. Not
> having HARC's to really play with extensively, I don't know how to do
> this. :)
Isn't "hint assigned" on the netserver enough for it to support dynamic
filters? it should be.........
> --
> Jeff McAdams Email: jeffm@iglou.com
> Head Network Administrator Voice: (502) 966-3848
> IgLou Internet Services (800) 436-4456
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:Re: (usr-tc) DUN1.3 and MPP From: Brian <signal@shreve.net> Date: 1998-11-13 10:39:00
On Fri, 13 Nov 1998, Jeff Mcadams wrote:
> Thus spake Brian
> >I have a user trying to connect using DUN1.3 and Win98. He is trying to
> >do MPP using 2 USR 56k external sportsters.
>
> <nit>Should be just MP, MPP is an Ascend proprietary thing</nit>
>
> >Should this work? Both modems connect but then the 2nd one disconnects.
> >I have him setup for dual channel, so i am not sure what is going on
> >exactly and wasn't sure if their are some known issues here.
>
> Check to see if the Windows machine is dialing both modems at the same
> time. I know NT does this, and it can give the NAS fits. I believe
> turning off LCP extensions in NT turns off this behavior, but am not
> sure.
The dialilng is about 30 seconds spaced. They authenticate, but right
after that the 2nd channel drops.
> --
> Jeff McAdams Email: jeffm@iglou.com
> Head Network Administrator Voice: (502) 966-3848
> IgLou Internet Services (800) 436-4456
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:RE: (usr-tc) Solution for idle time limits and dual logins From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-13 10:41:18
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian K McIntire
|Sent: Friday, November 13, 1998 11:28 AM
|To: usr-tc@lists.xmission.com
|Subject: RE: (usr-tc) Solution for idle time limits and dual logins
|
|
|I have seen the default user on the HiPer ARC settings not apply to RADIUS
|users. Maybe it was broke in one of the versions of code I had then. I'll
|take another look
|
It could have been.. But I havent seen it.. The only time default user settings
would not apply is if there was a conflicting RADIUS attribute. Then RADIUS wins
the tie. For example. Default user has idle time set to 600 and the Radius
attribute "Idle-Timeout=1200" is in the access accept. The idle timeout will be
1200 for that user.
-M
Subject:RE: (usr-tc) Solution for idle time limits and dual logins From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-13 11:28:25
I have seen the default user on the HiPer ARC settings not apply to RADIUS
users. Maybe it was broke in one of the versions of code I had then. I'll
take another look
========================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= Your Remote Access and Security Experts! =
= =
= Our experienced technicians can design and install your high speed =
= network for you. From Operating Systems and servers to routers and =
= firewalls nobody does it better than CommNetPlus. Our Technical =
= Support staff is available to you 24 hours a day to meet your needs.=
= Offering remote management and monitoring packages for dedicated =
= clients. (Specializing in ISP's). "Let us do the work for you!" =
= =
= Firm partnerships est. with current and emerging leaders of today's =
= technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
= Serving North America and parts of Canada. Reselling to the world. =
========================================================================
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Mike Wronski
Sent: Friday, November 13, 1998 10:01 AM
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian K McIntire
|Sent: Friday, November 13, 1998 12:21 AM
|To: usr-tc@lists.xmission.com
|Subject: RE: (usr-tc) Solution for idle time limits and dual logins
|
|The NetServer does not support session timeout without RADIUS
|for idletimeout set all idletime 15 (Where 15 is in minutes)
|
|For the HiPer ARC just apply the values you want to the default user. All
|users created after that will have the correct settings.
|
|This is very sloppy in my opinion. Better to use the RADIUS.
Sloppy? Default user applies to all dial-in users. Not just locally created
ones.
This is not sloppy. It's quite convenient. If you set the same Idle timeout
or
session timeout for all/most of your users. Just set that on the default
user and
remove the attributes from your default user in the RADIUS users file. If a
particular user or set of users gets different values then include them in
RADIUS. RADIUS attributes always take precedence over default user
settings.
-M
|=======================================================================
|= Brian K. McIntire bmcintire@commnet.com =
|= Systems Engineer III 317-558-5050 x113 =
|= CommNetPlus, Indianapolis, IN http://www.commnet.com =
|= =
|= Experts in Remote Access and all areas of NetWork Design, =
|= Security, and Implementation. Offering both installation and =
|= support, along with remote management and monitoring packages for =
|= dedicated clients (specializing in ISP's). =
|= Firm partnerships established with all the key players. =
|= Sales 800-845-2981 x117 (Aggressive Pricing strategies) =
|= =
|= Serving North America and parts of Canada. Reselling to the world. =
|=======================================================================
|
Isn't this signature a little long?? :) (4/5 lines max in netiquette)
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:RE: (usr-tc) DUN1.3 and MPP From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-13 11:31:01
Yes you can bundle channels together from 2 external sportsters. I do not
know if it works with DUN 1.3. Haven't tried it
========================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= Your Remote Access and Security Experts! =
= =
= Our experienced technicians can design and install your high speed =
= network for you. From Operating Systems and servers to routers and =
= firewalls nobody does it better than CommNetPlus. Our Technical =
= Support staff is available to you 24 hours a day to meet your needs.=
= Offering remote management and monitoring packages for dedicated =
= clients. (Specializing in ISP's). "Let us do the work for you!" =
= =
= Firm partnerships est. with current and emerging leaders of today's =
= technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
= Serving North America and parts of Canada. Reselling to the world. =
========================================================================
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian
Sent: Friday, November 13, 1998 10:28 AM
I have a user trying to connect using DUN1.3 and Win98. He is trying to
do MPP using 2 USR 56k external sportsters.
Should this work? Both modems connect but then the 2nd one disconnects.
I have him setup for dual channel, so i am not sure what is going on
exactly and wasn't sure if their are some known issues here.
Brian
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:Re: (usr-tc) DUN1.3 and MPP From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-13 11:32:54
Thus spake Brian
>I have a user trying to connect using DUN1.3 and Win98. He is trying to
>do MPP using 2 USR 56k external sportsters.
<nit>Should be just MP, MPP is an Ascend proprietary thing</nit>
>Should this work? Both modems connect but then the 2nd one disconnects.
>I have him setup for dual channel, so i am not sure what is going on
>exactly and wasn't sure if their are some known issues here.
Check to see if the Windows machine is dialing both modems at the same
time. I know NT does this, and it can give the NAS fits. I believe
turning off LCP extensions in NT turns off this behavior, but am not
sure.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:(usr-tc) Total Controll for Sale From: Tony Loosle <tony@tcsourceone.com> Date: 1998-11-13 13:22:23
For Sale:
USR Total Control
6 Quad Digital Modems, 33.6 and isdn. will do x2 with upgrade.
Dual PRI card
Net management card
Net server card
New, never in service.
Make an offer on parts or the entire box!!
Tony
Subject:Re: (usr-tc) DUN1.3 and MPP From: Jason W <jwatkins@iland.net> Date: 1998-11-13 13:46:18
Make sure that you have for the port type 'diail in'
only. If you try to use dial in & dial out, or two way,
it will not work properly. Also you need to make sure
and set up MPIP so bundles can be established
across multiple chassis'.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Jason Watkins jwatkins@iland.net
I-Land NOC Tech http://www.iland.net
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
-----Original Message-----
:Do you have MPIP configured properly. Otherwise it won't work unless you
:hit the same rack
:
:========================================================================
:= Brian K. McIntire bmcintire@commnet.com =
:= Systems Engineer III 317-558-5050 x113 =
:= CommNetPlus, Indianapolis, IN http://www.commnet.com =
:= Your Remote Access and Security Experts! =
:= =
:= Our experienced technicians can design and install your high speed =
:= network for you. From Operating Systems and servers to routers and =
:= firewalls nobody does it better than CommNetPlus. Our Technical =
:= Support staff is available to you 24 hours a day to meet your needs.=
:= Offering remote management and monitoring packages for dedicated =
:= clients. (Specializing in ISP's). "Let us do the work for you!" =
:= =
:= Firm partnerships est. with current and emerging leaders of today's =
:= technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
:= Serving North America and parts of Canada. Reselling to the world. =
:========================================================================
:
:
:-----Original Message-----
:From: owner-usr-tc@lists.xmission.com
:[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Terry Kennedy
:Sent: Friday, November 13, 1998 12:08 PM
:To: usr-tc@lists.xmission.com
:Subject: RE: (usr-tc) DUN1.3 and MPP
:
:
:Ours doesn't work unless it hits the same rack.
:
:> -----Original Message-----
:> From: owner-usr-tc@lists.xmission.com
:> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian
:> Sent: Friday, November 13, 1998 8:28 AM
:> To: USRobotics TC Mailing List
:> Subject: (usr-tc) DUN1.3 and MPP
:>
:>
:> I have a user trying to connect using DUN1.3 and Win98. He is trying to
:> do MPP using 2 USR 56k external sportsters.
:>
:> Should this work? Both modems connect but then the 2nd one disconnects.
:> I have him setup for dual channel, so i am not sure what is going on
:> exactly and wasn't sure if their are some known issues here.
:>
:> Brian
:>
:>
:> -------------------------------------------------------------------------
-
:> Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service
:> Provider
:> Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
:> signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
:> (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
:>
:>
:> -
:> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
:> with "unsubscribe usr-tc" in the body of the message.
:> For information on digests or retrieving files and old messages send
:> "help" to the same address. Do not use quotes in your message.
:>
:
:
:-
: To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
: with "unsubscribe usr-tc" in the body of the message.
: For information on digests or retrieving files and old messages send
: "help" to the same address. Do not use quotes in your message.
:
:
:-
: To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
: with "unsubscribe usr-tc" in the body of the message.
: For information on digests or retrieving files and old messages send
: "help" to the same address. Do not use quotes in your message.
:
Subject:RE: (usr-tc) DUN1.3 and MPP From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-13 14:04:32
Do you have MPIP configured properly. Otherwise it won't work unless you
hit the same rack
========================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= Your Remote Access and Security Experts! =
= =
= Our experienced technicians can design and install your high speed =
= network for you. From Operating Systems and servers to routers and =
= firewalls nobody does it better than CommNetPlus. Our Technical =
= Support staff is available to you 24 hours a day to meet your needs.=
= Offering remote management and monitoring packages for dedicated =
= clients. (Specializing in ISP's). "Let us do the work for you!" =
= =
= Firm partnerships est. with current and emerging leaders of today's =
= technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
= Serving North America and parts of Canada. Reselling to the world. =
========================================================================
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Terry Kennedy
Sent: Friday, November 13, 1998 12:08 PM
Ours doesn't work unless it hits the same rack.
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian
> Sent: Friday, November 13, 1998 8:28 AM
> To: USRobotics TC Mailing List
> Subject: (usr-tc) DUN1.3 and MPP
>
>
> I have a user trying to connect using DUN1.3 and Win98. He is trying to
> do MPP using 2 USR 56k external sportsters.
>
> Should this work? Both modems connect but then the 2nd one disconnects.
> I have him setup for dual channel, so i am not sure what is going on
> exactly and wasn't sure if their are some known issues here.
>
> Brian
>
>
> --------------------------------------------------------------------------
> Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service
> Provider
> Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:(usr-tc) front face plates, model numbers From: Laszlo Vecsey <laszlo@internexus.net> Date: 1998-11-13 18:20:21
Does anyone have the model numbers for the front face plates on the usr-tc
chasis, both one and two wide in particular -- does the most recent
netserver to hyper upgrade bundle include one?
(minor point, but it'll keep the dust out.. the most recent bundle in
question includes just one hyperdsp, leaving one slot free)
-L
Subject:RE: (usr-tc) DUN1.3 and MPP From: Brian <signal@shreve.net> Date: 1998-11-13 20:26:16
On Fri, 13 Nov 1998, Terry Kennedy wrote:
> Ours doesn't work unless it hits the same rack.
so mpip of win98 mp doesnt work?
>
> > -----Original Message-----
> > From: owner-usr-tc@lists.xmission.com
> > [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian
> > Sent: Friday, November 13, 1998 8:28 AM
> > To: USRobotics TC Mailing List
> > Subject: (usr-tc) DUN1.3 and MPP
> >
> >
> > I have a user trying to connect using DUN1.3 and Win98. He is trying to
> > do MPP using 2 USR 56k external sportsters.
> >
> > Should this work? Both modems connect but then the 2nd one disconnects.
> > I have him setup for dual channel, so i am not sure what is going on
> > exactly and wasn't sure if their are some known issues here.
> >
> > Brian
> >
> >
> > --------------------------------------------------------------------------
> > Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service
> > Provider
> > Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> > signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> > (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:Re: (usr-tc) DUN1.3 and MPP From: Brian <signal@shreve.net> Date: 1998-11-13 20:27:29
On Fri, 13 Nov 1998, Jason W wrote:
> Make sure that you have for the port type 'diail in'
> only. If you try to use dial in & dial out, or two way,
> it will not work properly. Also you need to make sure
> and set up MPIP so bundles can be established
> across multiple chassis'.
mpip works fine here, with isdn et al. But its just this w98 user having
problems, and someone posted to the list that their is in fact known 3com
issues with w98 mp.
>
>
> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> Jason Watkins jwatkins@iland.net
> I-Land NOC Tech http://www.iland.net
> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> -----Original Message-----
> From: Brian K McIntire <bmcintire@commnet.com>
> To: usr-tc@lists.xmission.com <usr-tc@lists.xmission.com>
> Date: Friday, November 13, 1998 1:09 PM
> Subject: RE: (usr-tc) DUN1.3 and MPP
>
>
> :Do you have MPIP configured properly. Otherwise it won't work unless you
> :hit the same rack
> :
> :========================================================================
> := Brian K. McIntire bmcintire@commnet.com =
> := Systems Engineer III 317-558-5050 x113 =
> := CommNetPlus, Indianapolis, IN http://www.commnet.com =
> := Your Remote Access and Security Experts! =
> := =
> := Our experienced technicians can design and install your high speed =
> := network for you. From Operating Systems and servers to routers and =
> := firewalls nobody does it better than CommNetPlus. Our Technical =
> := Support staff is available to you 24 hours a day to meet your needs.=
> := Offering remote management and monitoring packages for dedicated =
> := clients. (Specializing in ISP's). "Let us do the work for you!" =
> := =
> := Firm partnerships est. with current and emerging leaders of today's =
> := technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
> := Serving North America and parts of Canada. Reselling to the world. =
> :========================================================================
> :
> :
> :-----Original Message-----
> :From: owner-usr-tc@lists.xmission.com
> :[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Terry Kennedy
> :Sent: Friday, November 13, 1998 12:08 PM
> :To: usr-tc@lists.xmission.com
> :Subject: RE: (usr-tc) DUN1.3 and MPP
> :
> :
> :Ours doesn't work unless it hits the same rack.
> :
> :> -----Original Message-----
> :> From: owner-usr-tc@lists.xmission.com
> :> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian
> :> Sent: Friday, November 13, 1998 8:28 AM
> :> To: USRobotics TC Mailing List
> :> Subject: (usr-tc) DUN1.3 and MPP
> :>
> :>
> :> I have a user trying to connect using DUN1.3 and Win98. He is trying to
> :> do MPP using 2 USR 56k external sportsters.
> :>
> :> Should this work? Both modems connect but then the 2nd one disconnects.
> :> I have him setup for dual channel, so i am not sure what is going on
> :> exactly and wasn't sure if their are some known issues here.
> :>
> :> Brian
> :>
> :>
> :> -------------------------------------------------------------------------
> -
> :> Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service
> :> Provider
> :> Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> :> signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> :> (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
> :>
> :>
> :> -
> :> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> :> with "unsubscribe usr-tc" in the body of the message.
> :> For information on digests or retrieving files and old messages send
> :> "help" to the same address. Do not use quotes in your message.
> :>
> :
> :
> :-
> : To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> : with "unsubscribe usr-tc" in the body of the message.
> : For information on digests or retrieving files and old messages send
> : "help" to the same address. Do not use quotes in your message.
> :
> :
> :-
> : To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> : with "unsubscribe usr-tc" in the body of the message.
> : For information on digests or retrieving files and old messages send
> : "help" to the same address. Do not use quotes in your message.
> :
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:RE: (usr-tc) Total Controll for Sale From: Jim Logan <jim@top.net> Date: 1998-11-13 21:24:25
For which Piece :)
(Ok, so it's Friday the 13th, I guess)
At 10:17 PM 11/13/98 -0500, you wrote:
>Will offer $2K
>John
>
>On Friday, November 13, 1998 3:22 PM, Tony Loosle
[SMTP:tony@tcsourceone.com] wrote:
>> For Sale:
>> USR Total Control
>>
>> 6 Quad Digital Modems, 33.6 and isdn. will do x2 with upgrade.
>> Dual PRI card
>> Net management card
>> Net server card
>>
>> New, never in service.
>>
>> Make an offer on parts or the entire box!!
>>
>> Tony
>>
>>
>>
>>
>>
>>
>> -
>> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
>> with "unsubscribe usr-tc" in the body of the message.
>> For information on digests or retrieving files and old messages send
>> "help" to the same address. Do not use quotes in your message.
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
***** Top Net InterNet Services *****
Omaha, Nebraska Husker Heaven
www.top.net (402) 291-1542
Visit Our BBS at: www.hawgwild.com
Subject:RE: (usr-tc) Total Controll for Sale From: John Verreault <verreaul@aei.ca> Date: 1998-11-13 22:17:04
Will offer $2K
John
On Friday, November 13, 1998 3:22 PM, Tony Loosle [SMTP:tony@tcsourceone.com] wrote:
> For Sale:
> USR Total Control
>
> 6 Quad Digital Modems, 33.6 and isdn. will do x2 with upgrade.
> Dual PRI card
> Net management card
> Net server card
>
> New, never in service.
>
> Make an offer on parts or the entire box!!
>
> Tony
>
>
>
>
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
have a netserver advertising 1.2.3.0/32 into RIP. strange,
strange.
1.2.3.0 /32 1.2.3.3 HD 2 net0 0
(1.2.3.0 is the network, 1.2.3.4 is ether0 of the netserver)
when my cisco sees it it quits speaking rip with it. can't find
out why it's there (no deubg tools afaik), no netmasks or static
routes anywhere except for the default. really strange stuff.
anyone ever seen this?
thanks!
|o|----------------------------------------------------------------------|o|
|o| bryan s. blank (203)-351-1178 voice |o|
|o| senior systems analyst (203)-351-1186 fax |o|
|o| discovernet, incorporated (203)-979-5126 emerg |o|
On Sat, 14 Nov 1998, bryan s. blank wrote:
> have a netserver advertising 1.2.3.0/32 into RIP. strange,
> strange.
>
> 1.2.3.0 /32 1.2.3.3 HD 2 net0 0
>
This is a host dynamic route learnt by the Netserver from the ethernet.
Either you have a workstation or a router doing this or someone in your
network is giving this route. Check your netmask on your netserver and
on the other devices on your network.
krish
> (1.2.3.0 is the network, 1.2.3.4 is ether0 of the netserver)
>
> when my cisco sees it it quits speaking rip with it. can't find
> out why it's there (no deubg tools afaik), no netmasks or static
> routes anywhere except for the default. really strange stuff.
>
> anyone ever seen this?
>
> thanks!
>
> |o|----------------------------------------------------------------------|o|
> |o| bryan s. blank (203)-351-1178 voice |o|
> |o| senior systems analyst (203)-351-1186 fax |o|
> |o| discovernet, incorporated (203)-979-5126 emerg |o|
>
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) rip strangeness From: Brian <signal@shreve.net> Date: 1998-11-14 18:30:35
On Sat, 14 Nov 1998, bryan s. blank wrote:
> have a netserver advertising 1.2.3.0/32 into RIP. strange,
> strange.
>
> 1.2.3.0 /32 1.2.3.3 HD 2 net0 0
>
> (1.2.3.0 is the network, 1.2.3.4 is ether0 of the netserver)
>
> when my cisco sees it it quits speaking rip with it. can't find
> out why it's there (no deubg tools afaik), no netmasks or static
> routes anywhere except for the default. really strange stuff.
the cisco can show you where it is coming from and you can run a debug rip
on it.
>
> anyone ever seen this?
>
> thanks!
>
> |o|----------------------------------------------------------------------|o|
> |o| bryan s. blank (203)-351-1178 voice |o|
> |o| senior systems analyst (203)-351-1186 fax |o|
> |o| discovernet, incorporated (203)-979-5126 emerg |o|
>
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
|o| This is a host dynamic route learnt by the Netserver from the ethernet.
|o| Either you have a workstation or a router doing this or someone in your
|o| network is giving this route. Check your netmask on your netserver and
|o| on the other devices on your network.
extremely strange.
3 netservers going in to one switch, interface ethernet 1/1 on a
cisco 7206. netserver is 1.2.3.3
router rip
version 2
timers basic 30 90 5 120
passive-interface ... (everything but ethernet 1/1)
....
network 1.2.3.0
no auto-summary
the switch it's on doesn't know how to speak rip.
crisco says:
6d03h: RIP: received v2 update from 1.2.3.3 on Ethernet1/1
6d03h: 1.2.4.192/27 -> 0.0.0.0 in 1 hops
6d03h: 1.2.3.0/32 (illegal address)
6d03h: RIP: Update contains 1 routes
nobody else is broadcasting 1.2.3.0/32 ... i'm quite stumped.
thanks so much for your help, as always.
|o|----------------------------------------------------------------------|o|
|o| bryan s. blank (203)-351-1178 voice |o|
|o| senior systems analyst (203)-351-1186 fax |o|
|o| discovernet, incorporated (203)-979-5126 emerg |o|
Subject:Re: (usr-tc) Total Controll for Sale From: Dale Levendoski <dale-lev@mwt.net> Date: 1998-11-14 21:32:39
--Cyberdog-MixedBoundary-0023658F
X-Fontfamily: Courier
X-Fontsize: 12
Content-Type: text/enriched; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
<FIXED>Tony,
Why are you selling this? Is it sold yet?
Dale-Lev@MWT.Net
On Fri, Nov 13, 1998 2:22 PM, </FIXED>
--Cyberdog-MixedBoundary-0023658F
Content-Type: application/X-url
Content-Transfer-Encoding: base64
Content-Description: Tony Loosle
bWFpbHRvOnRvbnlAdGNzb3VyY2VvbmUuY29t
--Cyberdog-MixedBoundary-0023658F
X-Fontfamily: Courier
X-Fontsize: 12
Content-Type: text/enriched; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
<FIXED> wrote:</FIXED><FIXED>
</FIXED><FIXED>For Sale:
USR Total Control
6 Quad Digital Modems, 33.6 and isdn. will do x2 with upgrade.
Dual PRI card
Net management card
Net server card
New, never in service.
Make an offer on parts or the entire box!!</FIXED>
--Cyberdog-MixedBoundary-0023658F--
Subject:(usr-tc) Confirmation From: Terry Kennedy <terry@olypen.com> Date: 1998-11-16 11:52:56
In the process of trying to turn of mlppp ( god I hope that's what
I'm talking about - the ability for people to call into our system
and bond channels ) off, I've got what I think I need. If someone could
confirm this, I would appreciate it.
On the netserver it was "set mp off"
On the ARC is it " set network user default ppp max_channels 1 " ?
Or am I way off here..
Terry Kennedy
I have the following equipment available for sale.
7- Netserver 16
1- Netserver 16I
1- Netserver 8I
2- MP8
4- MP16
All sold guaranteed working.
Steve Rivera
WRCA, Inc. http://www.wrca.net
4 Kinney Rd mailto:sales@wrca.net
Manalapan, NJ 07726 Online Orders Accepted
p: 732-462-0062 f: 732-462-8455
"Everything DataCom, Nothing but Net"
Hello,
Who can give me detail meaning of the disconnect reason code 62--received
disconnect command from gateway?
An explanation that we already know is that the dialin user was
disconnected by the netserver/HiperArc card for the incorrect password or
incompatible network protocol.
But we found several recorder indicating the disconnect happened after
the call established for several minutes.
Does it mean that the netserver/hiper Arc card raise the disconnect for
unknow reason?
Allen
Subject:(usr-tc) accounting problem through NMC From: allen_lai@3com.com Date: 1998-11-16 13:44:22
Hello,all
We have installed a USR S/A server in a NT workstation(Intel pro233) to
collect the accounting informaiton from NMC.
We created about 60 NMC/chassis---raduis accounting client in one S/A
server.
At first, everything works OK.We got accounting information from all
created clients.
But gradually, some clients will disappered.We can't got any accounting
information of these disappered client now from S/A now.
When ping the NMC, it's ok.We can also managned these chassis through TCM.
Any hints?
Allen
Subject:Re: (usr-tc) Confirmation From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-16 14:59:57
Thus spake Terry Kennedy
>In the process of trying to turn of mlppp ( god I hope that's what
>I'm talking about - the ability for people to call into our system
>and bond channels ) off, I've got what I think I need. If someone could
>confirm this, I would appreciate it.
MP (commonly called MLPPP...multi-link PPP being the full name and
meaning the ability to bundle multiple channels together to effectively
aggregate the bandwidth of all the channels) is what you're refer'ing
to.
>On the netserver it was "set mp off"
This will actually turn off MP option negotiation...this might
break things.
>On the ARC is it " set network user default ppp max_channels 1 " ?
This doesn't turn off MP option negotiation, but limits the client side
to only connecting one channel in a bundle. This is probably more along
the lines of what you want, and the same thing can be done on the
NETServer, either through RADIUS or through the local users table I
believe.
The RADIUS attribute to limit this to one channel is:
Port-Limit = 1
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
At 02:59 PM 11/16/98 -0500, Jeff McAdams wrote:
>>On the ARC is it " set network user default ppp max_channels 1 " ?
>
>This doesn't turn off MP option negotiation, but limits the client side
>to only connecting one channel in a bundle. This is probably more along
>the lines of what you want, and the same thing can be done on the
>NETServer, either through RADIUS or through the local users table I
>believe.
Will this work to prohibit simultaneous logins, or just multiple logins
using 'bonded' channels? I'd like to have some way of preventing multiple
logins since S&A Server can't be counted on to do so properly.
Thanks,
Kirk
Kirk Mitchell-General Manager sysadmin@keyconn.net
Keystone Connect http://www.keyconn.net
Altoona, PA 814-941-5000 We Unlock the World
Subject:Re: (usr-tc) Confirmation From: Mike Andrews <mandrews@termfrost.org> Date: 1998-11-16 15:21:32
On Mon, 16 Nov 1998, Jeff Mcadams wrote:
> The RADIUS attribute to limit this to one channel is:
> Port-Limit = 1
This seems to be broken on current ARC releases up to and including at
least 4.1.78; you have to also use "Max_Channels = 1" (which is a VSA) to
do the right thing there. On the NETserver, it's fine. Go back through
the archives a month or two and you can see me whining away about this :)
We actually have both Port-Limit AND Max_Channels in the Radius users
file. That seems to take care of things.
Mike Andrews (MA12) icq 6602506 -------------- mandrews@termfrost.org
VP 'n' Systems/Network Administrator --------------- mandrews@dcr.net
Digital Crescent, Frankfort, KY ----------- http://www.termfrost.org/
# view;touch;unzip;finger;mount;mv;mv;mv;yes;mv;yes;mv;yes;umount;sleep
Subject:RE: (usr-tc) accounting problem through NMC From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-16 16:21:01
Hello,all
We have installed a USR S/A server in a NT workstation(Intel pro233) to
collect the accounting informaiton from NMC.
We created about 60 NMC/chassis---raduis accounting client in one S/A
server.
At first, everything works OK.We got accounting information from all
created clients.
But gradually, some clients will disappered.
*********
What do you mean by this?
Have you used TCM to open the NMC Logging Group info and make sure you see
the Primary Logging Server listed?
We can't got any accounting
information of these disappered client now from S/A now.
When ping the NMC, it's ok.We can also managned these chassis through TCM.
Any hints?
Allen
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:(usr-tc) PRI settings blown away by NMC From: Chris Hanes <chris@internetcreations.com> Date: 1998-11-16 16:36:58
Hey all,
I'm trying to map modems to slots under chassis slot device
configuration on a dual pri nac. The settings take so long
as my NMC is not plugged in. Once plugged in the NMC obscures these
settings and prevents me from reentering the settings on the pri card
(i.e. the settings won't take when the NMC is plugged in). I've been
through a couple pri cards and figure I must be doing something stupid.
I'm running tcs 3.3 (3.02 on the PRI nac and 5.5.5 on the NMC)
Any help would be IMMENSELY appreciated.
Thanks,
Chris Hanes
Internet Creations
Subject:RE: (usr-tc) PRI settings blown away by NMC From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-16 16:53:54
Sounds like you have "Auto config upon card initialization" enabled on your
NMC.
1. Load TCM
2. Bring up your chassis
3. Click on the NMC
4. Click on Configure/program settings/configuration group
5. Make sure auto config is disabled.
6. If it was enabled set it and click ok.
7. go to configure/action commands and save chassis to nvram (don't really
need this one but do it anyway)
8. Save UI to eeprom
9. Reboot
10. If the setting was disabled then restore your NMC (not chassis) from
default by clicking on the NMC and going to configure/action commands and
choosing restore NMC from default
11. then perform steps 7 & 8 and then 9 from above
This should solve your problem. Remember you will need to reconfigure the
Logging group on the NMC if you have restored it from default and wish to do
accounting
========================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= Your Remote Access and Security Experts! =
= =
= Our experienced technicians can design and install your high speed =
= network for you. From Operating Systems and servers to routers and =
= firewalls nobody does it better than CommNetPlus. Our Technical =
= Support staff is available to you 24 hours a day to meet your needs.=
= Offering remote management and monitoring packages for dedicated =
= clients. (Specializing in ISP's). "Let us do the work for you!" =
= =
= Firm partnerships est. with current and emerging leaders of today's =
= technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
= Serving North America and parts of Canada. Reselling to the world. =
========================================================================
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Chris Hanes
Sent: Monday, November 16, 1998 3:37 PM
Hey all,
I'm trying to map modems to slots under chassis slot device
configuration on a dual pri nac. The settings take so long
as my NMC is not plugged in. Once plugged in the NMC obscures these
settings and prevents me from reentering the settings on the pri card
(i.e. the settings won't take when the NMC is plugged in). I've been
through a couple pri cards and figure I must be doing something stupid.
I'm running tcs 3.3 (3.02 on the PRI nac and 5.5.5 on the NMC)
Any help would be IMMENSELY appreciated.
Thanks,
Chris Hanes
Internet Creations
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:Re: (usr-tc) PRI settings blown away by NMC From: Chris Hanes <chris@internetcreations.com> Date: 1998-11-16 18:05:16
Hey Brian,
Thanks for the reponse. Auto config was disabled. I've tried restoring
NMC from factory default several times to no avail. When I boot the
box,
I've been watching the pri cards chassis slot settings. The pri card
seems
to boot quickly and I'm able to see the settings briefly before they are
wiped out, leaving the pri card seeing only itself in slot 1 and the
netserver
card in slot 16. Any other ideas?
Thanks,
Chris Hanes
Brian K McIntire wrote:
>
> Sounds like you have "Auto config upon card initialization" enabled on your
> NMC.
>
> 1. Load TCM
> 2. Bring up your chassis
> 3. Click on the NMC
> 4. Click on Configure/program settings/configuration group
> 5. Make sure auto config is disabled.
> 6. If it was enabled set it and click ok.
> 7. go to configure/action commands and save chassis to nvram (don't really
> need this one but do it anyway)
> 8. Save UI to eeprom
> 9. Reboot
> 10. If the setting was disabled then restore your NMC (not chassis) from
> default by clicking on the NMC and going to configure/action commands and
> choosing restore NMC from default
> 11. then perform steps 7 & 8 and then 9 from above
> This should solve your problem. Remember you will need to reconfigure the
> Logging group on the NMC if you have restored it from default and wish to do
> accounting
>
> ========================================================================
> = Brian K. McIntire bmcintire@commnet.com =
> = Systems Engineer III 317-558-5050 x113 =
> = CommNetPlus, Indianapolis, IN http://www.commnet.com =
> = Your Remote Access and Security Experts! =
> = =
> = Our experienced technicians can design and install your high speed =
> = network for you. From Operating Systems and servers to routers and =
> = firewalls nobody does it better than CommNetPlus. Our Technical =
> = Support staff is available to you 24 hours a day to meet your needs.=
> = Offering remote management and monitoring packages for dedicated =
> = clients. (Specializing in ISP's). "Let us do the work for you!" =
> = =
> = Firm partnerships est. with current and emerging leaders of today's =
> = technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
> = Serving North America and parts of Canada. Reselling to the world. =
> ========================================================================
Subject:Re: (usr-tc) PRI settings blown away by NMC From: Jamie Orzechowski <mhz@recorder.ca> Date: 1998-11-16 18:17:25
have you tried the setting the config via the CONSOLE port?? ... I had the
same problem that I could get to the config via TCM but settings would never
"take" ... I did it via console (restore from default, config, save to
nvram, reset) ... and worked fine ...
-----Original Message-----
>Hey Brian,
>
>Thanks for the reponse. Auto config was disabled. I've tried restoring
>NMC from factory default several times to no avail. When I boot the
>box,
>I've been watching the pri cards chassis slot settings. The pri card
>seems
>to boot quickly and I'm able to see the settings briefly before they are
>wiped out, leaving the pri card seeing only itself in slot 1 and the
>netserver
>card in slot 16. Any other ideas?
>
>Thanks,
>Chris Hanes
>
>Brian K McIntire wrote:
>>
>> Sounds like you have "Auto config upon card initialization" enabled on
your
>> NMC.
>>
>> 1. Load TCM
>> 2. Bring up your chassis
>> 3. Click on the NMC
>> 4. Click on Configure/program settings/configuration group
>> 5. Make sure auto config is disabled.
>> 6. If it was enabled set it and click ok.
>> 7. go to configure/action commands and save chassis to nvram (don't
really
>> need this one but do it anyway)
>> 8. Save UI to eeprom
>> 9. Reboot
>> 10. If the setting was disabled then restore your NMC (not chassis) from
>> default by clicking on the NMC and going to configure/action commands and
>> choosing restore NMC from default
>> 11. then perform steps 7 & 8 and then 9 from above
>> This should solve your problem. Remember you will need to reconfigure
the
>> Logging group on the NMC if you have restored it from default and wish to
do
>> accounting
>>
>> ========================================================================
>> = Brian K. McIntire bmcintire@commnet.com =
>> = Systems Engineer III 317-558-5050 x113 =
>> = CommNetPlus, Indianapolis, IN http://www.commnet.com =
>> = Your Remote Access and Security Experts! =
>> = =
>> = Our experienced technicians can design and install your high speed =
>> = network for you. From Operating Systems and servers to routers and =
>> = firewalls nobody does it better than CommNetPlus. Our Technical =
>> = Support staff is available to you 24 hours a day to meet your needs.=
>> = Offering remote management and monitoring packages for dedicated =
>> = clients. (Specializing in ISP's). "Let us do the work for you!" =
>> = =
>> = Firm partnerships est. with current and emerging leaders of today's =
>> = technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
>> = Serving North America and parts of Canada. Reselling to the world. =
>> ========================================================================
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) Confirmation From: K Mitchell <mitch@keyconn.net> Date: 1998-11-16 18:26:14
At 02:59 PM 11/16/98 -0500, Jeff McAdams wrote:
>>On the ARC is it " set network user default ppp max_channels 1 " ?
>
>This doesn't turn off MP option negotiation, but limits the client side
>to only connecting one channel in a bundle. This is probably more along
>the lines of what you want, and the same thing can be done on the
>NETServer, either through RADIUS or through the local users table I
>believe.
Will this work to prohibit simultaneous logins, or just multiple logins
using 'bonded' channels? I'd like to have some way of preventing multiple
logins since S&A Server can't be counted on to do so properly.
Thanks,
Kirk
PS If this post comes again, I apologize, I had sent it with a different
return address and wasn't sure if it got rejected or not.
Kirk Mitchell-General Manager sysadmin@keyconn.net
Keystone Connect http://www.keyconn.net
Altoona, PA 814-941-5000 We Unlock the World
Subject:Re: (usr-tc) Confirmation From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-16 19:00:20
Thus spake K Mitchell
>At 02:59 PM 11/16/98 -0500, Jeff McAdams wrote:
>>This doesn't turn off MP option negotiation, but limits the client side
>>to only connecting one channel in a bundle. This is probably more along
>>the lines of what you want, and the same thing can be done on the
>>NETServer, either through RADIUS or through the local users table I
>>believe.
>Will this work to prohibit simultaneous logins, or just multiple logins
>using 'bonded' channels?
Just multiple MP channels in a bundle. This will not prevent multiple
logins. There is no clean way to prevent multiple logins cross-chassis
at this point. Perhaps some NAS vendor would like to come up with a
login-limiting protocol that will really make their NAS boxen act as one
unit? Noone has done this yet to my knowledge.
>I'd like to have some way of preventing multiple
>logins since S&A Server can't be counted on to do so properly.
There is no way to do this easily.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:(usr-tc) NMC rebooting From: Brian Elfert <brian@citilink.com> Date: 1998-11-16 19:16:14
I picked up a used 1866 bundle.
The NMC card keeps rebooting. It gets far enough to display the menu on
the console, but it reboots before anything can be entered.
It appears to be running 5.4.95 on the NMC.
Any ideas to fix this?
Brian
This could be because
1. you have programmed a 0.0.0.0 as ip address of the NMC
2. You have a bad flash
Try this.
Download 4.3.9 nmc code ( or any 4.x code ) use a program called pcsdl (
comes with the code ) and download that code to the NMC and then program
the proper ip address save it and then upgrade it to 5.x code
If this does not help then your flash may be corrupted.
krish
\ T.S.V. Krishnan \
\ Network System Engineer \ ( : - : )
\ 3Com ............ \
----------------------------------------------/
tkrishna@bubba.ae.usr.com
----------------------------/ http://interproc.ae.usr.com ----/
The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
Any Sufficiently advanced bug is indistinguishable for a feature.
- Rick Kulawiec
On Mon, 16 Nov 1998, Brian Elfert wrote:
> I picked up a used 1866 bundle.
>
> The NMC card keeps rebooting. It gets far enough to display the menu on
> the console, but it reboots before anything can be entered.
>
> It appears to be running 5.4.95 on the NMC.
>
> Any ideas to fix this?
>
> Brian
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) PRI or not? From: Marcelo Souza <mpsouza@centroin.com.br> Date: 1998-11-16 20:34:32
What must I do to change my dual E1/CAS cards to use with PRI
(ISDN)?
And the DSPs ?
Should I make some change in the modems also?
Can I have in the same chassis some DSP with PRI and others not?
- Marcelo
Subject:Re: (usr-tc) NMC rebooting From: Alan D. Criado <acriado@elink.net> Date: 1998-11-16 20:41:23
Hi Brian.
The same thing happened to me on my NMC once, when I was first setting it
up. I had placed 0.0.0.0 for one of the IP addresses and it causes the NMC
to perpetually boot. The fix was to load an older version of the software
on the NMC (can't remember the version), reconfigure the IP Address so that
it wasn't zeros and then it would boot up normally. At this point, I was
able to upload the latest version of the software back on the NMC.
I don't if this is the same problem you are having, but is sounds similar.
Take care.
Alan D. Criado
At 07:16 PM 11/16/98 -0600, you wrote:
>I picked up a used 1866 bundle.
>
>The NMC card keeps rebooting. It gets far enough to display the menu on
>the console, but it reboots before anything can be entered.
>
>It appears to be running 5.4.95 on the NMC.
>
>Any ideas to fix this?
>
>Brian
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) Two subnets on one hiperARC From: Laurie J Collinsworth <ljc1@cornell.edu> Date: 1998-11-16 21:04:05
Hello,
I have added a second subnet to an existing hiperARC setup to
the eth:2 port. I would like to have two ip pools one for each
subnet routed out the appropriate port.
Currently all I have done is configured the second ethernet port
eth:2. I cannot connect to the hiperARC using the new ip address
on eth:2 interface. If I do a traceroute, it finds the eth:1
interface and continues looking but never finds the new ip address.
Some sort for rip announcement is occuring for the gateway to send
the request to the eth:1 interface instead of the eth:2 interface.
Perhaps what I want is not possible or could be done with filters.
Or perhaps the hiperARC will only answer to the "LAN Host address"
but will pass data on to modems connected with the new ip subnet
properly.
In case people are curious, we have a chassis with 11 hiperDSP cards,
2 hiperARCS, and not enough ip addresses on a 24 bit subnet. We could
just split the two subnets and have one on each hiperARC but we would
like the redundancy of letting one hiperARC take over all the traffic
if the other should fail.
Ain't life fun.
-Laurie
Subject:RE: (usr-tc) v.90 connect problems From: Mario M. Bustamante <mario@accesspro.net> Date: 1998-11-17 02:50:41
I posted the messages below at a time when we were having an
unacceptable number of customer complaints. At present we
consider the problems to be resolved.
The biggest change came when we removed the Hiper ARC from our
network and co located it with a CLEC we are working with. When
we had it in our main POP (Miami), we had channelized T-1's and
when we moved it and used PRI's the problems disappeared. I am
convinced that we must have had some setting wrong, because the
performance increase was dramatic and nothing else changed.
We found that almost all connection problems dealing with
Sportster modems disappeared. We contacted every single customer
that reported a connection problem with USR modems, and asked
them to call the new access number, and without exception, they
connected without a problem (to the same equipment!).
We were also able to take care of most of the other customers by
helping them upgrade their modems to the latest drivers. Almost
all complaints were from LT Winmodems, and their latest code
works fine.
The performance of the Hiper Arc is great. We are slowly moving
more customers to the new POP, and we are getting a lot of
positive comments.
If this keeps up we will upgrade all our Net Servers to Hiper
Arc's win the coming months. The latest trade up program looks
very appealing...
Thanks,
_______________________________________________
Mario M. Bustamante, President
AccessPro Communications Inc.
Miami, Florida
http://www.AccessPro.net mario@accesspro.net
_______________________________________________
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of
> Mario M. Bustamante
> Sent: Tuesday, November 10, 1998 9:34 PM
> To: usr-tc@lists.xmission.com
> Subject: RE: (usr-tc) v.90 connect problems
>
>
> I posted a message earlier describing connection problems with
> Sportster modems. Up until recently, we assumed that our
> customers were complaining because they were using the
> cheap V90
> modems, and the problem was on their end.
>
> As the complaints started to add up, I had out support people
> fill out a form to gather information from problem users. The
> process included asking the end user to run "winipcfg.exe" on
> their Windows 95 or 98 systems and keep a log of good and bad
> connections. Since we have mostly Net Server / Quad modem
> equipment and only one Hiper ARC / DSP, we wanted to know what
> equipment was causing the most problems and with what kind of
> modem (on the customer end).
>
> We have somewhat over 2,200 users, and we have always
> recommended
> that they buy USR/3Com modems, so we were shocked at the number
> of Sportster users having problems (and they were mad at us for
> making them spend extra money to buy more problems).
> We were also
> shocked to find that a lot of them were connecting to BellSouth
> with a much better V90 connection (they use Ascend equipment).
>
> I do not want to post results publicly, but after compiling the
> results of almost 200 complaints, I can tell you that
> most of the
> problems were related to the Hiper ARC / DSP, and that the
> percentage of complaints from Sportster users was alarming.
>
> I would love to get someone that knows Sportsters on a
> conference
> call with someone that knows Hiper ARCs so we can see where the
> problems lie. I know that there are a lot of bad phone
> lines and
> a lot of people who set up their modems with the wrong
> settings,
> but somebody has to help us give them tech support.
>
> Any ideas?
>
> Thanks,
> _______________________________________________
>
> Mario M. Bustamante, President
> AccessPro Communications Inc.
> Miami, Florida
>
> Internet Service Providers, Web Hosting & Design
> Microsoft Certified Web Presence Providers
> Wide Area Networks, Security, Intranets.
> http://www.AccessPro.net mario@accesspro.net
> _______________________________________________
>
> > -----Original Message-----
> > From: owner-usr-tc@lists.xmission.com
> > [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of
> > John Powell
> > Sent: Saturday, November 07, 1998 10:15 PM
> > To: usr-tc@lists.xmission.com
> > Subject: Re: (usr-tc) v.90 connect problems
> >
> >
> > >OK, but a fair number of the people having problems
> > have Sportsters.
> > >(Some Winmodems, some not.) I wasn't aware there was
> > any new Sportster
> > >code out... if there is, where's it hidden? If
> > there isn't, is this one
> > >of the server-side problems?
> >
> > I was responding to your comments on LT Winmodems, you
> > did not mention
> > Sportsters. I know of no significant issues with
> > interop with Sportsters
> > (Winmodem or controller based) and Total Control modems.
> >
> > There will likely be improvements with the client side
> > (all vendors) for
> > quite some time. There are quite a few tweeks to
> > enhance performance and
> > coverage in the labs already. Heck, we are still
> > improving V.34 4 years
> > later.
> >
> > If you can provide details on problems with
> > Sportsters, I will gladly take
> > this off-line and put you in contact with my peers on
> > the PCD (home of the
> > Sportster) side of the house.
> >
> > JP
> >
> > PS. No, there are no recent releases of Sportster
> > code related to V.90
> > fixes. Overall, they are working quite well.
> >
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to
> > "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and
> > old messages send
> > "help" to the same address. Do not use quotes in
> > your message.
> >
>
>
> -
> To unsubscribe to usr-tc, send an email to
> "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and
> old messages send
> "help" to the same address. Do not use quotes in
> your message.
>
Subject:Re: (usr-tc) PRI settings blown away by NMC From: Ralph Helfenberger <rhelfenberger@comlight.ch> Date: 1998-11-17 08:21:27
Did you check the AutoResponse settings? I had a similar problem.
AutoResponse was responsible.
Ralph
Chris Hanes wrote:
>
> Hey Brian,
>
> Thanks for the reponse. Auto config was disabled. I've tried restoring
> NMC from factory default several times to no avail. When I boot the
> box,
> I've been watching the pri cards chassis slot settings. The pri card
> seems
> to boot quickly and I'm able to see the settings briefly before they are
> wiped out, leaving the pri card seeing only itself in slot 1 and the
> netserver
> card in slot 16. Any other ideas?
>
> Thanks,
> Chris Hanes
>
> Brian K McIntire wrote:
> >
> > Sounds like you have "Auto config upon card initialization" enabled on your
> > NMC.
> >
> > 1. Load TCM
> > 2. Bring up your chassis
> > 3. Click on the NMC
> > 4. Click on Configure/program settings/configuration group
> > 5. Make sure auto config is disabled.
> > 6. If it was enabled set it and click ok.
> > 7. go to configure/action commands and save chassis to nvram (don't really
> > need this one but do it anyway)
> > 8. Save UI to eeprom
> > 9. Reboot
> > 10. If the setting was disabled then restore your NMC (not chassis) from
> > default by clicking on the NMC and going to configure/action commands and
> > choosing restore NMC from default
> > 11. then perform steps 7 & 8 and then 9 from above
> > This should solve your problem. Remember you will need to reconfigure the
> > Logging group on the NMC if you have restored it from default and wish to do
> > accounting
> >
> > ========================================================================
> > = Brian K. McIntire bmcintire@commnet.com =
> > = Systems Engineer III 317-558-5050 x113 =
> > = CommNetPlus, Indianapolis, IN http://www.commnet.com =
> > = Your Remote Access and Security Experts! =
> > = =
> > = Our experienced technicians can design and install your high speed =
> > = network for you. From Operating Systems and servers to routers and =
> > = firewalls nobody does it better than CommNetPlus. Our Technical =
> > = Support staff is available to you 24 hours a day to meet your needs.=
> > = Offering remote management and monitoring packages for dedicated =
> > = clients. (Specializing in ISP's). "Let us do the work for you!" =
> > = =
> > = Firm partnerships est. with current and emerging leaders of today's =
> > = technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
> > = Serving North America and parts of Canada. Reselling to the world. =
> > ========================================================================
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
Subject:RE: (usr-tc) Two subnets on one HiPerARC From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-17 09:15:03
>Hello,
>I have added a second subnet to an existing hiperARC setup to
>the eth:2 port. I would like to have two ip pools one for each
>subnet routed out the appropriate port.
Easy enough. Create an additional pool on the HiPer ARC and specify that
pool name under the user setting in RADIUS for the users you wish to have
it. Make the pool private or specify pool names for everyone.
>Currently all I have done is configured the second ethernet port
>eth:2. I cannot connect to the hiperARC using the new ip address
>on eth:2 interface. If I do a traceroute, it finds the eth:1
>interface and continues looking but never finds the new ip address.
Some sort for rip announcement is occuring for the gateway to send
the request to the eth:1 interface instead of the eth:2 interface.
Perhaps what I want is not possible or could be done with filters.
Or perhaps the hiperARC will only answer to the "LAN Host address"
but will pass data on to modems connected with the new ip subnet
properly.
In case people are curious, we have a chassis with 11 hiperDSP cards,
2 hiperARCS, and not enough ip addresses on a 24 bit subnet. We could
just split the two subnets and have one on each hiperARC but we would
like the redundancy of letting one hiperARC take over all the traffic
if the other should fail.
Ain't life fun.
-Laurie
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:RE: (usr-tc) NMC rebooting From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-17 09:15:04
You may also want to flip dips 5 & 6 on and reseat before you flash it with
PCSDL. It shouldn't make a difference but I have seen it make one over a
hundred times
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Tatai SV Krishnan
Sent: Monday, November 16, 1998 8:14 PM
Cc: usr-tc@xmission.com
This could be because
1. you have programmed a 0.0.0.0 as ip address of the NMC
2. You have a bad flash
Try this.
Download 4.3.9 nmc code ( or any 4.x code ) use a program called pcsdl
comes with the code ) and download that code to the NMC and then program
the proper ip address save it and then upgrade it to 5.x code
If this does not help then your flash may be corrupted.
krish
\ T.S.V. Krishnan \
\ Network System Engineer \ ( : - : )
\ 3Com ............ \
----------------------------------------------/
tkrishna@bubba.ae.usr.com
----------------------------/ http://interproc.ae.usr.com ----/
The Yadda Yadda Search - for simple anwers -
http://interproc.ae.usr.com/tkb.html
Any Sufficiently advanced bug is indistinguishable for a feature.
- Rick Kulawiec
On Mon, 16 Nov 1998, Brian Elfert wrote:
> I picked up a used 1866 bundle.
>
> The NMC card keeps rebooting. It gets far enough to display the menu on
> the console, but it reboots before anything can be entered.
>
> It appears to be running 5.4.95 on the NMC.
>
> Any ideas to fix this?
>
> Brian
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:RE: (usr-tc) v.90 connect problems From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-17 09:17:26
PRI's and T1's are two different animals. Who knows how your T1 was
configured. Your Telco could have been padding it. The Receiver gain could
have been wrong. The provisioning may have been wrong. Don't believe this
is related to some setting your chassis. I've seen this a thousand times on
a thousand different AOL chassis. PRI's, by design, are superior to T1's
and have significantly less things that could go wrong (assuming your
provisioned has any notion of what he is doing) It was probably your T1.
-----Original Message-----
[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Mario M. Bustamante
Sent: Tuesday, November 17, 1998 1:51 AM
I posted the messages below at a time when we were having an
unacceptable number of customer complaints. At present we
consider the problems to be resolved.
The biggest change came when we removed the Hiper ARC from our
network and co located it with a CLEC we are working with. When
we had it in our main POP (Miami), we had channelized T-1's and
when we moved it and used PRI's the problems disappeared. I am
convinced that we must have had some setting wrong, because the
performance increase was dramatic and nothing else changed.
We found that almost all connection problems dealing with
Sportster modems disappeared. We contacted every single customer
that reported a connection problem with USR modems, and asked
them to call the new access number, and without exception, they
connected without a problem (to the same equipment!).
We were also able to take care of most of the other customers by
helping them upgrade their modems to the latest drivers. Almost
all complaints were from LT Winmodems, and their latest code
works fine.
The performance of the Hiper Arc is great. We are slowly moving
more customers to the new POP, and we are getting a lot of
positive comments.
If this keeps up we will upgrade all our Net Servers to Hiper
Arc's win the coming months. The latest trade up program looks
very appealing...
Thanks,
_______________________________________________
Mario M. Bustamante, President
AccessPro Communications Inc.
Miami, Florida
http://www.AccessPro.net mario@accesspro.net
_______________________________________________
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of
> Mario M. Bustamante
> Sent: Tuesday, November 10, 1998 9:34 PM
> To: usr-tc@lists.xmission.com
> Subject: RE: (usr-tc) v.90 connect problems
>
>
> I posted a message earlier describing connection problems with
> Sportster modems. Up until recently, we assumed that our
> customers were complaining because they were using the
> cheap V90
> modems, and the problem was on their end.
>
> As the complaints started to add up, I had out support people
> fill out a form to gather information from problem users. The
> process included asking the end user to run "winipcfg.exe" on
> their Windows 95 or 98 systems and keep a log of good and bad
> connections. Since we have mostly Net Server / Quad modem
> equipment and only one Hiper ARC / DSP, we wanted to know what
> equipment was causing the most problems and with what kind of
> modem (on the customer end).
>
> We have somewhat over 2,200 users, and we have always
> recommended
> that they buy USR/3Com modems, so we were shocked at the number
> of Sportster users having problems (and they were mad at us for
> making them spend extra money to buy more problems).
> We were also
> shocked to find that a lot of them were connecting to BellSouth
> with a much better V90 connection (they use Ascend equipment).
>
> I do not want to post results publicly, but after compiling the
> results of almost 200 complaints, I can tell you that
> most of the
> problems were related to the Hiper ARC / DSP, and that the
> percentage of complaints from Sportster users was alarming.
>
> I would love to get someone that knows Sportsters on a
> conference
> call with someone that knows Hiper ARCs so we can see where the
> problems lie. I know that there are a lot of bad phone
> lines and
> a lot of people who set up their modems with the wrong
> settings,
> but somebody has to help us give them tech support.
>
> Any ideas?
>
> Thanks,
> _______________________________________________
>
> Mario M. Bustamante, President
> AccessPro Communications Inc.
> Miami, Florida
>
> Internet Service Providers, Web Hosting & Design
> Microsoft Certified Web Presence Providers
> Wide Area Networks, Security, Intranets.
> http://www.AccessPro.net mario@accesspro.net
> _______________________________________________
>
> > -----Original Message-----
> > From: owner-usr-tc@lists.xmission.com
> > [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of
> > John Powell
> > Sent: Saturday, November 07, 1998 10:15 PM
> > To: usr-tc@lists.xmission.com
> > Subject: Re: (usr-tc) v.90 connect problems
> >
> >
> > >OK, but a fair number of the people having problems
> > have Sportsters.
> > >(Some Winmodems, some not.) I wasn't aware there was
> > any new Sportster
> > >code out... if there is, where's it hidden? If
> > there isn't, is this one
> > >of the server-side problems?
> >
> > I was responding to your comments on LT Winmodems, you
> > did not mention
> > Sportsters. I know of no significant issues with
> > interop with Sportsters
> > (Winmodem or controller based) and Total Control modems.
> >
> > There will likely be improvements with the client side
> > (all vendors) for
> > quite some time. There are quite a few tweeks to
> > enhance performance and
> > coverage in the labs already. Heck, we are still
> > improving V.34 4 years
> > later.
> >
> > If you can provide details on problems with
> > Sportsters, I will gladly take
> > this off-line and put you in contact with my peers on
> > the PCD (home of the
> > Sportster) side of the house.
> >
> > JP
> >
> > PS. No, there are no recent releases of Sportster
> > code related to V.90
> > fixes. Overall, they are working quite well.
> >
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to
> > "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and
> > old messages send
> > "help" to the same address. Do not use quotes in
> > your message.
> >
>
>
> -
> To unsubscribe to usr-tc, send an email to
> "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and
> old messages send
> "help" to the same address. Do not use quotes in
> your message.
>
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:(usr-tc) New Features From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-17 09:23:28
Does anyone know if 3COM posts information on features that will be
supported in the next full featured release. I'm fairly sure they didn't in
the past, but then it didn't matter. I had the inside scoop. Now, where do
I go for the answers. I have a lot of customer's with questions. I can
answer most of them because I'm already fairly well versed in what will be
included in TCS 3.5. I would love to see an "Expected to be released" paper
or something. If there is no such thing then my question is, Why not? You
don't have to guarantee the release of the feature. Just put a list out
there stating what will be included should the feature be completed and
tested on time. If the customer's get upset when a feature is delayed that
is their problem.
On Mon, 16 Nov 1998, Tatai SV Krishnan wrote:
> Download 4.3.9 nmc code ( or any 4.x code ) use a program called pcsdl (
> comes with the code ) and download that code to the NMC and then program
> the proper ip address save it and then upgrade it to 5.x code
How exactly can I load the code if the thing never boots up fully?
Brian
Subject:(usr-tc) MultiLink PPP and ISDN From: Taino d Johnston <usr-list@accesscom.com> Date: 1998-11-17 09:41:30
A few quick questions:
What needs to be done to enable MultiLink PPP and dual channel ISDN? Does
'set mp on' do it and that is it? Will there be any problems if one call
comes say into chassis #1 and the second call comes into chassis #2?
We are running a three TC chassis containing Quad Digital & Digital/Analog
modem cards with PRI lines. All our Quad modem cards, NETserver PRI cards,
Dual PRI cards and NMC cards are running the current versions of the
software.
Taino Johnston
Manager, Technical Support
Access Internet Communications
+----------------------------------------------------------------+
| Taino d Johnston | Phone: (408) 777-8190 |
| Manager, Technical Support | Tech Support: (408) 342-0551 |
| | Fax: (408) 777-8191 |
| Access Internet Communications | http://www.accesscom.com/ |
| tdj@accesscom.com | support@accesscom.com |
+----------------------------------------------------------------+
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian Elfert
|Sent: Tuesday, November 17, 1998 9:26 AM
|To: Tatai SV Krishnan
|Cc: usr-tc@xmission.com
|Subject: Re: (usr-tc) NMC rebooting
|
|
|
|
|On Mon, 16 Nov 1998, Tatai SV Krishnan wrote:
|
|> Download 4.3.9 nmc code ( or any 4.x code ) use a program called pcsdl (
|> comes with the code ) and download that code to the NMC and then program
|> the proper ip address save it and then upgrade it to 5.x code
|
|How exactly can I load the code if the thing never boots up fully?
|
PCSDL.
Subject:RE: (usr-tc) accounting problem through NMC From: allen_lai@3com.com Date: 1998-11-17 09:58:23
Brian
Thanks for your response.
Fro your questions,the answer is yes!
I'm sure the NT server IP address is in the logging server list.
My question is why the S/A in NT server lost their clients.
Allen
"Brian K McIntire" <bmcintire@commnet.com> on 11/17/98 06:21:01 AM
Please respond to usr-tc@lists.xmission.com
cc: (Allen Lai/CN/3Com)
Note: Some recipients have been dropped due to syntax errors.
Please refer to the "$AdditionalHeaders" item for the complete headers.
Hello,all
We have installed a USR S/A server in a NT workstation(Intel pro233) to
collect the accounting informaiton from NMC.
We created about 60 NMC/chassis---raduis accounting client in one S/A
server.
At first, everything works OK.We got accounting information from all
created clients.
But gradually, some clients will disappered.
*********
What do you mean by this?
Have you used TCM to open the NMC Logging Group info and make sure you see
the Primary Logging Server listed?
We can't got any accounting
information of these disappered client now from S/A now.
When ping the NMC, it's ok.We can also managned these chassis through TCM.
Any hints?
Allen
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
That is where you use pcsdl - This program will run on a pc and as soon
as you start will force the download of the code to the card.
krish
\ T.S.V. Krishnan \
\ Network System Engineer \ ( : - : )
\ 3Com ............ \
----------------------------------------------/
tkrishna@bubba.ae.usr.com
----------------------------/ http://interproc.ae.usr.com ----/
The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
Any Sufficiently advanced bug is indistinguishable for a feature.
- Rick Kulawiec
On Tue, 17 Nov 1998, Brian Elfert wrote:
>
>
> On Mon, 16 Nov 1998, Tatai SV Krishnan wrote:
>
> > Download 4.3.9 nmc code ( or any 4.x code ) use a program called pcsdl (
> > comes with the code ) and download that code to the NMC and then program
> > the proper ip address save it and then upgrade it to 5.x code
>
> How exactly can I load the code if the thing never boots up fully?
>
> Brian
>
Open a ticket and send the crash dump ( sho board crash ) to a tech.
They can analyze the crash and tell you what the problem is.
krish
\ T.S.V. Krishnan \
\ Network System Engineer \ ( : - : )
\ 3Com ............ \
----------------------------------------------/
tkrishna@bubba.ae.usr.com
----------------------------/ http://interproc.ae.usr.com ----/
The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
Any Sufficiently advanced bug is indistinguishable for a feature.
- Rick Kulawiec
On Tue, 17 Nov 1998, Marcelo Souza wrote:
>
> Yesterday I had a very curious problem.
> I discovered that some ports om my TCS had no filter set up, so I
> used the command:
>
> set interface slot:1/mod:[1-30] input_filter dial.in
>
> And I could see that some modems had a wrong value in the input
> filter field: @mailbo (don't ask me why)
> Then I send the commend again, and send it two times for each
> slot, when I was configuring the last slot (#6) the TC hang and rebooted.
>
> I'm using HARC 4.0.29.
>
> Have any one seem some thing like this?
>
> - Marcelo
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) MultiLink PPP and ISDN From: Taino d Johnston <usr-list@accesscom.com> Date: 1998-11-17 10:07:06
You say I need to enable a MPIP server and configure MPIP functionality on
all your NETServers. What exactly does that mean? Is a MPIP server
something extra that needs to be purchased? What configuration needs to be
done on the NETServers?
Taino Johnston
Manager, Technical Support
Access Internet Communications
>Thus spake Taino d Johnston
>>What needs to be done to enable MultiLink PPP and dual channel ISDN? Does
>>'set mp on' do it and that is it? Will there be any problems if one call
>>comes say into chassis #1 and the second call comes into chassis #2?
>
>You need to enable an MPIP server and configure MPIP functionality on
>all your NETServers. MPIP is the protocol that controls cross-platform
>channel bundling like you're wanting to do.
>--
>Jeff McAdams Email: jeffm@iglou.com
>Head Network Administrator Voice: (502) 966-3848
>IgLou Internet Services (800) 436-4456
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
+----------------------------------------------------------------+
| Taino d Johnston | Phone: (408) 777-8190 |
| Manager, Technical Support | Tech Support: (408) 342-0551 |
| | Fax: (408) 777-8191 |
| Access Internet Communications | http://www.accesscom.com/ |
| tdj@accesscom.com | support@accesscom.com |
+----------------------------------------------------------------+
Subject:Re: (usr-tc) PRI settings blown away by NMC From: Chris Hanes <chris@internetcreations.com> Date: 1998-11-17 11:27:14
I'm still having no luck preventing the NMC card from blowing
away my PRI slot device configuration. Autoconfig on the NMC is
disabled
and I'm configuring the PRI from the console. Any further suggestions?
Thanks,
Chris
Chris Hanes wrote:
>
> Hey all,
>
> I'm trying to map modems to slots under chassis slot device
> configuration on a dual pri nac. The settings take so long
> as my NMC is not plugged in. Once plugged in the NMC obscures these
> settings and prevents me from reentering the settings on the pri card
> (i.e. the settings won't take when the NMC is plugged in). I've been
> through a couple pri cards and figure I must be doing something stupid.
> I'm running tcs 3.3 (3.02 on the PRI nac and 5.5.5 on the NMC)
> Any help would be IMMENSELY appreciated.
>
> Thanks,
> Chris Hanes
> Internet Creations
We use a ping monitor program to check that interfaces are alive, and have
been seeing one Netserver keep going up and down in the monitor. We can
reach it via telnet, but when we ping it we see the following-
(caseyc@inca):/etc/raddb>ping tsbrn2
PING tsbrn2.gate.net: (207.36.0.99): 56 data bytes
64 bytes from 207.36.0.99: icmp_seq=1 ttl=244 time=65 ms
64 bytes from 207.36.0.99: icmp_seq=2 ttl=244 time=79 ms
64 bytes from 207.36.0.99: icmp_seq=3 ttl=244 time=72 ms
64 bytes from 207.36.0.99: icmp_seq=4 ttl=244 time=1955 ms
wrong data byte #24 should be 0x18 but was 0x4a
36 51 a6 76 0 b 7e 4 8 9 a b c d e f 10 11 12 13 14 15 16 17 4a 45
1a 1b
20 21 22 23 24 25 26 27 28 29 2a 2b 2c 2d 2e 2f
64 bytes from 207.36.0.99: icmp_seq=6 ttl=244 time=68 ms
64 bytes from 207.36.0.99: icmp_seq=7 ttl=244 time=77 ms
64 bytes from 207.36.0.99: icmp_seq=8 ttl=244 time=81 ms
64 bytes from 207.36.0.99: icmp_seq=10 ttl=244 time=79 ms
wrong data byte #24 should be 0x18 but was 0x30
36 51 a6 7c 0 b 8d 38 8 9 a b c d e f 10 11 12 13 14 15 16 17 30
30 31 3
20 21 22 23 24 25 26 27 28 29 2a 2b 2c 2d 2e 2f
64 bytes from 207.36.0.99: icmp_seq=13 ttl=244 time=73 ms
wrong data byte #42 should be 0x2a but was 0x41
36 51 a6 7f 0 b 98 f5 8 9 a b c d e f 10 11 12 13 14 15 16 17 18
19 1a 1
20 21 22 23 24 25 26 27 28 29 41 43 2c 2d 2e 2f
64 bytes from 207.36.0.99: icmp_seq=14 ttl=244 time=65 ms
64 bytes from 207.36.0.99: icmp_seq=6 ttl=243 time=8645 ms (DUP!)
64 bytes from 207.36.0.99: icmp_seq=16 ttl=244 time=69 ms
64 bytes from 207.36.0.99: icmp_seq=8 ttl=243 time=9067 ms (DUP!)
64 bytes from 207.36.0.99: icmp_seq=18 ttl=244 time=146 ms
wrong data byte #26 should be 0x1a but was 0xb3
36 51 a6 84 0 b a7 96 8 9 a b c d e f 10 11 12 13 14 15 16 17 18
19 b3 c
20 21 22 23 0 0 0 0 0 0 0 0 0 0 2e 2f
64 bytes from 207.36.0.99: icmp_seq=19 ttl=244 time=55 ms
^C
----tsbrn2.gate.net PING Statistics----
21 packets transmitted, 13 packets received, +2 duplicates, 38% packet
loss
round-trip min/avg/max = 55/1373/9067 ms
Subject:(usr-tc) Stuck in ringRcvd From: Brian <signal@shreve.net> Date: 1998-11-17 11:46:36
We had a modem that when people dialed our number, it would just be dead
silence. I narrowed it down to a card, and did a Program Settings on it,
and saw that the operational status for one of the modems was stuck in
ringRcvd. I tried to software reset this modem, and it stayed in
ringRcvd. The only fix was to hardware reset the card.
This happens quite a bit.
Has anyone else seen this? I am running Release hdm/arc code
(1.2.5/4.1.11).
What is the cause? The fix?
brian
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:RE: (usr-tc) NMC rebooting From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-17 12:11:58
FYI,
Actually, what happens is as the NMC boots up it will initialize hardware
(factory settings on a chip) and then software. Your card is probably
rebooting because of a software problem. Before it initializes completely,
runs into problems, and begins rebooting (NMC's usually only reboot when the
final initialization failed) it will be in a state (thanks to the factory
chip) where it will respond to PCSDL and take a software download. Your
PCSDL program will continue attempting communication until the NMC boots up
to the point where the factory chip has told it how to respond to PCSDL.
Therefore, this works whether you have software on your NMC or not.
For those of you that have tried to access the console while your NMC is
doing something like this you were not able to because the NMC needs to be
past the point of initializing the software in order to respond to commands
on the console.
>That is where you use pcsdl - This program will run on a pc and as soon
>as you start will force the download of the code to the card.
>
>krish
>
>-----------------------------------------
> \ T.S.V. Krishnan \
> \ Network System Engineer \ ( : - : )
> \ 3Com ............ \
> ----------------------------------------------/
>tkrishna@bubba.ae.usr.com
>----------------------------/ http://interproc.ae.usr.com ----/
>The Yadda Yadda Search - for simple anwers -
http://interproc.ae.usr.com/tkb.html
>-------------------------------------------------------------------------\
> Any Sufficiently advanced bug is indistinguishable for a feature.
> - Rick Kulawiec
>-------------------------------------------------------------------------/
On Tue, 17 Nov 1998, Brian Elfert wrote:
>
>
> On Mon, 16 Nov 1998, Tatai SV Krishnan wrote:
>
> > Download 4.3.9 nmc code ( or any 4.x code ) use a program called pcsdl
> > comes with the code ) and download that code to the NMC and then program
> > the proper ip address save it and then upgrade it to 5.x code
>
> How exactly can I load the code if the thing never boots up fully?
>
> Brian
>
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:Re: (usr-tc) Stuck in ringRcvd From: sagarwal@crash.ae.usr.com Date: 1998-11-17 12:37:09
What does "List Interfaces " command on Hiper ARC show. Does it show that
the int is down in oper status?
Sanjay
On Tue, 17 Nov 1998, Brian wrote:
> We had a modem that when people dialed our number, it would just be dead
> silence. I narrowed it down to a card, and did a Program Settings on it,
> and saw that the operational status for one of the modems was stuck in
> ringRcvd. I tried to software reset this modem, and it stayed in
> ringRcvd. The only fix was to hardware reset the card.
>
> This happens quite a bit.
>
> Has anyone else seen this? I am running Release hdm/arc code
> (1.2.5/4.1.11).
>
> What is the cause? The fix?
>
> brian
>
>
> --------------------------------------------------------------------------
> Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) Stuck in ringRcvd From: sagarwal@crash.ae.usr.com Date: 1998-11-17 12:37:09
What does "List Interfaces " command on Hiper ARC show. Does it show that
the int is down in oper status?
Sanjay
On Tue, 17 Nov 1998, Brian wrote:
> We had a modem that when people dialed our number, it would just be dead
> silence. I narrowed it down to a card, and did a Program Settings on it,
> and saw that the operational status for one of the modems was stuck in
> ringRcvd. I tried to software reset this modem, and it stayed in
> ringRcvd. The only fix was to hardware reset the card.
>
> This happens quite a bit.
>
> Has anyone else seen this? I am running Release hdm/arc code
> (1.2.5/4.1.11).
>
> What is the cause? The fix?
>
> brian
>
>
> --------------------------------------------------------------------------
> Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) Stuck in ringRcvd From: Brian <signal@shreve.net> Date: 1998-11-17 12:38:32
On Tue, 17 Nov 1998 sagarwal@crash.ae.usr.com wrote:
>
> What does "List Interfaces " command on Hiper ARC show. Does it show that
> the int is down in oper status?
>
> Sanjay
I have already reset the card. In the past usually the first thing I
check for when we get fast busies or problems from the modems is a list
int. I do not recall ever seeing the interface down when this condition
arises, usually it is up.
Brian
>
>
> On Tue, 17 Nov 1998, Brian wrote:
>
> > We had a modem that when people dialed our number, it would just be dead
> > silence. I narrowed it down to a card, and did a Program Settings on it,
> > and saw that the operational status for one of the modems was stuck in
> > ringRcvd. I tried to software reset this modem, and it stayed in
> > ringRcvd. The only fix was to hardware reset the card.
> >
> > This happens quite a bit.
> >
> > Has anyone else seen this? I am running Release hdm/arc code
> > (1.2.5/4.1.11).
> >
> > What is the cause? The fix?
> >
> > brian
> >
> >
> > --------------------------------------------------------------------------
> > Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> > Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> > signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> > (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:Re: (usr-tc) Stuck in ringRcvd From: Brian <signal@shreve.net> Date: 1998-11-17 12:38:32
On Tue, 17 Nov 1998 sagarwal@crash.ae.usr.com wrote:
>
> What does "List Interfaces " command on Hiper ARC show. Does it show that
> the int is down in oper status?
>
> Sanjay
I have already reset the card. In the past usually the first thing I
check for when we get fast busies or problems from the modems is a list
int. I do not recall ever seeing the interface down when this condition
arises, usually it is up.
Brian
>
>
> On Tue, 17 Nov 1998, Brian wrote:
>
> > We had a modem that when people dialed our number, it would just be dead
> > silence. I narrowed it down to a card, and did a Program Settings on it,
> > and saw that the operational status for one of the modems was stuck in
> > ringRcvd. I tried to software reset this modem, and it stayed in
> > ringRcvd. The only fix was to hardware reset the card.
> >
> > This happens quite a bit.
> >
> > Has anyone else seen this? I am running Release hdm/arc code
> > (1.2.5/4.1.11).
> >
> > What is the cause? The fix?
> >
> > brian
> >
> >
> > --------------------------------------------------------------------------
> > Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> > Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> > signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> > (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:Re: (usr-tc) MultiLink PPP and ISDN From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-17 12:54:06
Thus spake Taino d Johnston
>What needs to be done to enable MultiLink PPP and dual channel ISDN? Does
>'set mp on' do it and that is it? Will there be any problems if one call
>comes say into chassis #1 and the second call comes into chassis #2?
You need to enable an MPIP server and configure MPIP functionality on
all your NETServers. MPIP is the protocol that controls cross-platform
channel bundling like you're wanting to do.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) MultiLink PPP and ISDN From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-17 13:13:29
Thus spake Taino d Johnston
>You say I need to enable a MPIP server and configure MPIP functionality on
>all your NETServers. What exactly does that mean? Is a MPIP server
>something extra that needs to be purchased? What configuration needs to be
>done on the NETServers?
MPIP is built into the NETServers, you should be able to check out the
documentation to find the commands to enable it. One NETServer will
need to be designated as an MPIP server and requires some special
configuration. It should all be in the docs...unfortunately, I don't
have the information in front of me at the moment.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:(usr-tc) Double Pings ans Slow Downloads From: Carl Jagerski <carll@forcomm.net> Date: 1998-11-17 13:15:11
Hello Everyone,
I currently have pop sites setup with both T.C. HiperDSPs and regular
analog lines to 33,600 modems. These problems are at all pop sites.
When you dial into the analog lines, download speeds range from 1 to 3kbsp
But, when you dial into the TC boxes, download speeds start at around 1 and
slowly drop to almost nothing. This is very consistent and well test by us
with
different modems (ie. 33,600, X2, V.90 on a few different brands).
We also noticed that the pings are different through the digital than the
analog.
This is a typical ping through the analog modems:
PING coos.dartmouth.edu (129.170.16.50): 56 data bytes
64 bytes from 129.170.16.50: icmp_seq=0 ttl=235 time=463.6 ms
64 bytes from 129.170.16.50: icmp_seq=1 ttl=235 time=310.8 ms
64 bytes from 129.170.16.50: icmp_seq=2 ttl=235 time=600.8 ms
64 bytes from 129.170.16.50: icmp_seq=3 ttl=235 time=370.7 ms
This is the same ping dialed into the TC Boxes from the same modem:
PING coos.dartmouth.edu (129.170.16.50): 56 data bytes
64 bytes from 129.170.16.50: icmp_seq=0 ttl=235 time=196.9 ms
64 bytes from 129.170.16.50: icmp_seq=0 ttl=234 time=206.1 ms (DUP!)
64 bytes from 129.170.16.50: icmp_seq=1 ttl=235 time=310.8 ms
64 bytes from 129.170.16.50: icmp_seq=1 ttl=234 time=320.0 ms (DUP!)
64 bytes from 129.170.16.50: icmp_seq=2 ttl=235 time=250.8 ms
64 bytes from 129.170.16.50: icmp_seq=2 ttl=234 time=262.7 ms (DUP!)
64 bytes from 129.170.16.50: icmp_seq=3 ttl=235 time=190.9 ms
64 bytes from 129.170.16.50: icmp_seq=3 ttl=234 time=200.0 ms (DUP!)
What is causing the double pings? Could this be related to the slowing
of the download? We are totally confused here. Any ideas would help at this
point. 3COM has already been through our boxes and said the setting all
looked good to them. (but what do they know?)
Carl Jagerski
Network Administrator, Forward Communications
carll@forcomm.net
412-378-4490 Fax: 412-375-0156
Subject:Re: (usr-tc) MultiLink PPP and ISDN From: Jason W <jwatkins@iland.net> Date: 1998-11-17 13:19:07
Below I have pasted an email that a 3com
support tech sent me on how to properly
configure MPIP on Netserver cards.
You can also find this information on the
release notes for Netserver version 3.4
Hope this helps.
Solution ID: 112.0.3158865.1615069
Type: published-Public
Status: Stds & tech reviews complete
Shared: Yes
Title:
How to set up MPIP on a Total Control NETServer PRI
Goal:
How to set up MPIP on a Total Control NETServer PRI
Facts:
Total Control NETServer PRI
Total Control Dual PRI
MPIP Unix Control Server
MPIP
Symptoms:
Problems with setting up MPIP
Cannot do MPIP with 2 Total Control NETServer PRI's
Setting up MPIP with a Unix control server
Fixes:
How to set up MPIP with 2 Total Control NETServer PRI, and a
MPIP Control Server
MPIP - Multilink PPP Interspan Protocol (MPIP)
1) Configure both Total Control NETServer PRI 1 and Total
Control Netserver PRI 2 for use with MPIP Control Server 1 by
entering the following command at both netserver's:
set mpipserver 1.1.1.1/1030 mpipsecret
2) Add both Total Control NETServer PRI 1 and Total Control
NETServer PRI 2 as clients of MPIP Control Server 1 by
entering the commands below at the Total Control NETServer PRI
1:add mpipclient 1.1.1.1 mpipsecret ------------> adds the Total
Control NETServer 1 as its own client
add mpipclient 1.1.1.2 mpipsecret -------------> add the Total
Control NETServer 2 as a client
3)Set the NTP time server across the network by entering the
following command at the MPIP Control Server's command prompt:
set time 1.1.1.3
Then you can reset the time by entering this command:
reset time
4) Turn the MPIP Control Server on, at the command prompt of
the Total Control NETServer PRI 1, by entering the following
command:
set mpipserver on
MPIP is now up and running with the Total Control NETServer
PRI 1 serving as the MPIP Control Server and both NETServer 1
and NETServer 2 acting as clients.
Setting the Total Control NETServer PRI as a MPIP server:
1) Setup the chosen Total Control NETServer PRI as the MPIP
server and configure the MPIP clients. The ip addresses do the
work here.
2) The chosen MPIP server has to be its own client, Hence, the
MPIP client list on the MPIP server will also include its own
ip address.
3) Synchronize the clocks on the server and the clients by
pointing them to one NTP time server (and not 2 time servers).
4) The times on the 2 Total Control NETServer PRI's can be
synchronized by either rebooting them at the same time, or by
doing reset time.
5) Once the environment is figured out, then make sure that
you can test each NETServer for mlppp connections.
6) Once step 5 is preformed, then start up your DUN (dial-up
networking) and upon configuring the basic parameters, the
number dialing process should be followed as below.
a. if the numbers are different, then dial
8475882300&8475882301
b. if there is only one number, then dial that number
after making sure that one channel will land on one chassis
and the next will land on the other chassis.
7) Follow the above steps 2,3,&4 on the command syntax for
setting up MPIP server and client.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Jason Watkins jwatkins@iland.net
I-Land NOC Tech http://www.iland.net
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
-----Original Message-----
:Thus spake Taino d Johnston
:>You say I need to enable a MPIP server and configure MPIP functionality on
:>all your NETServers. What exactly does that mean? Is a MPIP server
:>something extra that needs to be purchased? What configuration needs to
be
:>done on the NETServers?
:
:MPIP is built into the NETServers, you should be able to check out the
:documentation to find the commands to enable it. One NETServer will
:need to be designated as an MPIP server and requires some special
:configuration. It should all be in the docs...unfortunately, I don't
:have the information in front of me at the moment.
:--
:Jeff McAdams Email: jeffm@iglou.com
:Head Network Administrator Voice: (502) 966-3848
:IgLou Internet Services (800) 436-4456
:
:-
: To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
: with "unsubscribe usr-tc" in the body of the message.
: For information on digests or retrieving files and old messages send
: "help" to the same address. Do not use quotes in your message.
:
Yesterday I had a very curious problem.
I discovered that some ports om my TCS had no filter set up, so I
used the command:
set interface slot:1/mod:[1-30] input_filter dial.in
And I could see that some modems had a wrong value in the input
filter field: @mailbo (don't ask me why)
Then I send the commend again, and send it two times for each
slot, when I was configuring the last slot (#6) the TC hang and rebooted.
I'm using HARC 4.0.29.
Have any one seem some thing like this?
- Marcelo
Subject:(usr-tc) New Hiper ARC service release From: sagarwal@crash.ae.usr.com Date: 1998-11-17 13:50:42
All,
Check out a new Hiper ARC service release code version 4.1.72 at
http://totalservice.usr.com
Thanks
Sanjay Agarwal
On Tuesday, November 17, 1998 12:43 PM, Casey Cook [SMTP:caseyc@gate.net]
wrote:
> 20 21 22 23 0 0 0 0 0 0 0 0 0 0 2e 2f
> 64 bytes from 207.36.0.99: icmp_seq=19 ttl=244 time=55 ms
>
>
> Any thoughts on those 'wrong data byte' lines?
>
I've never seen this with a TC box but I have seen it on an NT box. It
turned out to be a bad PCI bus. My guess is that you have a bad card or
something else...
Be Seeing You...
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Matthew Stainforth<>Technical Services Manager<>BrunNet Inc.<>(506)450-4562
A quart of wheat for a day's wages, and three quarts of barley for a day's
wages, and do not damage the oil and the wine!
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Subject:RE: (usr-tc) New Hiper ARC service release From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-17 15:20:36
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Mike Andrews
|Sent: Tuesday, November 17, 1998 2:52 PM
|To: usr-tc@lists.xmission.com
|Subject: Re: (usr-tc) New Hiper ARC service release
|
|
|Is there a list of what's changed between 4.1.78 and 4.1.72?
|
All changes are noted in the PDF's on totalservice. The list shows the delta from
4.1.11
All fixes in 4.1.78 are in there.
-M
"Brian K McIntire" <bmcintire@commnet.com> writes:
> is related to some setting your chassis. I've seen this a thousand times on
> a thousand different AOL chassis.
Just for a minor counterpoint, on most of our chassis using
channelized T1 spans, we don't really see all that much of a
difference. In fact, since telcos often rob bits even within short
call distances, the fact that the channelized T1 implies a robbed bit
doesn't necessarily hurt all that much - often the PRIs don't even
bring an advantage to the 56K protocols like x2 or V.90. (For
example, calling to a local T1 in my office gets me the same
connections as calling a PRI located only about 10 miles west of me,
and both connected to the same telco).
Of course, it is important how you configure your T1s - specifically
the signaling. If you opt for ground or loop start (we've only ever
used ground), you're much more likely to have issues and I'm more
likely to agree with your point, but not because of the digital T1
span itself. It's because in virtually all cases, the telco is going
to provision that at the concentrator end with a series of analog
copper pair circuits feeding from their switch into a channel bank
that handles the T1 framing. In such a configuration you really have
an extra analog hop, individual pairs that can go bad, etc, and what
you're hurt by is the central office configuration.
But an E&M configuration is almost always the opposite - digital
straight into the switch - and in my experience behaves just as nicely
as a PRI circuit and very closely even to the call rates achieved.
> PRI's, by design, are superior to T1's
Pedantic point - unless you are at an international site using E1s, a
"PRI" _is_ a "T1" fundamentally. Same framing. It's the signaling that
is different, so you need to qualify the T1 as, for example,
channelized with E&M signaling.
> and have significantly less things that could go wrong (assuming your
> provisioned has any notion of what he is doing) It was probably your T1.
Heh - try to tell that to someone trying to troubleshoot D channel
signaling or getting the non-standard "service message" support to
work everywhere. PRIs aren't automatically better in all
circumstances ;-)
Don't get me wrong - I love PRIs as much as the next guy, and we're
certainly using them as the standard configuration for all new
equipment, but it wasn't all that long ago when they were too
expensive in various regions and we still have a whole lot of
channelized T1 (far preferring E&M but we've still got the occasional
ground start - economic reasons again) working in the network, and in
general behaving just as nicely as the PRI circuits.
So I wouldn't automatically assume that a channelized T1 configuration
has to be inferior. It can be configured wrong (but so can a PRI) and
it can have different types of failures than a PRI, but it can also
work just as well.
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
Subject:Re: (usr-tc) New Hiper ARC service release From: Mike Andrews <mandrews@termfrost.org> Date: 1998-11-17 15:52:22
Is there a list of what's changed between 4.1.78 and 4.1.72?
Mike Andrews (MA12) icq 6602506 -------------- mandrews@termfrost.org
VP 'n' Systems/Network Administrator --------------- mandrews@dcr.net
Digital Crescent, Frankfort, KY ----------- http://www.termfrost.org/
# view;touch;unzip;finger;mount;mv;mv;mv;yes;mv;yes;mv;yes;umount;sleep
On Tue, 17 Nov 1998 sagarwal@crash.ae.usr.com wrote:
> All,
>
> Check out a new Hiper ARC service release code version 4.1.72 at
>
> http://totalservice.usr.com
>
> Thanks
>
> Sanjay Agarwal
Subject:RE: (usr-tc) New Hiper ARC service release From: Mike Andrews <mandrews@termfrost.org> Date: 1998-11-17 16:22:53
Duh, sorry... hand me the big pointy dunce hat for today. Thanks...
Mike Andrews (MA12) icq 6602506 -------------- mandrews@termfrost.org
VP 'n' Systems/Network Administrator --------------- mandrews@dcr.net
Digital Crescent, Frankfort, KY ----------- http://www.termfrost.org/
# view;touch;unzip;finger;mount;mv;mv;mv;yes;mv;yes;mv;yes;umount;sleep
On Tue, 17 Nov 1998, Mike Wronski wrote:
>
>
> |-----Original Message-----
> |From: owner-usr-tc@lists.xmission.com
> |[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Mike Andrews
> |Sent: Tuesday, November 17, 1998 2:52 PM
> |To: usr-tc@lists.xmission.com
> |Subject: Re: (usr-tc) New Hiper ARC service release
> |
> |
> |Is there a list of what's changed between 4.1.78 and 4.1.72?
> |
>
> All changes are noted in the PDF's on totalservice. The list shows the delta from
> 4.1.11
> All fixes in 4.1.78 are in there.
>
> -M
>
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) RADIUS 6.0.8: Authentication into NT Domain creates errors From: Ralph Helfenberger <rhelfenberger@comlight.ch> Date: 1998-11-17 18:52:54
Hi all
I try to setup 3COM RADIUS Server to authenticate all users, wich are
not found in the RADIUS database, in the NT Domain. For that reason I'm
working with the template NTUSER. The problem I encounter: There are
script errors in the logfile. Has anyone done something similar?
RADIUS 6.0.8
Script Section: CHECK-PASSWORD nLine: 534 bIgnore: 0
Script Section: VALIDATE-LOCALPASSWD nLine: 3345 bIgnore: 0
Trace: Attempting User: Tinu-H, Domain: LEO-DOMAIN [11 17 18:44:40]
Script Section: CHECK-EXPIRATION nLine: 535 bIgnore: 0
Script Section: CHECK-CALL-SOURCE nLine: 542 bIgnore: 0
Script Section: CHECK-TUNNEL-NAME nLine: 547 bIgnore: 0
Script error on line 1781 :Variable (THISGROUP) not found
Line is : if(length(thisGroup)>0)
Script error on line 1781 :Parameter has a bad value
Line is : if(length(thisGroup)>0)
Script Section: COLLECT-USER-RESPONSE nLine: 549 bIgnore: 0
Script Section: GET-STANDARD-ATTRIBUTES nLine: 671 bIgnore: 0
Script Section: GET-USER-SERVICE-TYPE nLine: 7257 bIgnore: 0
Script Section: GET-STANDARD-FRAMED-ATTRIBUTES nLine: 7276 bIgnore: 0
Script Section: GET-FRAMED-PROTOCOL nLine: 7319 bIgnore: 0
Script error on line 2696 :Variable (THISGROUP) not found
Line is : if(length(thisGroup)>0)
Script error on line 2696 :Parameter has a bad value
Line is : if(length(thisGroup)>0)
Script Section: GET-FRAMED-ADDRESS nLine: 7320 bIgnore: 0
Script error on line 2753 :Variable (THISGROUP) not found
Line is : if(length(thisGroup)>0)
Script error on line 2753 :Parameter has a bad value
Line is : if(length(thisGroup)>0)
Script error on line 2769 :Variable (THISGROUP) not found
Line is : if(length(thisGroup)>0)
Script error on line 2769 :Parameter has a bad value
Line is : if(length(thisGroup)>0)
Script Section: GET-FRAMED-NETMASK nLine: 7321 bIgnore: 0
Script error on line 2785 :Variable (THISGROUP) not found
Line is : if(length(thisGroup)>0)
and so on....
Subject:Re: (usr-tc) New Hiper ARC service release From: Ricky Beam <jfbeam@interpath.net> Date: 1998-11-17 19:11:35
sagarwal@crash.ae.usr.com was heard to say:
>Check out a new Hiper ARC service release code version 4.1.72 at
>
>http://totalservice.usr.com
I love the version number... ne041727.zip :-) I also like the way I don't
need a service contract to grab it (or do I?)
--Ricky
Subject:(usr-tc) Will Trade ISP Billing Software for Dual T1 PRI card, etc. From: Dwight G. Jones <djones@imagen.net> Date: 1998-11-17 19:21:44
We are developers of strong ISP billing software, and we need 2 Dual T1 PRI
cards as well as V.90 code for some used USR-TC racks that we use to keep
in touch with the family..
:^D
djones@imagen.net
www.imagen.net
I just noticed the small fan at the rear of my chassis stopped spinning.
Does anyone know if these are standard sized fans that can be readily
found, or if replacement of it requires a custom (costly) part swap from
3com.. pointers appreciated.
-L.
On Tue, 17 Nov 1998, Tatai SV Krishnan wrote:
|Open a ticket and send the crash dump ( sho board crash ) to a tech.
|They can analyze the crash and tell you what the problem is.
I don't have a support contract with 3com.
- Marcelo
|
|krish
|
|-----------------------------------------
| \ T.S.V. Krishnan \
| \ Network System Engineer \ ( : - : )
| \ 3Com ............ \
| ----------------------------------------------/
|tkrishna@bubba.ae.usr.com
|----------------------------/ http://interproc.ae.usr.com ----/
|The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
|-------------------------------------------------------------------------\
| Any Sufficiently advanced bug is indistinguishable for a feature.
| - Rick Kulawiec
|-------------------------------------------------------------------------/
|
|On Tue, 17 Nov 1998, Marcelo Souza wrote:
|
|>
|> Yesterday I had a very curious problem.
|> I discovered that some ports om my TCS had no filter set up, so I
|> used the command:
|>
|> set interface slot:1/mod:[1-30] input_filter dial.in
|>
|> And I could see that some modems had a wrong value in the input
|> filter field: @mailbo (don't ask me why)
|> Then I send the commend again, and send it two times for each
|> slot, when I was configuring the last slot (#6) the TC hang and rebooted.
|>
|> I'm using HARC 4.0.29.
|>
|> Have any one seem some thing like this?
|>
|> - Marcelo
|>
|>
|> -
|> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
|> with "unsubscribe usr-tc" in the body of the message.
|> For information on digests or retrieving files and old messages send
|> "help" to the same address. Do not use quotes in your message.
|>
|
- Marcelo
Is there a toggle to have the ppp start message display on the hiperarc?
'show ppp' lists the following, running on 4.1.11 and 4.1.72
DIAL_IN Users Authenticate: ANY
PPP Authentication Preference: PAP
PPP session start message: PPP session from %server_ip to %client beginning...
but when i log in manually with a terminal over dialup and log in for ppp,
it doesnt display it at all. also, does this default start message match
the netserver one.. i'd like to keep it the same.
-L
Subject:Re: (usr-tc) ppp start message From: Carl Litt <carl@snel.execulink.com> Date: 1998-11-17 20:52:37
enable authen hint_assigned
On Tue, 17 Nov 1998, Laszlo Vecsey wrote:
> Is there a toggle to have the ppp start message display on the hiperarc?
> 'show ppp' lists the following, running on 4.1.11 and 4.1.72
>
> DIAL_IN Users Authenticate: ANY
> PPP Authentication Preference: PAP
>
> PPP session start message: PPP session from %server_ip to %client beginning...
>
> but when i log in manually with a terminal over dialup and log in for ppp,
> it doesnt display it at all. also, does this default start message match
> the netserver one.. i'd like to keep it the same.
>
> -L
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
I have a Netserver Card hardware version 4.0.0 running software 3.3.3. Is
it possible to upgrade the software? It tries to load le*.nac software and
3.3.3 is the latest 3Com has on their site. It has a really annoying habit
of dropping all the connections and not telling my radius server.
Thanks,
Ted
Subject:Re: (usr-tc) Stuck in ringRcvd From: Brian <signal@shreve.net> Date: 1998-11-17 22:26:50
On Tue, 17 Nov 1998 sagarwal@crash.ae.usr.com wrote:
>
> What does "List Interfaces " command on Hiper ARC show. Does it show that
> the int is down in oper status?
>
It shows its "UP UP"
I just had this happen again...........
> Sanjay
>
>
> On Tue, 17 Nov 1998, Brian wrote:
>
> > We had a modem that when people dialed our number, it would just be dead
> > silence. I narrowed it down to a card, and did a Program Settings on it,
> > and saw that the operational status for one of the modems was stuck in
> > ringRcvd. I tried to software reset this modem, and it stayed in
> > ringRcvd. The only fix was to hardware reset the card.
> >
> > This happens quite a bit.
> >
> > Has anyone else seen this? I am running Release hdm/arc code
> > (1.2.5/4.1.11).
> >
> > What is the cause? The fix?
> >
> > brian
> >
> >
> > --------------------------------------------------------------------------
> > Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> > Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> > signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> > (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:Re: (usr-tc) Stuck in ringRcvd From: Brian <signal@shreve.net> Date: 1998-11-17 22:26:50
On Tue, 17 Nov 1998 sagarwal@crash.ae.usr.com wrote:
>
> What does "List Interfaces " command on Hiper ARC show. Does it show that
> the int is down in oper status?
>
It shows its "UP UP"
I just had this happen again...........
> Sanjay
>
>
> On Tue, 17 Nov 1998, Brian wrote:
>
> > We had a modem that when people dialed our number, it would just be dead
> > silence. I narrowed it down to a card, and did a Program Settings on it,
> > and saw that the operational status for one of the modems was stuck in
> > ringRcvd. I tried to software reset this modem, and it stayed in
> > ringRcvd. The only fix was to hardware reset the card.
> >
> > This happens quite a bit.
> >
> > Has anyone else seen this? I am running Release hdm/arc code
> > (1.2.5/4.1.11).
> >
> > What is the cause? The fix?
> >
> > brian
> >
> >
> > --------------------------------------------------------------------------
> > Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> > Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> > signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> > (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:Re: (usr-tc) Stuck in ringRcvd From: Charles Sprickman <spork@inch.com> Date: 1998-11-18 00:09:48
On Tue, 17 Nov 1998, Brian wrote:
> > What does "List Interfaces " command on Hiper ARC show. Does it show that
> > the int is down in oper status?
>
> It shows its "UP UP"
In a similar vein...
Besides 'list int' and 'list con', what else can give me more info on how
hosed a port may be? 'show all' on the netserver is handy for finding
renegade modems, what's the closest equivalent in HARC-land?
Thanks,
Charles
>
> I just had this happen again...........
>
>
> > Sanjay
> >
> >
> > On Tue, 17 Nov 1998, Brian wrote:
> >
> > > We had a modem that when people dialed our number, it would just be dead
> > > silence. I narrowed it down to a card, and did a Program Settings on it,
> > > and saw that the operational status for one of the modems was stuck in
> > > ringRcvd. I tried to software reset this modem, and it stayed in
> > > ringRcvd. The only fix was to hardware reset the card.
> > >
> > > This happens quite a bit.
> > >
> > > Has anyone else seen this? I am running Release hdm/arc code
> > > (1.2.5/4.1.11).
> > >
> > > What is the cause? The fix?
> > >
> > > brian
> > >
> > >
> > > --------------------------------------------------------------------------
> > > Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> > > Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> > > signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> > > (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
> > >
> > >
> > > -
> > > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > > with "unsubscribe usr-tc" in the body of the message.
> > > For information on digests or retrieving files and old messages send
> > > "help" to the same address. Do not use quotes in your message.
> > >
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
> --------------------------------------------------------------------------
> Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
=-----------------= =
| Charles Sprickman Internet Channel |
| INCH System Administration Team (212)243-5200 |
| spork@inch.com access@inch.com |
= =----------------=
Subject:Re: (usr-tc) Stuck in ringRcvd From: Richard Lorbieski <richard@alpha1.net> Date: 1998-11-18 02:48:42
If you have syslog enabled, does it say something like this?
Nov 13 14:14:47 bry409 At 14:13:50, Facility "GWC Modem Driver", Level
"CRITICAL":: GWCMDMDRV FSM illegal event interface slot:2/mod:18, state
Disconnecting, event CallArrivedReq
Charles Sprickman wrote:
>
> On Tue, 17 Nov 1998, Brian wrote:
>
> > > What does "List Interfaces " command on Hiper ARC show. Does it show that
> > > the int is down in oper status?
> >
> > It shows its "UP UP"
>
> In a similar vein...
>
Subject:Re: (usr-tc) New Hiper ARC service release From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-18 08:01:24
Thus spake Russ Miescke
>Is this a newer code than the ER 4.1.78? Does 4.1.78 contain all of the
>fixes from 4.1.72? I guess I am hesitant to change something that is sort
>of working. Anybody load this yet?
Nope, just the opposite, 4.1.72 has all the fixes from 4.1.78. ER code
is number from 99 down, so 72 is newer than 78.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) New Hiper ARC service release From: sagarwal@crash.ae.usr.com Date: 1998-11-18 09:03:43
4.1.72 is newer code than 4.1.78. It has all the fixes from 4.1.78 and
some additional fixes.
Thanks
Sanjay
On Thu, 19 Nov 1998, Russ Miescke wrote:
> Is this a newer code than the ER 4.1.78? Does 4.1.78 contain all of the
> fixes from 4.1.72? I guess I am hesitant to change something that is sort
> of working. Anybody load this yet?
>
>
> Russ Miescke
> Power Web Connect
> russm@powerweb.net
> http://www.powerweb.net
>
> ----- Original Message -----
> From: Ricky Beam <jfbeam@Interpath.net>
> To: <usr-tc@lists.xmission.com>
> Sent: Tuesday, November 17, 1998 6:11 PM
> Subject: Re: (usr-tc) New Hiper ARC service release
>
>
> sagarwal@crash.ae.usr.com was heard to say:
> >Check out a new Hiper ARC service release code version 4.1.72 at
> >
> >http://totalservice.usr.com
>
> I love the version number... ne041727.zip :-) I also like the way I don't
> need a service contract to grab it (or do I?)
>
> --Ricky
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) New Hiper ARC service release From: Dale Hege <fhege@sover.net> Date: 1998-11-18 10:50:31
How is the ER for the DSP comming? Does anyone know?
-Dale
On Wed, 18 Nov 1998 sagarwal@crash.ae.usr.com wrote:
> Date: Wed, 18 Nov 1998 09:03:43 -0600 (CST)
> From: sagarwal@crash.ae.usr.com
> Reply-To: usr-tc@lists.xmission.com
> To: Russ Miescke <russm@powerweb.net>
> Cc: usr-tc@lists.xmission.com
> Subject: Re: (usr-tc) New Hiper ARC service release
>
>
> 4.1.72 is newer code than 4.1.78. It has all the fixes from 4.1.78 and
> some additional fixes.
>
> Thanks
>
> Sanjay
>
> On Thu, 19 Nov 1998, Russ Miescke wrote:
>
> > Is this a newer code than the ER 4.1.78? Does 4.1.78 contain all of the
> > fixes from 4.1.72? I guess I am hesitant to change something that is sort
> > of working. Anybody load this yet?
> >
> >
> > Russ Miescke
> > Power Web Connect
> > russm@powerweb.net
> > http://www.powerweb.net
> >
> > ----- Original Message -----
> > From: Ricky Beam <jfbeam@Interpath.net>
> > To: <usr-tc@lists.xmission.com>
> > Sent: Tuesday, November 17, 1998 6:11 PM
> > Subject: Re: (usr-tc) New Hiper ARC service release
> >
> >
> > sagarwal@crash.ae.usr.com was heard to say:
> > >Check out a new Hiper ARC service release code version 4.1.72 at
> > >
> > >http://totalservice.usr.com
> >
> > I love the version number... ne041727.zip :-) I also like the way I don't
> > need a service contract to grab it (or do I?)
> >
> > --Ricky
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) upgrade for Netserver card available? From: Pete Ashdown <pashdown@xmission.com> Date: 1998-11-18 11:56:54
Theodore Cekan said once upon a time:
>
>I have a Netserver Card hardware version 4.0.0 running software 3.3.3. Is
>it possible to upgrade the software? It tries to load le*.nac software and
>3.3.3 is the latest 3Com has on their site. It has a really annoying habit
>of dropping all the connections and not telling my radius server.
You might need a memory upgrade. Check to see if your card has 16 megs or
more, if not, you will need to replace the 4 meg SIMM inside with a 16 meg
one.
Laszlo Vecsey said once upon a time:
> PPP session start message: PPP session from %server_ip to %client beginning...
>
>but when i log in manually with a terminal over dialup and log in for ppp,
>it doesnt display it at all. also, does this default start message match
>the netserver one.. i'd like to keep it the same.
For the most part, except that %server_ip on the Netserver is bracketted by
parenthesis. As stated in your other message Laszlo, I haven't been able
to get the parenthesis to go in either, so I just bagged them.
Subject:Re: (usr-tc) Confirmation From: Pete Ashdown <pashdown@xmission.com> Date: 1998-11-18 12:06:32
Mike Andrews said once upon a time:
>This seems to be broken on current ARC releases up to and including at
>least 4.1.78; you have to also use "Max_Channels = 1" (which is a VSA) to
>do the right thing there. On the NETserver, it's fine. Go back through
>the archives a month or two and you can see me whining away about this :)
>We actually have both Port-Limit AND Max_Channels in the Radius users
>file. That seems to take care of things.
According to the fix documentation, 4.1.72 follows the RFC on this. Does
anyone know if that returns things to normal? ;-)
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Pete Ashdown <pashdown@xmission.com> Date: 1998-11-18 12:11:28
Mike Wronski said once upon a time:
>It could have been.. But I havent seen it.. The only time default user settings
>would not apply is if there was a conflicting RADIUS attribute. Then RADIUS wins
>the tie. For example. Default user has idle time set to 600 and the Radius
>attribute "Idle-Timeout=1200" is in the access accept. The idle timeout will be
>1200 for that user.
But Mike, what good is a time-limit when just about anything can reset it?
ICQ, RIP, ping, all manner of useless broadcasts, in fact any traffic at
all, no matter how mundane resets the timeout. We need to be able to
configure what counts as activity on the ARC.
Once upon a time Marcelo Souza shaped the electrons to say...
> How can I set ARC to log the packets dropped from a filter?
I'm not sure if they kept this the same from the NS - but if so you just
add the word 'log' on the end of the filter rule and any packet that hits
that rule causes a syslog entry.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Once upon a time GTI x2 Tech shaped the electrons to say...
> We have several TC's, and are looking to implement a way to
> keep users from connecting more than once at a time (using RADIUS).
You need a RADIUS server that has this intelligence built in. I highly
recommend Cistron RADIUS for this - it works very well.
Others include Lucent RADIUS ABM, Radiator, and IEA's RadiusNT.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Thanks Pete,
It has 4 meg. Will an upgrade allow me to install later software or will it
just help performance?
Ted
-----Original Message-----
>Theodore Cekan said once upon a time:
>>
>>I have a Netserver Card hardware version 4.0.0 running software 3.3.3. Is
>>it possible to upgrade the software? It tries to load le*.nac software
and
>>3.3.3 is the latest 3Com has on their site. It has a really annoying
habit
>>of dropping all the connections and not telling my radius server.
>
>You might need a memory upgrade. Check to see if your card has 16 megs or
>more, if not, you will need to replace the 4 meg SIMM inside with a 16 meg
>one.
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Pete Ashdown <pashdown@xmission.com> Date: 1998-11-18 13:05:08
Mike Wronski said once upon a time:
>I think the configurable idle timeout filter sounds like a good idea until you
>really look at its implementation and wheather it will even do what you want. It
>would work for a short time and then the users will figure out a way around it. I
>think there are many other features that time & resources could be spent
>developing that would benefit the ISP more than this.
I'm not talking about the savvy abuser though. I'm talking about my DAD
who leaves his connection open by accident just because he's running ICQ.
We go after the savvy abusers by looking at total time used. However, I
would still say that we have a large problem with non-technical users
accidentally leaving their computers connected.
I don't know the internals of your code Mike, but wouldn't it be as simple
as a boolean operation on existing filter code with a specified name? I
think if 3com did a poll (the horror!) of ISPs feature demands, this would
rate rather highly. Its #1 with me.
Subject:Re: (usr-tc) upgrade for Netserver card available? From: Pete Ashdown <pashdown@xmission.com> Date: 1998-11-18 13:05:52
Theodore Cekan said once upon a time:
>
>Thanks Pete,
>
>It has 4 meg. Will an upgrade allow me to install later software or will it
>just help performance?
Both, if you install the later software. 3.3.3 will not take advantage of
the extra memory.
Subject:Re: (usr-tc) HELP: need to route IP addresses to a customer over TC From: MegaZone <megazone@megazone.org> Date: 1998-11-18 13:06:01
Once upon a time C Thompson shaped the electrons to say...
>We have not had to do this before, but I need to route a Class C address
>across an ISDN connection on a USR/3com Total Control hub to one of
>our customers. Also, for future reference, it would be helpful to know
>what to do for a subnet as well.
>
>Can anyone provide helpful pointers on what I need to do
>1) in RADIUS
See below.
>2) in the TC Hub (if anything)
I'm not sure what needs to be done to get it to recognize the netmask as
sent by RADIUS. For example a PM has 'set user-netmask on' for this.
>3) in my router
Well, if you are running an active routing protocol that can handle this
then it should propigate ok.
Now, as to radius. I speak primarily Lucent RADIUS or Cistron RDIUS, so
translate as need be.
eris Auth-Type = System
Service-Type = Framed-User,
Framed-Protocol = PPP
Framed-Routing = None,
Framed-MTU = 1500,
Framed-Compression = Van-Jacobson-TCP-IP,
Framed-IP-Address = 192.168.1.1,
Framed-IP-Netmask = 255.255.255.0
With a NAS that recognizes netmasks, that should route the user the
192.168.1.0/24 network, with their end of the line being the .1 address.
Now, if your NAS doesn't listen to RADIUS netmasks (unfortunate) you can
try this kludge.
eris Auth-Type = System
Service-Type = Framed-User,
Framed-Protocol = PPP
Framed-Routing = None,
Framed-MTU = 1500,
Framed-Compression = Van-Jacobson-TCP-IP,
Framed-IP-Address = 192.168.1.1,
Framed-IP-Netmask = 255.255.255.0
Framed-Route = "192.168.1.0/24 192.168.1.1 1"
That should 'manually' install the network route for you.
For a subnet, just change the netmask(s) in the entry.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Subject:Re: (usr-tc) RADIUS & user limitations From: Dan Hollis <goemon@sasami.anime.net> Date: 1998-11-18 13:07:10
On Wed, 18 Nov 1998, MegaZone wrote:
> Once upon a time GTI x2 Tech shaped the electrons to say...
> > We have several TC's, and are looking to implement a way to
> > keep users from connecting more than once at a time (using RADIUS).
> You need a RADIUS server that has this intelligence built in. I highly
> recommend Cistron RADIUS for this - it works very well.
> Others include Lucent RADIUS ABM, Radiator, and IEA's RadiusNT.
I heard Livingston wasnt too keen on the free radius server clones (eg
merit, etc) which is why they made the Radius2.0 license so restrictive.
How does Lucent view cistron, especially since its based off old
Livingston code?
-Dan
Subject:(usr-tc) I'm back (administrative message) From: Pete Ashdown <pashdown@xmission.com> Date: 1998-11-18 13:07:20
I'm back from a month-long honeymoon. This is why you are seeing an influx
of old messages that required approval to the list. I'll also be cleaning
up any bad addresses and adding all the new subscription requests that
weren't automatically processed.
Ok, just a follow up on the problem I was having and the solution, for
those that are interested.
There are certain mac dialers (configppp in particular i think) that check
for the existence of "beginning...." in the script. Now by default the
hiperarc puts out a ppp session start string that includes that text, but
apparently the PPP protocol (garbled) binary text that follows includes a
backspace character of some sort, perhaps someone familiar with the
protocol could confirm this, so that what the client ends up seeing is one
less dot.
The solution was to add a carriage return to the start string
set ppp session_start "PPP session from %server_ip to %client_ip beginning....\n"
I wonder, does the Netserver send a carriage return, and exactly what does
its PPP string look like anyway? I'd like to match up the hiperarc string
exactly the way my former netserver was running, less there be other
quirks I overlooked..
Now sometime back on the list there was some discussion of putting
paranthesis around the server_ip and client_ip, and quirks with respect to
that. Maybe I didnt archive all the messages or deleted them by accident,
but I found that you need a space character after either %server_ip or
%client_ip for it to match and be parsed out, you cant follow it directly
with a ')' character. Also when it does get parsed out, theres a space
inserted before the IP automatically.. so I dont see a way to get the
paranthesis to line up right to the IP's. Not that I even need this, but
just curious.
On Tue, 17 Nov 1998, Carl Litt wrote:
>
> enable authen hint_assigned
>
> On Tue, 17 Nov 1998, Laszlo Vecsey wrote:
>
> > Is there a toggle to have the ppp start message display on the hiperarc?
> > 'show ppp' lists the following, running on 4.1.11 and 4.1.72
> >
> > DIAL_IN Users Authenticate: ANY
> > PPP Authentication Preference: PAP
> >
> > PPP session start message: PPP session from %server_ip to %client beginning...
> >
> > but when i log in manually with a terminal over dialup and log in for ppp,
> > it doesnt display it at all. also, does this default start message match
> > the netserver one.. i'd like to keep it the same.
> >
> > -L
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) Confirmation From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-18 13:24:46
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Pete Ashdown
|Sent: Wednesday, November 18, 1998 1:07 PM
|To: usr-tc@lists.xmission.com
|Subject: Re: (usr-tc) Confirmation
|
|
|Mike Andrews said once upon a time:
|
|>This seems to be broken on current ARC releases up to and including at
|>least 4.1.78; you have to also use "Max_Channels = 1" (which is a VSA) to
|>do the right thing there. On the NETserver, it's fine. Go back through
|>the archives a month or two and you can see me whining away about this :)
|>We actually have both Port-Limit AND Max_Channels in the Radius users
|>file. That seems to take care of things.
|
|According to the fix documentation, 4.1.72 follows the RFC on this. Does
|anyone know if that returns things to normal? ;-)
Yes.. Port-Limit & Max_Channels now function the same. They only apply to MLPPP
bundles and not to concurrent session limiting.. HARC & NS work the same now. You
can remove your double entries or workarounds.
-M
Subject:RE: (usr-tc) HiperARC.. [resend, first one never hit the list?] From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-18 13:39:59
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of System
|Administrator
|Sent: Wednesday, November 11, 1998 1:17 PM
|To: usr-tc@lists.xmission.com
|Subject: (usr-tc) HiperARC.. [resend, first one never hit the list?]
|
|
|Config'ing first HiperARC (btw, must say that CLI is MUCH better than
|netserver), couple of questions:
|
|1. Is there still a known problem with chassis awareness (w/ latest non-ER
|code). Only one HARC, so should I do chassis slot assignments manually?
This is not an issue with one HARC. It works fine automatically.
|2. Am I to understand that the Idle-Timeout RADIUS attribute does not work
|correctly with the HARC, and that I must use a VA (we have a highly customized
|radius daemon, which supports loads of goodies, but not VA)?
No.. Idle-Timeout works fine and there is no VSA for it anyway.
NOTE: These statements reflect the latest code available 4.1.72-7. I don't wish
to list does/does not in previous versions.
-M
Subject:RE: (usr-tc) Solution for idle time limits and dual logins From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-18 13:40:00
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Pete Ashdown
|Sent: Wednesday, November 18, 1998 1:11 PM
|To: usr-tc@lists.xmission.com
|Subject: Re: (usr-tc) Solution for idle time limits and dual logins
|
|
|Mike Wronski said once upon a time:
|
|>It could have been.. But I havent seen it.. The only time default user settings
|>would not apply is if there was a conflicting RADIUS attribute. Then
|RADIUS wins
|>the tie. For example. Default user has idle time set to 600 and the Radius
|>attribute "Idle-Timeout=1200" is in the access accept. The idle timeout will be
|>1200 for that user.
|
|But Mike, what good is a time-limit when just about anything can reset it?
|ICQ, RIP, ping, all manner of useless broadcasts, in fact any traffic at
|all, no matter how mundane resets the timeout. We need to be able to
|configure what counts as activity on the ARC.
|
I don't have a good answer for that. Idle-timeouts are only usefull against the
non-savy user and those are getting fewer by the day.
So what do you count as activity that should reset the idle timer?
No matter what you define as idle, someone will come up with a way to use data
you don't check to keep the channel open.. I can think a simple way to use HTTP
gets to keep the line open.. I think you are fighting a battle that cant be won..
The user who wants to, will nail their connection up.. A resonable session limit
may be your only hope.
I think the configurable idle timeout filter sounds like a good idea until you
really look at its implementation and wheather it will even do what you want. It
would work for a short time and then the users will figure out a way around it. I
think there are many other features that time & resources could be spent
developing that would benefit the ISP more than this.
-M
Subject:RE: (usr-tc) CONNECTION PROBLEM From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-18 13:45:06
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jeff Binkley
|Sent: Tuesday, November 10, 1998 7:56 AM
|To: USR-TC@lists.xmission.com
|Subject: (usr-tc) CONNECTION PROBLEM
|
|
|
|
|We have a customer who has Brother NB-80C Geobook trying to connect with us
|on a HiperArc running 4.1.78 code. The call continually fails in the LCP
|negotiation stage and we never see the userid and password get sent. The
|Geobook is a replacement because he had so many problems with the old one.
|With the old one he was able to connect with us but since then we have done
|a couple of HiPerArc release upgrades. Below is a copy of a MONITOR PPP
|trace from the HiPerArc. I am concerned this may be a HiPerArc problem
|but I want to be sure. We have sent the trace on to Brother too. They do
|include a diskette from Earthlink and it does work with them. Does anyone
|have a document for decoding these traces and figuring out what is going
|on ?
|
|We spoke to Brother and they were no help. Basically mumbling things like if
|you are trying to connect to an NT server it won't work and we only support
|dialing into Unix machines. They said send them to Earthlink. I don't think
|we want to start eliminating groups of users from our collective ISps based
|upon the equipment the end user has. That doesn't seem to be good business
|practice.
|
This looks like a problem we had with ASYNC_MAP of all 0's.. This is resolved in
the 4.1.72-7 code.
-M
Subject:Re: (usr-tc) PRI vs T1 ? From: MegaZone <megazone@megazone.org> Date: 1998-11-18 13:46:01
Once upon a time Robb Bryn shaped the electrons to say...
> are now pricing PRI vs T1. In our area T1's are priced about
> $600/month cheaper than a PRI. Aside from PRI's being easier to
> setup, I've caught rumors that T1's are not as fast as PRI (when
> considering a v.90 call & all other factors are equal). I have only
> one question... are these rumors true?
Yes - and no. Some users have shown that CT1 lines *for them* get about a
notch lower connect speeds than PRI. A good many users have reported this
with 3Com, Lucent, Ascend, etc. However, not ALL users see this - and there
hae been a couple of cases where users reported their CT1 lines were giving
better connections.
But based on what I've seen you do have a good chance of seeing *slightly*
lower speeds on a CT1.
There is also the fact you can't do ISDN DATA over a CT1, only ISDN DOSBS.
And I don't think the TC does DOSBS yet, does it? (If so, can it do ISDN
DOSBS *and* modems over the same CT1 at the same time?)
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Quad modem chassis running 4.2.1 on T1 card (running channelized T1's)
modem cards running 5.10.9, HiPer ARC 4.0.29, NMC running 5.5.5
When doing a DS0 performance monitor, the DS0 channels show idle, but the
modem connected to the DS0 status is "corresponding modem is unavailable."
I've tried restoring the chassis from defaults, and resetting everything,
to no avail. Has anyone seen this before??
| Curtis V. Shambeau | curt@execpc.com | http://www.execpc.com/~curt |
| Executive Vice President - Exec-PC, Inc. |
Subject:RE: (usr-tc) Preventing the answering of a 128K ISDN call From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-18 14:05:23
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Eric J. Lorenzo
|Sent: Friday, November 06, 1998 1:05 PM
|To: TC List
|Subject: (usr-tc) Preventing the answering of a 128K ISDN call
|
|
|How do you block in both a NETServer and a HARC the answering of a 128K
|ISDN calls? We don't want two modems being taken up by one user.
|
Set Port-Limit=1 in your RADIUS default user. Or without RADIUS:
Netserver: set mp off
HARC: set net user default ppp max_chan 1 (4.1.x)
set net user default max_chan 1 (4.0.x)
-M
Once upon a time Dan Hollis shaped the electrons to say...
>I heard Livingston wasnt too keen on the free radius server clones (eg
>merit, etc) which is why they made the Radius2.0 license so restrictive.
Not quite. Most of the issue was other vendors selling HW without a RADIUS
server - and telling customers to just go download it from Livingston. This
happened a LOT. Then they'd get calls from people wanting Livingston to
support their free RADIUS with a Computone, etc.
Since RADIUS was well into the RFC process, there were a number of free
servers out, a number of vendors had started produciing their own, AND
there were a growing number of commercial servers - when they produced 2.0
they decided it was time to tighten control a bit. They added things like
menuing, SecurID support, Group check - and now proxy, VSAs, ActivCard, etc
to the 2.x revisions. It didn't make sense to keep helping the competition
sell product, or let them produce a RADIUS server after Livingston, now
Lucent RABU, had invested quite a bit of resources in doing the ground work
for the new features.
>How does Lucent view cistron, especially since its based off old
>Livingston code?
Quite favorably actually. You'll see a number of Lucent RABU folks referring
people to Cistron if they need special features, or are looking for a free
server with wider support for a mixed vendor community - for example, the
latest Cistron supports 3Com VSAs.
It still has some Livingston 1.16 code in its guts, but there is an ongoing
effort to switch to libradius to create a truly public sourced RADIUS, which
would likely be GPL'd.
Free servers like Cistron don't detract in any way from the Lucent servers.
Cistron doesn't quite do everything - SecurID, ActivCard, and menuing are
major features not in the free server. And while Cistron, and some
commercial products liek Radiator, do simultaneous login control they
don't have the large scalability, nor the myriad of advanced features, in
a product like RADIUS ABM.
Actually, a free server like Cistron spares Lucent from having to
implement features not widely desired, or that could have scalability
issues. The server they release has to address a very wide array of users,
and it has their name on it. With Cistron you have a number of interesting
features - like fall through profiles, IP pools, etc - which a few users
like.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Pete Ashdown <pashdown@xmission.com> Date: 1998-11-18 14:08:52
Mike Wronski said once upon a time:
>I would propose that using SNMP, the %utilization over a time interval on a
>per-interface basis could be obtained. Then you can drop users if they have less
>than some % use for last time interval.. This is easy with HARC and possible with
>Netserver. Its also not *perfect* but it would work.
That would be very cool. Can we get it implemented into ARC code via
RADIUS? ;-)
Us to... We run this combo with no issues.
-----Original Message-----
>Jamie Orzechowski said once upon a time:
>>
>>Wondering if anyone has had any experience running 10 DSP's and 1 ARC card
>>in one chassis? ... any recommendations? ... any problems?
>>
>>I will be using the lastest ARC beta code ....
>
>I'm running 11. Overall, things seem to work fine. We have a bit of
>funkiness with slot 6 on a couple chassis, which may just be
>telco-coincidence, but other that, everything is great.
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) Solution for idle time limits and dual logins From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-18 14:56:08
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Pete Ashdown
|Sent: Wednesday, November 18, 1998 2:05 PM
|To: usr-tc@lists.xmission.com
|Subject: Re: (usr-tc) Solution for idle time limits and dual logins
|
|
|Mike Wronski said once upon a time:
|
|>I think the configurable idle timeout filter sounds like a good idea until you
|>really look at its implementation and wheather it will even do what
|you want. It
|>would work for a short time and then the users will figure out a way
|around it. I
|>think there are many other features that time & resources could be spent
|>developing that would benefit the ISP more than this.
|
|I'm not talking about the savvy abuser though. I'm talking about my DAD
|who leaves his connection open by accident just because he's running ICQ.
|We go after the savvy abusers by looking at total time used. However, I
|would still say that we have a large problem with non-technical users
|accidentally leaving their computers connected.
|I don't know the internals of your code Mike, but wouldn't it be as simple
|as a boolean operation on existing filter code with a specified name?
Its actually not that simple. The architecture has filters associated with IP
data and idle-timout with PPP data. By the time the filter gets the packet the
timer has already been reset. This implementation would require some
reorganization of the levels and thus not be a simple boolean check.
I would propose that using SNMP, the %utilization over a time interval on a
per-interface basis could be obtained. Then you can drop users if they have less
than some % use for last time interval.. This is easy with HARC and possible with
Netserver. Its also not *perfect* but it would work.
-M
Subject:RE: (usr-tc) upgrade for Netserver card available? From: Wayne Barber <barberw@tidewater.net> Date: 1998-11-18 15:00:21
You also need the Frame Relay NIC behind your Netserver. Without it, the
memory upgrade will cause your Netserver to randomly reboot every few
minutes. We finally bought a used Netserver PRI and FR NIC and have been
troublefree since.
Wayne Barber
Coastal Telco Services
> -----Original Message-----
> From: owner-usr-tc@lists.xmission.com
> [mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Pete Ashdown
> Sent: Wednesday, November 18, 1998 1:57 PM
> To: usr-tc@lists.xmission.com
> Subject: Re: (usr-tc) upgrade for Netserver card available?
>
>
> Theodore Cekan said once upon a time:
> >
> >I have a Netserver Card hardware version 4.0.0 running software
> 3.3.3. Is
> >it possible to upgrade the software? It tries to load le*.nac
> software and
> >3.3.3 is the latest 3Com has on their site. It has a really
> annoying habit
> >of dropping all the connections and not telling my radius server.
>
> You might need a memory upgrade. Check to see if your card has 16 megs or
> more, if not, you will need to replace the 4 meg SIMM inside with a 16 meg
> one.
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) Problems with brackets in session_start_message From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-18 15:01:14
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Rick Payne
|Sent: Monday, October 26, 1998 9:46 AM
|To: usr-tc@lists.xmission.com
|Subject: Re: (usr-tc) Problems with brackets in session_start_message
|
|
|Mike Wronski writes:
|
| > Have you tried escaping the '(' with back slashes??
|
|Yes.
|
| > set slip session_start_message "SL/IP session from \(%server_ip\) to
| > %client_ip beginning..."
|
|I entered it as above. Same thing:
|
|SL/IP session from ( 194.42.231.28 to 255.96.0.0 beginning..
|
|(assuming I don't need to relaod after setting the parameter).
|
|Code is 4.1.78 on the HiperARC.
|
Just place a space between '%server_ip' and the ')'. The result will be
SL/IP session from ( 194.42.231.28 ) to 255.96.0.0 beginning..
-M
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-18 15:05:22
Thus spake Mike Wronski
>I don't have a good answer for that. Idle-timeouts are only usefull
>against the
>non-savy user and those are getting fewer by the day.
>So what do you count as activity that should reset the idle timer?
Ideally, this should be administrator configurable...a filter list would
be perfect.
>No matter what you define as idle, someone will come up with a way to use data
>you don't check to keep the channel open.. I can think a simple way to use HTTP
>gets to keep the line open.. I think you are fighting a battle that cant be
>won..
>The user who wants to, will nail their connection up.. A resonable
>session limit
>may be your only hope.
Oh, so if you can't provide a *perfect* tool, then you won't give us
*any* tools to work with at all. Thanks a lot there!
You've obviously never run as ISP. Using the current filter rule setup
to define what traffic resets the idle timeout, I could define a filter
of about 5 to 10 lines that would eliminate 99.9% of our problems with
people camping on lines. Yes, there will be users that will work around
it, its not a perfect solution, but then again, RADIUS isn't a perfect
solution for authenticating users the way everyone needs either, so why
do you support RADIUS?
>I think the configurable idle timeout filter sounds like a good idea until you
>really look at its implementation and wheather it will even do what
>you want.
I *KNOW* it'll do what I want because I've used a pppd in the past that
supported this sort of thing and it worked like a charm!
>It
>would work for a short time and then the users will figure out a way
>around it. I
>think there are many other features that time & resources could be spent
>developing that would benefit the ISP more than this.
There are plenty of features that you could develop that would benefit
us greatly...let's come up with a list and I'll rank them how I feel
they would be most useful to us. I guarentee you that idle-timer filter
list would be *very* high up there.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Hi,
We have several TC's, and are looking to implement a way to
keep users from connecting more than once at a time (using RADIUS).
If anyone knows of any documentation on doing this or has experience
with it, any references/help would be appreciated.
Thanks.
(x2@gti.net)
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Marcelo Souza
|Sent: Wednesday, November 18, 1998 2:40 PM
|To: usr-tc@lists.xmission.com
|Subject: (usr-tc) Filter settings
|
|
|
| How can I set ARC to log the packets dropped from a filter?
set packet_loggin logging <option> packet_size <0-493>
This will send packet_size bytes of the dropped packet to the syslog.
-M
Subject:RE: (usr-tc) Solution for idle time limits and dual logins From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-18 15:15:57
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jeff Mcadams
|Sent: Wednesday, November 18, 1998 2:05 PM
|To: usr-tc@lists.xmission.com
|Subject: Re: (usr-tc) Solution for idle time limits and dual logins
|>So what do you count as activity that should reset the idle timer?
|
|Ideally, this should be administrator configurable...a filter list would
|be perfect.
|>No matter what you define as idle, someone will come up with a way to use data
|>you don't check to keep the channel open.. I can think a simple way to use HTTP
|>gets to keep the line open.. I think you are fighting a battle that cant be
|>won..
|>The user who wants to, will nail their connection up.. A resonable
|>session limit
|>may be your only hope.
|
|Oh, so if you can't provide a *perfect* tool, then you won't give us
|*any* tools to work with at all. Thanks a lot there!
I don't provide anything but support for the product. People at higher levels
have to wade through all the possible features and some fall out in favor of
others. Its just not possible to make a product everyone is in love with.. I
don't know for certain, but do any other vendors have this feature? I don't think
we are unique here.
|Yes, there will be users that will work around
|it, its not a perfect solution, but then again, RADIUS isn't a perfect
|solution for authenticating users the way everyone needs either, so why
|do you support RADIUS?
Agreed, but what is the alternative to RADIUS? TACACS? Is MS Anything perfect?
nope. What do most people use??
With that logic: I think LINUX is better than MS. Should I not support MS even
though its the defacto standard.
I don't think its fair to compare the justification for a new feature with the
use of the standard. I now expect a jump off the bridge reply from someone.. And
"No I wouldn't do it just because all my friends did! :)"
|There are plenty of features that you could develop that would benefit
|us greatly...let's come up with a list and I'll rank them how I feel
|they would be most useful to us. I guarantee you that idle-timer filter
|list would be *very* high up there.
You should do that. Put don't post it here. It is far more effective when
submitted formally as a Feature Request/CRA and submitted by many people. Like
most things, if enough people ask for it and it makes financial sense to have a
feature it tends to get more consideration. (OSPF flame expected here.. Don't
bother! I know its been a long time coming...But.. its coming...Really....No
fooling...Its not vaporware...)
I don't want to start a rant or flame war on this issue. I think that was is
desired is understood by those on this list.
I do my best to respond to issues on this list and have no desire to fight with
anyone over feature sets..
-M
Jamie Orzechowski said once upon a time:
>
>Wondering if anyone has had any experience running 10 DSP's and 1 ARC card
>in one chassis? ... any recommendations? ... any problems?
>
>I will be using the lastest ARC beta code ....
I'm running 11. Overall, things seem to work fine. We have a bit of
funkiness with slot 6 on a couple chassis, which may just be
telco-coincidence, but other that, everything is great.
Subject:RE: (usr-tc) Solution for idle time limits and dual logins From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-18 15:24:11
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Pete Ashdown
|Sent: Wednesday, November 18, 1998 3:09 PM
|To: usr-tc@lists.xmission.com
|Subject: Re: (usr-tc) Solution for idle time limits and dual logins
|
|
|Mike Wronski said once upon a time:
|
|>I would propose that using SNMP, the %utilization over a time interval on a
|>per-interface basis could be obtained. Then you can drop users if they
|have less
|>than some % use for last time interval.. This is easy with HARC and
|possible with
|>Netserver. Its also not *perfect* but it would work.
|
|That would be very cool. Can we get it implemented into ARC code via
|RADIUS? ;-)
|
Sure.. The new harc is also a Soda Machine and Cotton candy maker. Gets the ISP
into the lucrative Remote Carnival Access business..
But seriously.. This should not be very difficult to program in PERL and run on
UNIX or (CHOKE!) NT..
-m
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of rbryn@cape-fear.net
|Sent: Thursday, October 22, 1998 10:24 AM
|To: usr-tc@lists.xmission.com
|Subject: (usr-tc) Radius Accounting Problems
|
|
|I have a problem with Radius Accounting packets reporting the wrong NAS
|port. When the Radius server recieves a packet for authentication it will
|report the correct NAS port but when it sends the Accounting start/stop
|packet it always reports NASPort=1.
|
|This problem only seemed to occur after upgrading TCS 3.3
|
|Config is:
|Total Control Chassis w/
|NMC v5.5.5
|HARC v.4.1.11
|Hiper DSP v.1.2.5
|
Known issue with 4.1.11. Get the service relase.
-M
This was a bug in 4.1.11 code of Hiper ARC . Get the latest code 4.1.72
code from
http://totalservice.usr.com
The latest code fixes this issue of NAS port.
Sanjay
On Thu, 22 Oct 1998 rbryn@cape-fear.net wrote:
> I have a problem with Radius Accounting packets reporting the wrong NAS
> port. When the Radius server recieves a packet for authentication it will
> report the correct NAS port but when it sends the Accounting start/stop
> packet it always reports NASPort=1.
>
> This problem only seemed to occur after upgrading TCS 3.3
>
> Config is:
> Total Control Chassis w/
> NMC v5.5.5
> HARC v.4.1.11
> Hiper DSP v.1.2.5
>
> Radius = IEA Software's RadiusNT v2.5
>
>
> Thanks
> Robb Bryn
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) Interested in Linux on the Netserver? From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-18 16:01:52
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jeff Mcadams
|Sent: Wednesday, November 18, 1998 3:41 PM
|To: usr-tc@lists.xmission.com
|Subject: Re: (usr-tc) Interested in Linux on the Netserver?
|
|
|Thus spake Allen Marsalis
|>Of course if there are any real hardware limits, they could be
|>minimized and dealt with so udp performance would be as good or better
|>than it is now, which is good enough for many.. just not us I
|>suppose.. We house alot of Quake players..
|
|OK, this is a rather random thought that your message here
|triggered...pardon my rambling.
|
|A NETServer running PPP (ignoring for the moment filters...I'll get back
|to this) passing an IP packet between a dial-up line and the
|ethernet...has no reason to know if a packet is UDP, TCP, GRE, or
|whatever. Obviously, to do filtering, that would require looking
|farther into the packet to determine if its UDP, and even farther to
|port numbers and the like. In the absense of filters though (we don't
|use filters on our NETServers at all) conceivably, the code might not
|even know if a packet was a UDP packet or not.
Correct..PPP packet data is reassembled into the protcol that is encapsulated.
This packet is pushed out onto the wire. (wire could be frame or ethernet).
|So...is the problem really with *UDP* packets? Or is the problem more
|one of handling a rapid stream of small packets (like Quake produces).
|Conceivably, a similar stream of small IP packets could be generated in
|a TCP connection...though its less likely...I wonder if that is affected
|as well?
Most traffic through the NS does not follow the same model as quake traffic.
Quake is a high rate of very small packets. Each packet reguardless of size has a
specific overhead associated with it. And then there is the overhead of the size.
So when a large number of small packets, reguardless of protocol, pass through
the Netserver, there is latency. It is for this reason that other usage does not
appear to experience the problem. Most of your dialup traffic is bursty
(asyncronous) larger packets. Quake is ...($10 word) almost isochronous. Its
sending packets on a clock tic. So what we have discovered is that Quake is best
played across pure ATM(not LANE) and not ethernet. :)..
-M
Subject:RE: (usr-tc) Archives? From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-18 16:04:44
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jamie Orzechowski
|Sent: Wednesday, November 18, 1998 3:51 PM
|To: usr-tc@lists.xmission.com
|Subject: (usr-tc) Archives?
|
|
|Just wondering where archives of this list are stored ... I thought they
|were on interpric.ae.usr.com but that's down ...
I think interpric is an adult site..;) interproc.ae.usr.com does not contain
archives of this list.
-M
Subject:RE: (usr-tc) Excessive data transferred / DNS resolution From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-18 16:27:21
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of
|David.Perrin@mnco.com
|Sent: Tuesday, October 13, 1998 1:21 PM
|To: usr-tc@xmission.com
|Subject: (usr-tc) Excessive data transferred / DNS resolution
|
|
|
|Working with the TCS 3.5 release using the Hiper ARC version 4.1.11 ,
|I'm running into a couple of significant
|issues.
|
|We are currently using a NetServer card in production and are planning
|to move to the Hiper ARC card in a new
|chassis, but I'm having the following problems:
|
|1. There's an excessive amount of data transferred during a session
|with the Hiper ARC card versus the
| Netserver card connection. It's on order of 5 times the data
|transferred with a Hiper ARC connection versus
| the Netserver connection.
How are you quantifing this? What kind of traffic are you putting on the line and
do you have traces??
|2. DNS connectivity is acting strangely. DNS seems to be active and
|working OK when telnetted into the
| Hiper ARC console itself. The problem occurs when attempting to
|use DNS via a Windows 95 session
| connected remotely via PPP. For example I can ping a DNS host
|name successfully, when using the
| hostname along with the domain name (ie. abcdef.corp.mnco.com),
|but receive a Bad IP address response
| when trying to use the the hostname (ie. abcdef) . The DNS
|domain name is set properly on the server
| and as I said earlier I can sucessfully use DNS with just a
|hostname when telnetted into the Hiper ARC
| card.
|
So hostname lookup does not work from the 95 client? Did you configure the HARC
to send its DNS information to the user?
this is done with 'SET PPP DNS_USAGE SYSTEM'. Or you can configure separate DNS
servers for PPP from the one configured for the HARC. 'SET PPP PPPDNS_PRIMARY
...' and then 'SET PPP DNS_USAGE PPP'.
When you type 'winipcfg' on the client, do you see the correct DNS server
address?
-M
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-18 16:34:07
Thus spake Mike Wronski
>I would propose that using SNMP, the %utilization over a time interval on a
>per-interface basis could be obtained. Then you can drop users if they
>have less
>than some % use for last time interval.. This is easy with HARC and
>possible with
>Netserver. Its also not *perfect* but it would work.
And also potentially bump off customers that are slow typists logged in
via telnet to a shell server or something...
Unforutunately, I think that solution is considerably *more* fraught
with peril than not reset'ing the idle timeout based on filters that
aren't going to catch everything. :/
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) Interested in Linux on the Netserver? From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-18 16:41:22
Thus spake Allen Marsalis
>Of course if there are any real hardware limits, they could be
>minimized and dealt with so udp performance would be as good or better
>than it is now, which is good enough for many.. just not us I
>suppose.. We house alot of Quake players..
OK, this is a rather random thought that your message here
triggered...pardon my rambling.
A NETServer running PPP (ignoring for the moment filters...I'll get back
to this) passing an IP packet between a dial-up line and the
ethernet...has no reason to know if a packet is UDP, TCP, GRE, or
whatever. Obviously, to do filtering, that would require looking
farther into the packet to determine if its UDP, and even farther to
port numbers and the like. In the absense of filters though (we don't
use filters on our NETServers at all) conceivably, the code might not
even know if a packet was a UDP packet or not.
So...is the problem really with *UDP* packets? Or is the problem more
one of handling a rapid stream of small packets (like Quake produces).
Conceivably, a similar stream of small IP packets could be generated in
a TCP connection...though its less likely...I wonder if that is affected
as well?
Like I said...this is a very random thought, but thought I would throw
it out for all to see. :)
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Just wondering where archives of this list are stored ... I thought they
were on interpric.ae.usr.com but that's down ...
----- Original Message -----
Sent: Wednesday, November 18, 1998 4:34 PM
>Thus spake Mike Wronski
>>I would propose that using SNMP, the %utilization over a time interval on
a
>>per-interface basis could be obtained. Then you can drop users if they
>>have less
>>than some % use for last time interval.. This is easy with HARC and
>>possible with
>>Netserver. Its also not *perfect* but it would work.
>
>And also potentially bump off customers that are slow typists logged in
>via telnet to a shell server or something...
>
>Unforutunately, I think that solution is considerably *more* fraught
>with peril than not reset'ing the idle timeout based on filters that
>aren't going to catch everything. :/
>--
>Jeff McAdams Email: jeffm@iglou.com
>Head Network Administrator Voice: (502) 966-3848
>IgLou Internet Services (800) 436-4456
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
This is a bug in 4.1.11 code. Fixed in service release 4.1.72
Sanjay
On Tue, 6 Oct 1998, Rick Payne wrote:
> Dragan D. Vecerina writes:
>
> > Hi,
> > Is there any workaround or suggestions for loss of accounting records
> > from HiperArc Card (4.1.11).
> > Problem is that it doesn't send (sometimes) ACCT OFF information
> > to radius server when user disconnect.
>
> Indeed, and does this explain why the ARC loses track of how many IP's its
> assigned from the pool?
>
> On one of ours today, we had no-one connected, but it still reckoned 2
> IP's from the pool were in use. *sigh*.
>
> Rick
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) CONNECTION PROBLEM From: Jeff Binkley <jeff.binkley@asacomp.com> Date: 1998-11-18 17:10:00
-> |
-> |We spoke to Brother and they were no help. Basically mumbling things like
-> if |you are trying to connect to an NT server it won't work and we only
-> support |dialing into Unix machines. They said send them to Earthlink. I
-> don't think |we want to start eliminating groups of users from our
-> collective ISps based |upon the equipment the end user has. That doesn't
-> seem to be good business |practice.
-> |
->
-> This looks like a problem we had with ASYNC_MAP of all 0's.. This is
-> resolved in
-> the 4.1.72-7 code.
Actually since this message Krish has taken numerous debug traces and is
investigating. Just watch out for these Brother Geobooks until we hear
back with a resolution. Brother's technical support hasn't been helpful
so far.
Jeff Binkley
ASA Network Computing
Subject:Re: (usr-tc) Interested in Linux on the Netserver? From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-18 17:25:29
Thus spake Mike Wronski
>Most traffic through the NS does not follow the same model as quake traffic.
>Quake is a high rate of very small packets. Each packet reguardless of
>size has a
>specific overhead associated with it. And then there is the overhead
>of the size.
>So when a large number of small packets, reguardless of protocol, pass through
>the Netserver, there is latency. It is for this reason that other
>usage does not
>appear to experience the problem. Most of your dialup traffic is bursty
>(asyncronous) larger packets. Quake is ...($10 word) almost isochronous. Its
>sending packets on a clock tic. So what we have discovered is that
>Quake is best
>played across pure ATM(not LANE) and not ethernet. :)..
Heh...ok...so I was right in my thinking that its not strictly Quake, or
even UDP that causes it...but is just lots of small packets that cause
problems...understandable. Just wanted to get clarification of what
really the crux of the problem was.
Obviously, very few applications are going to create lot's of small
packets like Quake does...and this could have largely been avoided by
good coding in Quake to avoid large numbers of small
packets...but...that's water under the bridge.
Thanks for the info Mike.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) ISDN: Netserver vs. Modems From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-18 17:27:41
Thus spake Mark Lemmert
>3com has suggested that I terminate the ISDN calls on the quad modems
>instead of on the Netserver to solve this problem. Has any body else
>had this problem? Can anybody give an opinion on whether 3coms suggested
>fix is likely to have an affect on this? Thanks!
I think I remember this happening, and, yes, terminating to Quad's did
fix the problem (maybe just avoided it?) Anyway...you want to terminate
on Quad's anyway...it's an all around better solution. :)
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:RE: (usr-tc) Solution for idle time limits and dual logins From: K Mitchell <mitch@keyconn.net> Date: 1998-11-18 17:43:56
At 02:56 PM 11/18/98 -0600, Mike Wronski wrote:
>
>Its actually not that simple. The architecture has filters associated with IP
>data and idle-timout with PPP data. By the time the filter gets the packet
the
>timer has already been reset. This implementation would require some
>reorganization of the levels and thus not be a simple boolean check.
>
>
>I would propose that using SNMP, the %utilization over a time interval on a
>per-interface basis could be obtained. Then you can drop users if they
have less
>than some % use for last time interval.. This is easy with HARC and
possible with
>Netserver. Its also not *perfect* but it would work.
This may be a really stupid question, but I'll stick my neck out anyhow.
<nomex=on>
Instead of trying to figure out ways to determine the type of traffic being
passed in order to determine if the only process running is an automated
one, wouldn't it be easier to poll events for consistancy? For example;
Event Time Traffic
1 01:03.28 24305 octets
2 01:08.22 24298 octets
If event 3 occurs in same time frame(in 5 mins +/-30 seconds), and traffic
remains +/-5% of previous counts, it is determined to be automated and
punted. This should readily detect ICQ or other keep-alives that send a
signal on a regular basis.
Ok gentlemen(and ladies), you may now commence to start blowing my theory
out of the water :)
Kirk
Kirk Mitchell-General Manager sysadmin@keyconn.net
Keystone Connect http://www.keyconn.net
Altoona, PA 814-941-5000 We Unlock the World
Subject:Re: (usr-tc) PRI vs T1 ? From: David Bolen <db3l@ans.net> Date: 1998-11-18 18:04:37
MegaZone <megazone@megazone.org> writes:
> Yes - and no. Some users have shown that CT1 lines *for them* get about a
> notch lower connect speeds than PRI. A good many users have reported this
> with 3Com, Lucent, Ascend, etc. However, not ALL users see this - and there
> hae been a couple of cases where users reported their CT1 lines were giving
> better connections.
>
> But based on what I've seen you do have a good chance of seeing *slightly*
> lower speeds on a CT1.
I'd pretty much agree with this. Using a CT1 circuit guarantees you
at least one robbed bit path - that of the final CT1 circuit to your
equipment.
If everything else (the phone network, the user's call path, the
user's line, etc..) were exactly the same and ideal, your CT1 path
should probably result in a 'tick' or two less connection rates (say
up to around 2400 baud difference).
However, in real life, often the limiting factor of connections is not
your final circuit. So if, for example, your local telco is robbing
bits anyway along the path for analog calls that's probably going to
limit things as much if not more so than your own circuit. And you'd
be surprised how little a phone call (even a local call) has to travel
through some telco networks before it traverses a robbed bit path.
Additionally, if the user's own line is putting more limitation on the
connection than the phone network, they'd never see the difference
between CT1 and PRI anyway. Averaged around our network, there isn't
much noticeable difference (statistically speaking) between our CT1
and PRI circuits.
This of course, assumes you are going with CT1 configured for E&M
signaling (don't touch ground or loop start).
As for ease of setup, there's really not much more to CT1 - just
define the signaling and that's about it - don't even have to worry
about the remote switch type or anything like you do with PRI.
If you can afford the effort and cost, I suppose you could initially
get one of each in a test area, let them process user calls for some
period, and then look at the results to see how they compare for your
user base. If not though, as a data point from our perspective, we
currently prefer PRI ourselves, but used to primarily use CT1 due to
cost factors, and still do so without any problems in areas where the
delta in cost is significant.
> There is also the fact you can't do ISDN DATA over a CT1, only ISDN DOSBS.
Definitely a good point if ISDN data termination needs to be offered.
> And I don't think the TC does DOSBS yet, does it? (If so, can it do ISDN
> DOSBS *and* modems over the same CT1 at the same time?)
Don't believe so.
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
Subject:Re: (usr-tc) V.90 is it safe to put on non-hyper dsp Total Control From: David Bolen <db3l@ans.net> Date: 1998-11-18 18:11:52
"Mike Hamrich" <mhamrich@drfast.net> writes:
> Any advice for taking a X2 enabled year and a half old 2.5.X release Total
> Control unit up to V.90 successfully?
Just upgrade the code to the V.90 system release (at least TCS 3.1).
Your x2 key in the NMC will also enable V.90 automatically.
> The one thing I did hear was to reset modem to factory default. Have not
> read if you need to upgrade in steps with new then what in the unit but
> older then what current. If you need to put interim release on. How do you
> get them?
You can jump right from your 2.5.x release to 3.1 - that's the same
upgrade path we took (2.5.1->3.1).
For modems, I do recommend upgrading, then issuing a command to each
one to restore to factory defaults, then issuing any settings you
customize, and then saving to NVRAM. That's for any upgrade, not just
this one.
In terms of order, we do the NMC first (so it can support any newer
objects we like to customize), and then just sort of work left to
right in a chassis (circuit card, modems, and NETServer). The order
isn't really that critical, but if you do the NETServer before the
modems, remember to check your ports after doing the modems to ensure
they are all live on the packet bus.
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
Subject:Re: (usr-tc) using round robin & fixed assignment From: David Bolen <db3l@ans.net> Date: 1998-11-18 18:24:58
Taino d Johnston <usr-list@accesscom.com> writes:
> What I am trying to find out is why when configured for 'fixed assignment'
> we were giving out busy signals? The problems went away once we switched
> to 'round robin'.
It's a software bug - well, I consider it as such.
What happens is that when the call release message comes down the PRI
D channel, the PRI card informs the modem which tears down the call.
However, the PRI immediately acknowledges the release message without
waiting for the modem to indicate it has finished resetting. The
problem is that the modem takes time to completely reset and be
prepared for a new call (in some releases up to perhaps 250ms or
more). During that time the telco thinks that the B channel is idle
and may immediately send down a new call - especially if this is a
busy hunt group and the channel is low in the group and the telco is
itself using ascending hunting within the group.
The PRI gets the call, tries to signal the modem over the packet bus
and one of two things happens:
* The modem isn't ready at all, and the PRI times out in the
configuration step for a new call trying to talk to the modem.
the configuration step for the new call. The result is a delay to
the user followed by a busy when the PRI finally gives up. If
enabled, the PRI should generate a callTermFailedEvent trap for
this call.
* The modem is sort of ready but not fully - it acknowledges the PRI
indication, but when it gets ahold of the call can't properly
process it and the user perceives dead air. If enabled, the modem
should generate a DteRingNoAns trap for this call.
You can also see the counts of these sorts of failures on some of the
PRI debug summary screens for modem errors.
Why this tends to happen more on PRI than CT1 is probably a
combination of the faster and out-of-band signaling for call setup and
teardown, and the way in which the PRI communicates to modems OOB over
the packet bus for call setup as opposed to using TDM bus patterns for
the CT1 code, both of which affect the general setup and teardown
timing sequence during call processing.
The reason that changing to round-robin allocation "fixes" this
problem (in many cases) is that the newer call is not going to try to
hit the same modem, but one of the two "idle" modems (on a T1). So
instead of a limiting factor of the reset time of a single modem for
your call cycle time, you have divided that time by 3 (worst case).
Normally unless you lose several calls simultaneously, this is enough
to avoid the problem.
Having the telco distribute calls over your entire hunt group in a
statistical fashion (or even round-robin) can help too, but not once
the hunt group gets full or very busy since it's still likely to pick
the B channel just released for the next call.
The round robin selection also isn't a possibility in typical E1
configurations since in that case you don't have the automatic buffer
modems.
We had terrible problems with this at our E1 sites, and actually had
to remove 4 B channels from service to give ourselves enough of a
"buffer" to handle the call load.
The "fix" (which doesn't require round-robin) is to both have the
modem reset more quickly (at least to the point where it can start
accepting new call information) as well as have the PRI wait for the
modem to say it is done before acknowledging the release message from
the telco. True, the PRI does have some hard timing limits within
which it must acknowledge the message according to the D channel
protocol, but there's room in there to wait for the modem in most
cases.
I can't speak to the general availability of such a fix or not, but it
is possible. Of course, switching to round robin (for most T1
configurations) is generally pretty effective too.
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
Subject:Re: (usr-tc) using round robin & fixed assignment From: David Bolen <db3l@ans.net> Date: 1998-11-18 18:34:29
In a prior note, I wrote:
> (...)
> the user followed by a busy when the PRI finally gives up. If
> enabled, the PRI should generate a callTermFailedEvent trap for
> this call.
BTW, these call event traps that the PRI can generate can be used in
general to track call delivery problems that occur between the PRI
card and other components. At a minimum the callTermFailedEvent trap
will give you an indication of every call that arrived at the PRI
(from the telco) but for some reason was not properly serviced by the
target device (either modem or ISDN gateway).
Enabling callArriveEvent permits you to count all call arrivals seen
by the PRI card, so you can get a failure percentage. The others
(callConnectEvent, callTermNormalEvent, and callEvent) can also be
used to build a better picture of just how calls are flowing from the
PRI card to other chassis components.
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
How can I set ARC to log the packets dropped from a filter?
- Marcelo
Subject:Re: (usr-tc) Solution for idle time limits and dual logins From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-18 20:07:27
Thus spake Mike Wronski
>|Oh, so if you can't provide a *perfect* tool, then you won't give us
>|*any* tools to work with at all. Thanks a lot there!
>I don't provide anything but support for the product. People at higher levels
>have to wade through all the possible features and some fall out in favor of
>others.
Oh, believe me, I understand that...but please understand that we've
gotten the run-around on features before. :)
>Its just not possible to make a product everyone is in love with.. I
>don't know for certain, but do any other vendors have this feature? I
>don't think
>we are unique here.
I've not seen any NAS vendors that do, but as I mentioned previously,
I've used a pppd (Morningstar) that had this feature and it was a
*wonderful* feature to have.
>Agreed, but what is the alternative to RADIUS? TACACS? Is MS Anything perfect?
>nope. What do most people use??
>With that logic: I think LINUX is better than MS. Should I not support MS even
>though its the defacto standard.
You, of course, have a point...RADIUS is the best thing out there for
doing this... However, you see my point as well, right? We, the
customers need tools to be able to make our networks work the best we
can, and we're getting you (and I don't mean this to be a slam on you
personally...I'm *glad* you're responding to these postings) telling us
that our desires for features that we want are wrong. :) You can see
how that would be a little bit grating. :)
As I said...I've used this feature before, on a UNIX based pppd and it
was wonderful to have...I'd like to be able to do this on a NAS too. :)
>I now expect a jump off the bridge reply from someone.. And
>"No I wouldn't do it just because all my friends did! :)"
Blah...I've always thought that was a stupid argument. :)
>|There are plenty of features that you could develop that would benefit
>|us greatly...let's come up with a list and I'll rank them how I feel
>|they would be most useful to us. I guarantee you that idle-timer filter
>|list would be *very* high up there.
>You should do that. Put don't post it here. It is far more effective when
>submitted formally as a Feature Request/CRA and submitted by many people.
Hey, I'm more than happy to submit it through formal channels...but
let's also consider this as a possible discussion ground for what those
features that we request might be. Perhaps we could all work on a list
together and could use that as a base for features that we'd all like to
see put in, and possibly augment that with our own individual requests.
:)
>(OSPF flame expected here.. Don't
>bother! I know its been a long time coming...But.. its coming...Really....No
>fooling...Its not vaporware...)
Heh...wasn't gonna be a flame, but it *was* gonna get mentioned. :) I
know its coming...is it 4.1.72 perchance? I thought I had heard some
talk of it being available in ER's/SR's, but perhaps I'm just imagining
things.
>I do my best to respond to issues on this list and have no desire to fight with
>anyone over feature sets..
Please, your input is greatly appreciated...and you pointed out a very
real limitation on why this would be a difficult thing to implement
(idle-timer is reset before the IP header is even parsed)...that's very
understandable. At least when I go talk to Tom Goodman I'll understand
if it doesn't show up in the very next release. :)
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) Pt to Pt Test Link From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-18 20:09:37
Thus spake Mark Lemmert
>If you made an cat5 cable /w RJ45 ends with
>a special wiring scheme you can simulate make
>the two CSUs talk as though you had a telco circuit
>between them.
Well...it's not *quite* that simple as you have issues of clocking and
stuff to deal with...one of the systems would have to generate clock for
the other to see on the T1...so the config is slightly different than if
you had a telco simulator or something to do it.
>See your CSU manual to find out which pins are
>for transmit/receive and tip/ring.
You'll want 1 <-> 4, and 2 <-> 5, other pins aren't needed.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Pete Ashdown was heard to say:
>You might need a memory upgrade. Check to see if your card has 16 megs or
>more, if not, you will need to replace the 4 meg SIMM inside with a 16 meg
>one.
I've never seen a netserver with less than 4M on the board itself. That's
why alot of our netservers have 20M (and a few 32M)
--Ricky
Pete Ashdown was heard to say:
>>This seems to be broken on current ARC releases up to and including at
>>least 4.1.78; you have to also use "Max_Channels = 1" (which is a VSA) to
>>do the right thing there. On the NETserver, it's fine. Go back through
>>the archives a month or two and you can see me whining away about this :)
>>We actually have both Port-Limit AND Max_Channels in the Radius users
>>file. That seems to take care of things.
>
>According to the fix documentation, 4.1.72 follows the RFC on this. Does
>anyone know if that returns things to normal? ;-)
What fix docs? There wasn't anything in the zip file.
--Ricky
Subject:Re: (usr-tc) upgrade for Netserver card available? From: David Bolen <db3l@ans.net> Date: 1998-11-18 21:33:45
Ricky Beam <jfbeam@Interpath.net> writes:
> Pete Ashdown was heard to say:
> >You might need a memory upgrade. Check to see if your card has 16 megs or
> >more, if not, you will need to replace the 4 meg SIMM inside with a 16 meg
> >one.
>
> I've never seen a netserver with less than 4M on the board itself. That's
> why alot of our netservers have 20M (and a few 32M)
True, but it sounds like Pete is talking about an 8MB configuration
(4MB on the board itself and a 4MB SIMM) or else there wouldn't be a
4MB SIMM to remove.
8MB was the format used for a while when they first introduced the
ISDN capable NETServers (with the Munich daughtercard) and even for
non-ISDN NETServers. I'm not sure how long they actually shipped
NETServers with only the 4MB on the board, although we sure had a lot
of them to deal with later :-)
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
Subject:Re: (usr-tc) Interested in Linux on the Netserver? From: Ricky Beam <jfbeam@interpath.net> Date: 1998-11-18 21:48:36
Allen Marsalis was heard to say:
>Lets say you get USR to cooperate (1st feat). Then you get linux up
>on the netserver (2nd feat). What if Quake still lags? I mean Quake
>Lag is a symptom of a much bigger problem. And it might be hardware
>related and that has been hinted at.
A packet is a packet. I don't buy the "but it's the hardware" excuse.
--Ricky
Saw the same thing... Several things. Make sure if you are using
ODBC and SQL for backend to have the SESSIONIDENTIFIER in your
SQL table to at least 14 characters in length -- it jumped from 12 to 14
in this release (under radius NT). Also look at the radius config on the
Hyper ARC. We have set the radius attributes to Ascend style and
re-ported the SQL table to handle ports 0-23 in slot 1 1024 to 1024+23
in slot 2, 2048 to 2048+23 in slot 3, etc.. (for PRIs). Some of this
requires
SQL by hand. It works fine now after all this.. The biggest problem is
that
you have to drop the Calls table under IEAs Radius NT and recreate it
with the increased column width for session identifier.
-Chris
>-----Original Message-----
>From: rbryn@cape-fear.net
>Sent: Thursday, October 22, 1998 11:24 AM
>To: usr-tc@lists.xmission.com
>Subject: (usr-tc) Radius Accounting Problems
>
>
>I have a problem with Radius Accounting packets reporting the wrong NAS
>port. When the Radius server recieves a packet for authentication it
>will
>report the correct NAS port but when it sends the Accounting start/stop
>packet it always reports NASPort=1.
>
>This problem only seemed to occur after upgrading TCS 3.3
>
>Config is:
>Total Control Chassis w/
>NMC v5.5.5
>HARC v.4.1.11
>Hiper DSP v.1.2.5
>
>Radius = IEA Software's RadiusNT v2.5
>
>
>Thanks
>Robb Bryn
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
Subject:Re: (usr-tc) Interested in Linux on the Netserver? From: Mark R. Lindsey <mark@vielle.datasys.net> Date: 1998-11-18 22:35:53
: Allen Marsalis was heard to say:
: >Lets say you get USR to cooperate (1st feat). Then you get linux up
: >on the netserver (2nd feat). What if Quake still lags? I mean Quake
: >Lag is a symptom of a much bigger problem. And it might be hardware
: >related and that has been hinted at.
:
: A packet is a packet. I don't buy the "but it's the hardware" excuse.
Of Allen's original feats, the first has already started to happen,
and the second is much closer to done than you might imagine.
I'd bet a small sum of money that there's nothing in the hardware
specifically related to UDP. The Netserver is a general-purpose computer
with some nifty interfaces.
Subject:RE: (usr-tc) Solution for idle time limits and dual logins From: Charles Sprickman <spork@inch.com> Date: 1998-11-18 22:53:04
On Wed, 18 Nov 1998, Mike Wronski wrote:
> You should do that. Put don't post it here. It is far more effective when
> submitted formally as a Feature Request/CRA and submitted by many people. Like
> most things, if enough people ask for it and it makes financial sense to have a
> feature it tends to get more consideration. (OSPF flame expected here.. Don't
> bother! I know its been a long time coming...But.. its coming...Really....No
> fooling...Its not vaporware...)
How exactly does one make a feature request? I'll request my brains out,
just show me the way... I hope I don't need a support contract to do
that..
Thanks,
Charles
>
> I don't want to start a rant or flame war on this issue. I think that was is
> desired is understood by those on this list.
> I do my best to respond to issues on this list and have no desire to fight with
> anyone over feature sets..
>
> -M
>
>
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
=-----------------= =
| Charles Sprickman Internet Channel |
| INCH System Administration Team (212)243-5200 |
| spork@inch.com access@inch.com |
= =----------------=
Subject:RE: (usr-tc) Archives? From: Charles Sprickman <spork@inch.com> Date: 1998-11-18 22:57:32
> |Just wondering where archives of this list are stored ... I thought they
> |were on interpric.ae.usr.com but that's down ...
They were at http://usr-tc.datasys.net, but it looks like it hasn't been
updated and the search and index features have gone away...
Charles
>
> I think interpric is an adult site..;) interproc.ae.usr.com does not contain
> archives of this list.
ehhh...
>
>
> -M
>
>
>
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
=-----------------= =
| Charles Sprickman Internet Channel |
| INCH System Administration Team (212)243-5200 |
| spork@inch.com access@inch.com |
= =----------------=
Subject:Re: (usr-tc) New Hiper ARC service release From: Russ Miescke <russm@powerweb.net> Date: 1998-11-19 01:31:45
Is this a newer code than the ER 4.1.78? Does 4.1.78 contain all of the
fixes from 4.1.72? I guess I am hesitant to change something that is sort
of working. Anybody load this yet?
Russ Miescke
Power Web Connect
russm@powerweb.net
http://www.powerweb.net
----- Original Message -----
Sent: Tuesday, November 17, 1998 6:11 PM
sagarwal@crash.ae.usr.com was heard to say:
>Check out a new Hiper ARC service release code version 4.1.72 at
>
>http://totalservice.usr.com
I love the version number... ne041727.zip :-) I also like the way I don't
need a service contract to grab it (or do I?)
--Ricky
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
This is my current procedure for setting modems, for both hiperdsp and
quad modems hooked up to hiperarcs. Comments welcomed.
restore modems from default
save modems to nvram
on hiperarc: reset modem_group all
make following modem (not template) changes
Line interface options - carrier loss detect delay = 14
signal converter settings - v.42 selective reject = enable
Call control options - response to +++ = ignore esc code
save modems to nvram
hardware reset
verify settings if desired...
DTE Interface Settings
Modem Escape Char (S2) 255
Call Control Options
Result Codes (Qn) displayResult
Verbal/Numeric Result Codes numeric
Result Code Groups 0
ARQ Result Codes arqResultsDisabled
Response to +++ ignoreEscCode
Rings for Auto Answer 0
ATZ Handling over Packet Bus atZPbIgnore
Packet Bus Answer Only (S47.5) enable
Signal converter settings
v.42 selective reject enable
Line interface options
carrier loss detect delay 14
--
Aaron Nabil
Subject:Re: (usr-tc) Pt to Pt Test Link From: Greg Coffey <greg@coffey.com> Date: 1998-11-19 07:21:40
Thanks for the reply and info. BTW, we did get it to work using two
motorola CSU's. The compression worked pretty well too, now if we can just
get USWest to get the line changed correctly someday before the turn of the
century.............., we'll see how it works in the real world.
At 08:09 PM 11/18/98 -0500, you wrote:
>Thus spake Mark Lemmert
>>If you made an cat5 cable /w RJ45 ends with
>>a special wiring scheme you can simulate make
>>the two CSUs talk as though you had a telco circuit
>>between them.
>
>Well...it's not *quite* that simple as you have issues of clocking and
>stuff to deal with...one of the systems would have to generate clock for
>the other to see on the T1...so the config is slightly different than if
>you had a telco simulator or something to do it.
>
>>See your CSU manual to find out which pins are
>>for transmit/receive and tip/ring.
>
>You'll want 1 <-> 4, and 2 <-> 5, other pins aren't needed.
>--
>Jeff McAdams Email: jeffm@iglou.com
>Head Network Administrator Voice: (502) 966-3848
>IgLou Internet Services (800) 436-4456
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
______________________________________________________
Thanks,
Greg Coffey 307-234-5443 Fax 307-234-5446
CoffeyNet v.90 56k Access for Casper & Douglas
142 S. Center St.
Casper, WY 82601 http://www.coffey.com
Subject:Re: (usr-tc) Interested in Linux on the Netserver? From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-19 08:31:37
Thus spake Ricky Beam
>Allen Marsalis was heard to say:
>>Lets say you get USR to cooperate (1st feat). Then you get linux up
>>on the netserver (2nd feat). What if Quake still lags? I mean Quake
>>Lag is a symptom of a much bigger problem. And it might be hardware
>>related and that has been hinted at.
>A packet is a packet. I don't buy the "but it's the hardware" excuse.
See my other thread with Mike Wronski about the specifics of the
problem...a packet is not just a packet...there are other issues to deal
with...most significant in this situation is that its a lot of small
packets. That is very difficult for a router to deal with. Yes, a
486/100 *should* be able to handle 48 ports of this if the code were
decent (check out the cisco 2500's...they easily handle this much
traffic with a much less powerful cpu), but it may be so insideous that
its easier just to scrap the whole ComOS code base and do their own...as
they've done with Pilgrim. That, however, doesn't explain the
collosally stupid decision *not* to support Pilgrim on the
NETServer...that is, unless the Pilgrim code is as inefficient in packet
handling, so it would run into the same problems that the NETServer does
in this case. I doubt that's the case though since the ARC's seem to be
handling many DSP's without problem from reports that I've heard.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:RE: (usr-tc) v.90 connect problems From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-19 09:03:13
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of David Bolen
>Sent: Tuesday, November 17, 1998 2:30 PM
>To: usr-tc@lists.xmission.com
>Subject: RE: (usr-tc) v.90 connect problems
>
>
>"Brian K McIntire" <bmcintire@commnet.com> writes:
>
>> is related to some setting your chassis. I've seen this a
>thousand times on
>> a thousand different AOL chassis.
>
>Just for a minor counterpoint, on most of our chassis using
>Channelized T1 spans, we don't really see all that much of a
>difference.
The difference I'm referring to is not just in performance but reliability
and sometimes, on ease of turn up. On PRI's you do run this risk of losing
an entire span whenever a d-cannel goes down. Just another argument for
Backup D. One advantage, however insignificant to someone as experienced as
you, is normal customer's who do not understand the intricacies of T1's and
PRI do not have to fight the Telco's as much. Telco's all across the United
States are notorious for screwing up. Especially on initial turn up. This
may not be an issue for you since you give each telco a detailed spec of
exactly what you want and know how to work with them if it doesn't happen.
What about these other guys on this list? The one who have one or two
chassis who are just getting started. There are some very intelligent
people on this list but many have no idea what a T1 or a PRI is much less
how to troubleshoot problems. (Hence, the need for the list) Fact is, in
my experience, there are considerably more things the telco could screw up
on a T1 than on a PRI. By nature of setup there should be no padding on a
PRI. I've seen Telco's do it, but there should be none. All over the US
Telco's usually through padding on without thinking twice. On T1's, no
matter how many times you tell a telco you need Channelized T1, I've seen
them turn it up with all the provisioning wrong. Or the damn thing is going
through an additional analog to digital conversion. You won't see a channel
bank on a PRI! Even a complete moron wouldn't put a T1 through a channel
bank and call it a PRI.
>In fact, since telcos often rob bits even within short
>call distances, the fact that the Channelized T1 implies a robbed bit
>doesn't necessarily hurt all that much - often the PRI's don't even
>bring an advantage to the 56K protocols like x2 or V.90. (For
>example, calling to a local T1 in my office gets me the same
>connections as calling a PRI located only about 10 miles west of me,
>and both connected to the same telco).
>
>Of course, it is important how you configure your T1's - specifically
>the signaling. If you opt for ground or loop start (we've only ever
>used ground), you're much more likely to have issues and I'm more
>likely to agree with your point, but not because of the digital T1
>span itself. It's because in virtually all cases, the telco is going
>to provision that at the concentrator end with a series of analog
>copper pair circuits feeding from their switch into a channel bank
>that handles the T1 framing. In such a configuration you really have
>an extra analog hop, individual pairs that can go bad, etc, and what
>you're hurt by is the central office configuration.
Definitely
>
>But an E&M configuration is almost always the opposite - digital
>straight into the switch
Again, assuming your telco knows what the hell they are doing.
> - and in my experience behaves just as nicely
>as a PRI circuit and very closely even to the call rates achieved.
>
>> PRI's, by design, are superior to T1's
>
>Pedantic point - unless you are at an international site using E1's, a
>"PRI" _is_ a "T1" fundamentally. Same framing. It's the signaling that
>is different, so you need to qualify the T1 as, for example,
>Channelized with E&M signaling.
>
See above notes. Again, you are assuming I'm discussing nothing but
throughput or connect speeds.
>> and have significantly less things that could go wrong (assuming your
>> provisioned has any notion of what he is doing) It was probably your T1.
>
>Heh - try to tell that to someone trying to troubleshoot D channel
>signaling or getting the non-standard "service message" support to
>work everywhere. PRI's aren't automatically better in all
>circumstances ;-)
How often do you have to troubleshoot d-channel signaling? What percentage
of chassis on your network require you to troubleshot d-channel signaling
each month? I had several thousand chassis I was responsible for working
on. Over a period of several months I believe I may have had one or two
that required real troubleshooting. The rest of the PRI problems we had
were telco screw ups or problems after brown outs.
As far as the non standard message goes I definitely know what you are
talking about. It's a pain in the ass!
>
>Don't get me wrong - I love PRI's as much as the next guy, and we're
>certainly using them as the standard configuration for all new
>equipment, but it wasn't all that long ago when they were too
>expensive in various regions and we still have a whole lot of
>Channelized T1 (far preferring E&M but we've still got the occasional
>ground start - economic reasons again) working in the network, and in
>general behaving just as nicely as the PRI circuits.
>
>So I wouldn't automatically assume that a Channelized T1 configuration
>has to be inferior. It can be configured wrong (but so can a PRI) and
>it can have different types of failures than a PRI, but it can also
>work just as well.
See above. David, I appreciate your enthusiasm. In fact, I love it. Your
experience with this equipment is unique and everything but uninformed.
However, you must remember, when you are posting to this list you are
talking to a tremendous number of people no where near as informed as you.
They experience different kinds of problems from sheer lack of experience
and understanding. Again, I say, stop assuming your situation applies to
everyone else. It doesn't!
Brian
>
>-- David
>
>/-----------------------------------------------------------------------\
> \ David Bolen \ Internet: db3l@ans.net /
> | ANS Communications, Inc. \ Phone: (914) 701-5327 |
> / 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
>\-----------------------------------------------------------------------/
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) New Hiper ARC service release From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-19 09:05:15
Is there a way to get a list of what hasn't been fixed yet?
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Mike Wronski
>Sent: Tuesday, November 17, 1998 3:21 PM
>To: usr-tc@lists.xmission.com
>Subject: RE: (usr-tc) New Hiper ARC service release
>
>
>
>
>|-----Original Message-----
>|From: owner-usr-tc@lists.xmission.com
>|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Mike Andrews
>|Sent: Tuesday, November 17, 1998 2:52 PM
>|To: usr-tc@lists.xmission.com
>|Subject: Re: (usr-tc) New Hiper ARC service release
>|
>|
>|Is there a list of what's changed between 4.1.78 and 4.1.72?
>|
>
>All changes are noted in the PDF's on totalservice. The list shows
>the delta from
>4.1.11
>All fixes in 4.1.78 are in there.
>
>-M
>
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) Corresponding modem is unavailable From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-19 09:31:57
Not sure if someone answered this for you yet.
I have seen this many times. Are you running PRI or T1? Check the Line
Interface Options/Line Interface source. Make sure it is set to T1tdm or
PRItdm depending on what kind of circuit you have. Save it to NVRAM and
reboot the quads
========================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= Your Remote Access and Security Experts! =
= Firm partnerships est. with current and emerging leaders of today's =
= technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
= Serving North America and parts of Canada. Reselling to the world. =
========================================================================
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Curt Shambeau
>Sent: Wednesday, November 18, 1998 1:46 PM
>To: USR Total Control Mailing list
>Subject: (usr-tc) Corresponding modem is unavailable
>
>
>
>Quad modem chassis running 4.2.1 on T1 card (running channelized T1's)
>modem cards running 5.10.9, HiPer ARC 4.0.29, NMC running 5.5.5
>
>When doing a DS0 performance monitor, the DS0 channels show idle, but the
>modem connected to the DS0 status is "corresponding modem is unavailable."
>
>I've tried restoring the chassis from defaults, and resetting everything,
>to no avail. Has anyone seen this before??
>
>--------------------------------------------------------------------------
>| Curtis V. Shambeau | curt@execpc.com | http://www.execpc.com/~curt |
>| Executive Vice President - Exec-PC, Inc. |
>--------------------------------------------------------------------------
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) Corresponding modem is unavailable From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-19 09:31:57
Not sure if someone answered this for you yet.
I have seen this many times. Are you running PRI or T1? Check the Line
Interface Options/Line Interface source. Make sure it is set to T1tdm or
PRItdm depending on what kind of circuit you have. Save it to NVRAM and
reboot the quads
========================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= Your Remote Access and Security Experts! =
= Firm partnerships est. with current and emerging leaders of today's =
= technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
= Serving North America and parts of Canada. Reselling to the world. =
========================================================================
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Curt Shambeau
>Sent: Wednesday, November 18, 1998 1:46 PM
>To: USR Total Control Mailing list
>Subject: (usr-tc) Corresponding modem is unavailable
>
>
>
>Quad modem chassis running 4.2.1 on T1 card (running channelized T1's)
>modem cards running 5.10.9, HiPer ARC 4.0.29, NMC running 5.5.5
>
>When doing a DS0 performance monitor, the DS0 channels show idle, but the
>modem connected to the DS0 status is "corresponding modem is unavailable."
>
>I've tried restoring the chassis from defaults, and resetting everything,
>to no avail. Has anyone seen this before??
>
>--------------------------------------------------------------------------
>| Curtis V. Shambeau | curt@execpc.com | http://www.execpc.com/~curt |
>| Executive Vice President - Exec-PC, Inc. |
>--------------------------------------------------------------------------
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) Solution for idle time limits and dual logins From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-19 09:35:35
It's not a matter of can't be done. It's a matter of Can we make it better?
The answer is yes. So lets do it.
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jeff Mcadams
>Sent: Wednesday, November 18, 1998 2:05 PM
>To: usr-tc@lists.xmission.com
>Subject: Re: (usr-tc) Solution for idle time limits and dual logins
>
>
>Thus spake Mike Wronski
>>I don't have a good answer for that. Idle-timeouts are only usefull
>>against the
>>non-savy user and those are getting fewer by the day.
>
>>So what do you count as activity that should reset the idle timer?
>
>Ideally, this should be administrator configurable...a filter list would
>be perfect.
>
>>No matter what you define as idle, someone will come up with a
>way to use data
>>you don't check to keep the channel open.. I can think a simple
>way to use HTTP
>>gets to keep the line open.. I think you are fighting a battle
>that cant be
>>won..
>>The user who wants to, will nail their connection up.. A resonable
>>session limit
>>may be your only hope.
>
>Oh, so if you can't provide a *perfect* tool, then you won't give us
>*any* tools to work with at all. Thanks a lot there!
>
>You've obviously never run as ISP. Using the current filter rule setup
>to define what traffic resets the idle timeout, I could define a filter
>of about 5 to 10 lines that would eliminate 99.9% of our problems with
>people camping on lines. Yes, there will be users that will work around
>it, its not a perfect solution, but then again, RADIUS isn't a perfect
>solution for authenticating users the way everyone needs either, so why
>do you support RADIUS?
>
>>I think the configurable idle timeout filter sounds like a good
>idea until you
>>really look at its implementation and wheather it will even do what
>>you want.
>
>I *KNOW* it'll do what I want because I've used a pppd in the past that
>supported this sort of thing and it worked like a charm!
>
>>It
>>would work for a short time and then the users will figure out a way
>>around it. I
>>think there are many other features that time & resources could be spent
>>developing that would benefit the ISP more than this.
>
>There are plenty of features that you could develop that would benefit
>us greatly...let's come up with a list and I'll rank them how I feel
>they would be most useful to us. I guarentee you that idle-timer filter
>list would be *very* high up there.
>--
>Jeff McAdams Email: jeffm@iglou.com
>Head Network Administrator Voice: (502) 966-3848
>IgLou Internet Services (800) 436-4456
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) Confirmation From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-19 09:37:07
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Ricky Beam
|Sent: Wednesday, November 18, 1998 8:25 PM
|To: usr-tc@lists.xmission.com
|Subject: Re: (usr-tc) Confirmation
|
|
|Pete Ashdown was heard to say:
|>>This seems to be broken on current ARC releases up to and including at
|>>least 4.1.78; you have to also use "Max_Channels = 1" (which is a VSA) to
|>>do the right thing there. On the NETserver, it's fine. Go back through
|>>the archives a month or two and you can see me whining away about this :)
|>>We actually have both Port-Limit AND Max_Channels in the Radius users
|>>file. That seems to take care of things.
|>
|>According to the fix documentation, 4.1.72 follows the RFC on this. Does
|>anyone know if that returns things to normal? ;-)
|
|What fix docs? There wasn't anything in the zip file.
|
There on the web site.. Next to the link for the code..
-M
Subject:(usr-tc) Re: your mail From: Dominic Ciresi <dciresi@defunct.ae.usr.com> Date: 1998-11-19 09:47:08
Brian,
This is a known bug. There is a fix in SAS 6.0.92.(Solaris) or SAS 6.0.91
(HP). Open a ticket and I will deliver code to you.
Dominic
On Thu, 19 Nov 1998, Brian K McIntire wrote:
> This is directed to anyone out there who may be using Secure-ID with TC S/A
> for Unix. We have installed S/A 5.5.4 and secure-id works fine. With the
> same config on 6.0.7 it doesn't. The debug is useless. All it says is
> invalid pass code. We down grade back to 5.5.4 it works again. Any ideas?
> Is this a known bug?
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) using round robin & fixed assignment From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-19 09:58:00
The time to reset a b-channel at Telco by default is probably close to 20
ms. This time can be changed. The key is knowing precisely how long is the
perfect time. This would be easy to find out with experimentation and
statistics. Talk to your telco about it.
I would be very interested to find out what your research uncovers.
========================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= Your Remote Access and Security Experts! =
= Firm partnerships est. with current and emerging leaders of today's =
= technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
= Serving North America and parts of Canada. Reselling to the world. =
========================================================================
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Taino d Johnston
>Sent: Wednesday, October 14, 1998 5:23 PM
>To: usr-tc@lists.xmission.com
>Subject: (usr-tc) using round robin & fixed assignment
>
>
>We recently switched telcos, which then switched us from channelized T1
>lines to PRI lines. The telco switch was from PacBell to ICG. Upon
>switching we found ourselves facing a lot of problems, however, it was this
>last problem that we felt was quite strange and worth noting.
>
>In order to do some tests with the new service, we changed the
>configuration of our equipment from 'round robin' to 'fixed assignment'.
>After we thought the experienced problems were corrected we started getting
>reports of busy signals -- during time periods when there were modems
>available. The problem turned out to be that the PRI lines were resetting
>faster than our modems were. Essentially, this meant calls would be routed
>back down to a newly opened line on the PRI but would hit a modem that had
>not been reset. Our testing found that we were giving busy signals out
>about 20% of the time. Switching back to 'round robin' seems to have since
>corrected the problem and we have not given out any more busy signals.
>
>What I am trying to find out is why when configured for 'fixed assignment'
>we were giving out busy signals? The problems went away once we switched
>to 'round robin'.
>
>We are running a few TC chassis containing Quad Digital & Digital/Analog
>modem cards with PRI lines. All our Quad modem cards, NETserver PRI cards,
>Dual PRI cards and NMC cards are running the current versions of the
>software.
>
>Taino Johnston
>Manager, Technical Support
>Access Internet Communications
>
>+----------------------------------------------------------------+
>| Taino d Johnston | Phone: (408) 777-8190 |
>| Manager, Technical Support | Tech Support: (408) 342-0551 |
>| | Fax: (408) 777-8191 |
>| Access Internet Communications | http://www.accesscom.com/ |
>| tdj@accesscom.com | support@accesscom.com |
>+----------------------------------------------------------------+
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Already checked. I'm running Ch.T1's, and it's set to the T1tdm bus.
Subject:(no subject) From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-19 10:14:59
This is directed to anyone out there who may be using Secure-ID with TC S/A
for Unix. We have installed S/A 5.5.4 and secure-id works fine. With the
same config on 6.0.7 it doesn't. The debug is useless. All it says is
invalid pass code. We down grade back to 5.5.4 it works again. Any ideas?
Is this a known bug?
> What slot is your T1 card in? You may have some troubleshooting ahead. Try
> pulling every card out of the chassis except 1 quad and the t1 and try it
> again. If it maps correctly put one card at a time back into the chassis
> until it stops working. Once it stops working you have probably found a bad
> card
Hadn't thought of that... Thanks, I'll give it a try.
The T1 card is in the first slot of the chassis. It was actually a brand
new Quad chassis that I put an ARC into (trying to get all the netserver's
out of the network).
| Curtis V. Shambeau | curt@execpc.com | http://www.execpc.com/~curt |
| Executive Vice President - Exec-PC, Inc. |
> What slot is your T1 card in? You may have some troubleshooting ahead. Try
> pulling every card out of the chassis except 1 quad and the t1 and try it
> again. If it maps correctly put one card at a time back into the chassis
> until it stops working. Once it stops working you have probably found a bad
> card
Hadn't thought of that... Thanks, I'll give it a try.
The T1 card is in the first slot of the chassis. It was actually a brand
new Quad chassis that I put an ARC into (trying to get all the netserver's
out of the network).
| Curtis V. Shambeau | curt@execpc.com | http://www.execpc.com/~curt |
| Executive Vice President - Exec-PC, Inc. |
>They were at http://usr-tc.datasys.net, but it looks like it hasn't been
>updated and the search and index features have gone away...
>
>Charles
The archives are available at:
http://www.xmission.com/pub/lists/usr-tc/archive/
Latest posts are at the bottom of the page.
Blann
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:RE: (usr-tc) using round robin & fixed assignment From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-19 10:59:15
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of David Bolen
>Sent: Wednesday, November 18, 1998 5:25 PM
>To: usr-tc@lists.xmission.com
>Subject: Re: (usr-tc) using round robin & fixed assignment
>
>
>Taino d Johnston <usr-list@accesscom.com> writes:
>
>> What I am trying to find out is why when configured for 'fixed
>assignment'
>> we were giving out busy signals? The problems went away once we switched
>> to 'round robin'.
>
>It's a software bug - well, I consider it as such.
>
>What happens is that when the call release message comes down the PRI
>D channel, the PRI card informs the modem which tears down the call.
>However, the PRI immediately acknowledges the release message without
>waiting for the modem to indicate it has finished resetting. The
>problem is that the modem takes time to completely reset and be
>prepared for a new call (in some releases up to perhaps 250ms or
>more). During that time the telco thinks that the B channel is idle
>and may immediately send down a new call - especially if this is a
>busy hunt group and the channel is low in the group and the telco is
>itself using ascending hunting within the group.
>
>The PRI gets the call, tries to signal the modem over the packet bus
>and one of two things happens:
>
> * The modem isn't ready at all, and the PRI times out in the
> configuration step for a new call trying to talk to the modem.
> the configuration step for the new call. The result is a delay to
> the user followed by a busy when the PRI finally gives up. If
> enabled, the PRI should generate a callTermFailedEvent trap for
> this call.
>
> * The modem is sort of ready but not fully - it acknowledges the PRI
> indication, but when it gets ahold of the call can't properly
> process it and the user perceives dead air. If enabled, the modem
> should generate a DteRingNoAns trap for this call.
>
>You can also see the counts of these sorts of failures on some of the
>PRI debug summary screens for modem errors.
>
>Why this tends to happen more on PRI than CT1 is probably a
>combination of the faster and out-of-band signaling for call setup and
>teardown, and the way in which the PRI communicates to modems OOB over
>the packet bus for call setup as opposed to using TDM bus patterns for
>the CT1 code, both of which affect the general setup and teardown
>timing sequence during call processing.
>
>The reason that changing to round-robin allocation "fixes" this
>problem (in many cases) is that the newer call is not going to try to
>hit the same modem, but one of the two "idle" modems (on a T1). So
>instead of a limiting factor of the reset time of a single modem for
>your call cycle time, you have divided that time by 3 (worst case).
>Normally unless you lose several calls simultaneously, this is enough
>to avoid the problem.
>
>Having the telco distribute calls over your entire hunt group in a
>statistical fashion (or even round-robin) can help too, but not once
>the hunt group gets full or very busy since it's still likely to pick
>the B channel just released for the next call.
>
>The round robin selection also isn't a possibility in typical E1
>configurations since in that case you don't have the automatic buffer
>modems.
>
>We had terrible problems with this at our E1 sites, and actually had
>to remove 4 B channels from service to give ourselves enough of a
>"buffer" to handle the call load.
>
>The "fix" (which doesn't require round-robin) is to both have the
>modem reset more quickly (at least to the point where it can start
>accepting new call information) as well as have the PRI wait for the
>modem to say it is done before acknowledging the release message from
>the telco. True, the PRI does have some hard timing limits within
>which it must acknowledge the message according to the D channel
>protocol, but there's room in there to wait for the modem in most
>cases.
The "fix" may not be to wait for software features. Although the suggestion
you mention above sounds like it would work better than what I'm going to
suggest. The amount of time it takes to the telco to reset their side can
be tailored to your specifications depending on the equipment the telco
uses. Try experimenting with it by having the time to reset changed to
250ms. I believe the telco can adjust up to 1200ms. I may be wrong.
>
>I can't speak to the general availability of such a fix or not, but it
>is possible. Of course, switching to round robin (for most T1
>configurations) is generally pretty effective too.
>
>-- David
>
>/-----------------------------------------------------------------------\
> \ David Bolen \ Internet: db3l@ans.net /
> | ANS Communications, Inc. \ Phone: (914) 701-5327 |
> / 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
>\-----------------------------------------------------------------------/
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) Solution for idle time limits and dual logins From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-19 11:03:09
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jeff Mcadams
>Sent: Wednesday, November 18, 1998 7:07 PM
>To: usr-tc@lists.xmission.com
>Subject: Re: (usr-tc) Solution for idle time limits and dual logins
>
>
>Thus spake Mike Wronski
>>|Oh, so if you can't provide a *perfect* tool, then you won't give us
>>|*any* tools to work with at all. Thanks a lot there!
>
>>I don't provide anything but support for the product. People at
>higher levels
>>have to wade through all the possible features and some fall out
>in favor of
>>others.
>
>Oh, believe me, I understand that...but please understand that we've
>gotten the run-around on features before. :)
>
>>Its just not possible to make a product everyone is in love with.. I
>>don't know for certain, but do any other vendors have this feature? I
>>don't think
>>we are unique here.
>
>I've not seen any NAS vendors that do, but as I mentioned previously,
>I've used a pppd (Morningstar) that had this feature and it was a
>*wonderful* feature to have.
>
>>Agreed, but what is the alternative to RADIUS? TACACS? Is MS
>Anything perfect?
>>nope. What do most people use??
>>With that logic: I think LINUX is better than MS. Should I not
>support MS even
>>though its the defacto standard.
>
>You, of course, have a point...RADIUS is the best thing out there for
>doing this... However, you see my point as well, right? We, the
>customers need tools to be able to make our networks work the best we
>can, and we're getting you (and I don't mean this to be a slam on you
>personally...I'm *glad* you're responding to these postings) telling us
>that our desires for features that we want are wrong. :) You can see
>how that would be a little bit grating. :)
>
>As I said...I've used this feature before, on a UNIX based pppd and it
>was wonderful to have...I'd like to be able to do this on a NAS too. :)
>
>>I now expect a jump off the bridge reply from someone.. And
>>"No I wouldn't do it just because all my friends did! :)"
>
>Blah...I've always thought that was a stupid argument. :)
>
>>|There are plenty of features that you could develop that would benefit
>>|us greatly...let's come up with a list and I'll rank them how I feel
>>|they would be most useful to us. I guarantee you that idle-timer filter
>>|list would be *very* high up there.
>
>>You should do that. Put don't post it here. It is far more effective when
>>submitted formally as a Feature Request/CRA and submitted by many
>people.
>
>Hey, I'm more than happy to submit it through formal channels...but
>let's also consider this as a possible discussion ground for what those
>features that we request might be. Perhaps we could all work on a list
Absolutely. Lets get a list going on features we would like to see and get
it into the hands of someone who can make a decision. If it's a money
issue, I'm sure most customer's would pay a "little" more for a feature rich
RADIUS.
>together and could use that as a base for features that we'd all like to
>see put in, and possibly augment that with our own individual requests.
>:)
>
>>(OSPF flame expected here.. Don't
>>bother! I know its been a long time coming...But.. its
>coming...Really....No
>>fooling...Its not vaporware...)
>
>Heh...wasn't gonna be a flame, but it *was* gonna get mentioned. :) I
>know its coming...is it 4.1.72 perchance? I thought I had heard some
>talk of it being available in ER's/SR's, but perhaps I'm just imagining
>things.
>
>>I do my best to respond to issues on this list and have no desire
>to fight with
>>anyone over feature sets..
>
>Please, your input is greatly appreciated...and you pointed out a very
>real limitation on why this would be a difficult thing to implement
>(idle-timer is reset before the IP header is even parsed)...that's very
>understandable. At least when I go talk to Tom Goodman I'll understand
>if it doesn't show up in the very next release. :)
>--
>Jeff McAdams Email: jeffm@iglou.com
>Head Network Administrator Voice: (502) 966-3848
>IgLou Internet Services (800) 436-4456
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) Corresponding modem is unavailable From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-19 11:17:28
What slot is your T1 card in? You may have some troubleshooting ahead. Try
pulling every card out of the chassis except 1 quad and the t1 and try it
again. If it maps correctly put one card at a time back into the chassis
until it stops working. Once it stops working you have probably found a bad
card
========================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= Your Remote Access and Security Experts! =
= Firm partnerships est. with current and emerging leaders of today's =
= technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
= Serving North America and parts of Canada. Reselling to the world. =
========================================================================
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Curt Shambeau
>Sent: Thursday, November 19, 1998 10:13 AM
>To: usr-tc@lists.xmission.com
>Subject: RE: (usr-tc) Corresponding modem is unavailable
>
>
>Already checked. I'm running Ch.T1's, and it's set to the T1tdm bus.
>
>--------------
>
>> Not sure if someone answered this for you yet.
>> I have seen this many times. Are you running PRI or T1? Check the Line
>> Interface Options/Line Interface source. Make sure it is set to T1tdm or
>> PRItdm depending on what kind of circuit you have. Save it to NVRAM and
>> reboot the quads
>>
>> ========================================================================
>> = Brian K. McIntire bmcintire@commnet.com =
>> = Systems Engineer III 317-558-5050 x113 =
>> = CommNetPlus, Indianapolis, IN http://www.commnet.com =
>> = Your Remote Access and Security Experts! =
>> = Firm partnerships est. with current and emerging leaders of today's =
>> = technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
>> = Serving North America and parts of Canada. Reselling to the world. =
>> ========================================================================
>>
>>
>> >-----Original Message-----
>> >From: owner-usr-tc@lists.xmission.com
>> >[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Curt Shambeau
>> >Sent: Wednesday, November 18, 1998 1:46 PM
>> >To: USR Total Control Mailing list
>> >Subject: (usr-tc) Corresponding modem is unavailable
>> >
>> >
>> >
>> >Quad modem chassis running 4.2.1 on T1 card (running channelized T1's)
>> >modem cards running 5.10.9, HiPer ARC 4.0.29, NMC running 5.5.5
>> >
>> >When doing a DS0 performance monitor, the DS0 channels show
>idle, but the
>> >modem connected to the DS0 status is "corresponding modem is
>unavailable."
>> >
>> >I've tried restoring the chassis from defaults, and resetting
>everything,
>> >to no avail. Has anyone seen this before??
>> >
>>
>>--------------------------------------------------------------------------
>> >| Curtis V. Shambeau | curt@execpc.com |
http://www.execpc.com/~curt |
> >| Executive Vice President - Exec-PC, Inc.
|
>
>--------------------------------------------------------------------------
> >
> >
> >-
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
| Curtis V. Shambeau | curt@execpc.com | http://www.execpc.com/~curt |
| Executive Vice President - Exec-PC, Inc. |
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:Re: (usr-tc) Interested in Linux on the Netserver? From: Matthew Opoka <phantom@magnolia.net> Date: 1998-11-19 11:17:50
I tell you what.
I get better quake on a livingston portmaster 2e and an anolog modem
attach than I do on a netserver 486 with pri lines. The portmaster has
an AMD 386 DX40.
Jeff Mcadams wrote:
>
> Thus spake Ricky Beam
> >Allen Marsalis was heard to say:
> >>Lets say you get USR to cooperate (1st feat). Then you get linux up
> >>on the netserver (2nd feat). What if Quake still lags? I mean Quake
> >>Lag is a symptom of a much bigger problem. And it might be hardware
> >>related and that has been hinted at.
>
> >A packet is a packet. I don't buy the "but it's the hardware" excuse.
>
> See my other thread with Mike Wronski about the specifics of the
> problem...a packet is not just a packet...there are other issues to deal
> with...most significant in this situation is that its a lot of small
> packets. That is very difficult for a router to deal with. Yes, a
> 486/100 *should* be able to handle 48 ports of this if the code were
> decent (check out the cisco 2500's...they easily handle this much
> traffic with a much less powerful cpu), but it may be so insideous that
> its easier just to scrap the whole ComOS code base and do their own...as
> they've done with Pilgrim. That, however, doesn't explain the
> collosally stupid decision *not* to support Pilgrim on the
> NETServer...that is, unless the Pilgrim code is as inefficient in packet
> handling, so it would run into the same problems that the NETServer does
> in this case. I doubt that's the case though since the ARC's seem to be
> handling many DSP's without problem from reports that I've heard.
> --
> Jeff McAdams Email: jeffm@iglou.com
> Head Network Administrator Voice: (502) 966-3848
> IgLou Internet Services (800) 436-4456
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Laszlo Vecsey
|Sent: Thursday, November 19, 1998 11:08 AM
|To: usr-tc@xmission.com
|Subject: (usr-tc) hiperarc NAS-Port-Id
|
|
|With 4.1.72 on the hiperarc, theres an improvement with the NAS-Port-Id
|output (used to all be set to 1), but I think theres still a glitch:
|
| NAS-Port-Id = 1539
| NAS-Port-Id = 1028
| NAS-Port-Id = 1793
| NAS-Port-Id = 1026
| NAS-Port-Id = 514
| NAS-Port-Id = 1793
| NAS-Port-Id = 1794
| NAS-Port-Id = 912469510
| NAS-Port-Id = 1795
| NAS-Port-Id = 1796
| NAS-Port-Id = 258
| NAS-Port-Id = 2049
|
Can you show the entire record the Odd entry came from??
GNU GREP with -A 3 -B 3 ???
-M
Subject:RE: (usr-tc) Interested in Linux on the Netserver? From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-19 11:24:08
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Matthew Opoka
|Sent: Thursday, November 19, 1998 11:18 AM
|To: usr-tc@lists.xmission.com
|Subject: Re: (usr-tc) Interested in Linux on the Netserver?
|
|
|I tell you what.
|I get better quake on a livingston portmaster 2e and an anolog modem
|attach than I do on a netserver 486 with pri lines. The portmaster has
|an AMD 386 DX40.
How many ports??
-M
With 4.1.72 on the hiperarc, theres an improvement with the NAS-Port-Id
output (used to all be set to 1), but I think theres still a glitch:
NAS-Port-Id = 1539
NAS-Port-Id = 1028
NAS-Port-Id = 1793
NAS-Port-Id = 1026
NAS-Port-Id = 514
NAS-Port-Id = 1793
NAS-Port-Id = 1794
NAS-Port-Id = 912469510
NAS-Port-Id = 1795
NAS-Port-Id = 1796
NAS-Port-Id = 258
NAS-Port-Id = 2049
That was a sample grep from my detail file. Every now and then theres an
oddball entry..
?
Subject:Re: (usr-tc) RADIUS 6.0.8: Authentication into NT Domain creates errors From: Dominic Ciresi <dciresi@defunct.ae.usr.com> Date: 1998-11-19 13:49:57
Ralph,
There is as yet no way to uncheck the script debugging feature in
SAS 6.0.8. The workaround is to edit your \<windows>\radius.ini,
and change the value for "ScriptLog" to 0. Unfortunately, you will
have to do this every time before restarting the server.
Dominic
On Tue, 17 Nov 1998, Ralph Helfenberger wrote:
> Hi all
>
> I try to setup 3COM RADIUS Server to authenticate all users, wich are
> not found in the RADIUS database, in the NT Domain. For that reason I'm
> working with the template NTUSER. The problem I encounter: There are
> script errors in the logfile. Has anyone done something similar?
>
> RADIUS 6.0.8
>
> Script Section: CHECK-PASSWORD nLine: 534 bIgnore: 0
> Script Section: VALIDATE-LOCALPASSWD nLine: 3345 bIgnore: 0
> Trace: Attempting User: Tinu-H, Domain: LEO-DOMAIN [11 17 18:44:40]
> Script Section: CHECK-EXPIRATION nLine: 535 bIgnore: 0
> Script Section: CHECK-CALL-SOURCE nLine: 542 bIgnore: 0
> Script Section: CHECK-TUNNEL-NAME nLine: 547 bIgnore: 0
> Script error on line 1781 :Variable (THISGROUP) not found
> Line is : if(length(thisGroup)>0)
>
> Script error on line 1781 :Parameter has a bad value
> Line is : if(length(thisGroup)>0)
>
> Script Section: COLLECT-USER-RESPONSE nLine: 549 bIgnore: 0
> Script Section: GET-STANDARD-ATTRIBUTES nLine: 671 bIgnore: 0
> Script Section: GET-USER-SERVICE-TYPE nLine: 7257 bIgnore: 0
> Script Section: GET-STANDARD-FRAMED-ATTRIBUTES nLine: 7276 bIgnore: 0
> Script Section: GET-FRAMED-PROTOCOL nLine: 7319 bIgnore: 0
> Script error on line 2696 :Variable (THISGROUP) not found
> Line is : if(length(thisGroup)>0)
>
> Script error on line 2696 :Parameter has a bad value
> Line is : if(length(thisGroup)>0)
>
> Script Section: GET-FRAMED-ADDRESS nLine: 7320 bIgnore: 0
> Script error on line 2753 :Variable (THISGROUP) not found
> Line is : if(length(thisGroup)>0)
>
> Script error on line 2753 :Parameter has a bad value
> Line is : if(length(thisGroup)>0)
>
> Script error on line 2769 :Variable (THISGROUP) not found
> Line is : if(length(thisGroup)>0)
>
> Script error on line 2769 :Parameter has a bad value
> Line is : if(length(thisGroup)>0)
>
> Script Section: GET-FRAMED-NETMASK nLine: 7321 bIgnore: 0
> Script error on line 2785 :Variable (THISGROUP) not found
> Line is : if(length(thisGroup)>0)
>
> and so on....
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
We are looking into issues with 1.2.68 (and .67, .66, .65, and .64)
causing problems on some T3 muxes. There is not an issue on 1.2.5 or
engineering releases numbered above 1.2.68. It also does not appear to be
an issue in E1 codebases.
The issue seems contained to certain muxes, but is serious enough when it
does happen, that I highly recommend not using those releases on any
chassis that don't already have it until we completely nail this issue and
issue a patch.
If you are already using one of them, and are not experiencing
problems, there is
no need to back them off, this is not a problem that creeps up. Your
spans either have a problem with it, or they don't.
So, if you have 1.2.68 (that one is out there in large quantities), please
use great care and I do not suggest you expand its' use at all.
Regards,
JP
Subject:Re: (usr-tc) Interested in Linux on the Netserver? From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-19 14:07:48
Thus spake Mike Wronski
>|I tell you what.
>|I get better quake on a livingston portmaster 2e and an anolog modem
>|attach than I do on a netserver 486 with pri lines. The portmaster has
>|an AMD 386 DX40.
>How many ports??
A 2e has 30.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:(usr-tc) padding on trunk groups From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-19 16:09:13
Please forgive the following "book"
I'm faced with a problem. I'm working for a customer in an Ameritech
region. When ever I place a call with my courier to a PRI I see a 6db pad
being applied to my receive. (The PRI's transmit) I've been told there has
to be a 6db pad applied because Ameritech has not installed echo
cancellation in their telephone network. Makes some sense. I have further
been told echo cancellation is not currently accepted as a standard but
where ever it has been installed telephone companies remove the 6db pad my
client modem see's on the receive.
I understand there needs to be echo cancellation to go from a 4 to 2 wire.
(PRI to analog) If not there needs to be a pad. Does anyone have a
different opinion or know something different. Is there a way around this?
(Probably not) If not does anyone know if using echo cancellation will
become a standard?
The bottom line is, I have some customer's that connect at higher speeds
than others because of who their local telephone company is. The PRI I'm
working on is set up to be a local # for several cities with different
switches. One of the cities that dial this PRI applies no padding. The
others do. The problem is my customer's main competition is centrally
located in the local loop of the telco that does not pad. His customer's
are being called by his competition and given a chance to call a non-padded
PRI for free to prove their connect speeds will be better and steal
customers. Of course, connect speeds are always better when you compare
padded against non-padded.
Any positive ideas?
By the way, here's an interesting quirk. When I dial the PRI that goes to
our internal lab chassis from a trunk off a non-padded T1 being provided by
the same telco I see no pad and connect at 50666. When I dial AOL's local #
(PRI's being provided by the same telco) I see 6db padding and connect at
49333 or 48000.
When I dial either from home (from a different phone company "Ameritheft") I
connect at 45333 or lower. Sometimes I don't even connect at v.90. I see
37333 quite often.
David, do you have any comments on this. The AOL chassis here are yours.
========================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= Your Remote Access and Security Experts! =
= =
= Our experienced technicians can design and install your high speed =
= network for you. From Operating Systems and servers to routers and =
= firewalls nobody does it better than CommNetPlus. Our Technical =
= Support staff is available to you 24 hours a day to meet your needs.=
= Offering remote management and monitoring packages for dedicated =
= clients. (Specializing in ISP's). "Let us do the work for you!" =
= =
= Firm partnerships est. with current and emerging leaders of today's =
= technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
= Serving North America and parts of Canada. Reselling to the world. =
========================================================================
Subject:(usr-tc) FYI: 1.2.66 & PRI errors From: Jerry Kalligonis <jerryk@blazenet.net> Date: 1998-11-19 16:59:10
I just wanted to share a fun <grin> day I had with my DSP cards yesterday.
At one particular location I have a good many PRI's. I was running the
1.2.5 code, but had heard good things about the newer patch from other
posts on this list.
I upgraded 1 of my chassis' and tested @12am and everything seemed to be
OK. After finishing the rest of the chassis' I did some testing and
everything seemed fine. About 1 half hour later ALL of my PRI's appeared
to drop. I immediately called the telco and they reported to me that the
T3 that feeds us our PRI's had dropped and they were looking into it.
I pulled 5 of my PRI's out of a chassis and the T3 came back up. I put
them back in and it took the T3 down.
Anyway, after hours of trouble shooting and no help from 3com, we found
that the new code puts out a certain level of errors onto the PRI line.
When enough of these PRIs are combined the error level is enough to bring
down a T3.
Of course later that afternoon I got a call back from 3com saying, sorry I
couldn't get back to you this morning, but engineering says there may be a
problem with this code that puts out enough errors to take down a T3. Call
back to get DSP code 1.2.64
I haven't gotten the code yet, but I'll pass on the info when I try it....
Jerry Kalligonis
Brian K McIntire writes...
>The difference I'm referring to is not just in performance but reliability
>and sometimes, on ease of turn up. On PRI's you do run this risk of losing
>an entire span whenever a d-cannel goes down. Just another argument for
>Backup D.
This is stupid.
The probability of losing a D channel is miniscule compared to losing the
span to something like a looped repeater. When this happens having a backup
D channel will make you very, very sorry as your calls dissapear down
a black hole.
-a
Jerry Kalligonis writes...
>At one particular location I have a good many PRI's. I was running the
>1.2.5 code, but had heard good things about the newer patch from other
>posts on this list.
>
> . . .
>
>Anyway, after hours of trouble shooting and no help from 3com, we found
>that the new code puts out a certain level of errors onto the PRI line.
>When enough of these PRIs are combined the error level is enough to bring
>down a T3.
>
>Of course later that afternoon I got a call back from 3com saying, sorry I
>couldn't get back to you this morning, but engineering says there may be a
>problem with this code that puts out enough errors to take down a T3. Call
>back to get DSP code 1.2.64
You didn't mention what version the "newer patch" was that cause these
problems for you.
I was having trouble with false alarms coming from my DSP cards which
version 1.2.68 fixed.
--
Aaron Nabil
Subject:RE: (usr-tc) Interested in Line on the Netserver? From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-19 17:52:20
The fact that the 2e supports only 30 may partially explain how it can
handle things a little better. It doesn't have to do the work the USR
NetServer does
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jeff Mcadams
>Sent: Thursday, November 19, 1998 1:08 PM
>To: usr-tc@lists.xmission.com
>Subject: Re: (usr-tc) Interested in Linux on the Netserver?
>
>
>Thus spake Mike Wronski
>>|I tell you what.
>>|I get better quake on a livingston portmaster 2e and an anolog modem
>>|attach than I do on a netserver 486 with pri lines. The portmaster has
>>|an AMD 386 DX40.
>
>>How many ports??
>
>A 2e has 30.
>--
>Jeff McAdams Email: jeffm@iglou.com
>Head Network Administrator Voice: (502) 966-3848
>IgLou Internet Services (800) 436-4456
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) Re: your mail From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-19 17:56:35
Thanks for the info Dominic. I upgraded to the code you specified. This
time it doesn't work at all. Instead of seeing the error message incorrect
passcode the RADIUS isn't even communicating with Secure ID. Any other
Ideas. Debug was useless. All it showed was requests going out that were
un- answered. Again, when I down graded back to 5.5.4 it worked like a
charm.
========================================================================
= Brian K. McIntire bmcintire@commnet.com =
= Systems Engineer III 317-558-5050 x113 =
= CommNetPlus, Indianapolis, IN http://www.commnet.com =
= Your Remote Access and Security Experts! =
= Firm partnerships est. with current and emerging leaders of today's =
= technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
= Serving North America and parts of Canada. Reselling to the world. =
========================================================================
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Dominic Ciresi
>Sent: Thursday, November 19, 1998 9:47 AM
>To: usr tc
>Subject: (usr-tc) Re: your mail
>
>
>Brian,
>This is a known bug. There is a fix in SAS 6.0.92.(Solaris) or SAS 6.0.91
>(HP). Open a ticket and I will deliver code to you.
>
>Dominic
>
>On Thu, 19 Nov 1998, Brian K McIntire wrote:
>
>> This is directed to anyone out there who may be using Secure-ID
>with TC S/A
>> for Unix. We have installed S/A 5.5.4 and secure-id works fine.
> With the
>> same config on 6.0.7 it doesn't. The debug is useless. All it says is
>> invalid pass code. We down grade back to 5.5.4 it works again.
>Any ideas?
>> Is this a known bug?
>>
>>
>> -
>> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
>> with "unsubscribe usr-tc" in the body of the message.
>> For information on digests or retrieving files and old messages send
>> "help" to the same address. Do not use quotes in your message.
>>
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) PPTP on HiPer ARC From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-19 18:02:22
I'm trying to set up PPTP on a HiPer ARC. The Tunnel will be going to an NT
SERVER that is used for PPTP from several NETServers. The NETServers work
fine. I followed the instructions outlined in the HiPer ARC manual. When I
list PPTP tunnels I don't see anything. I'm sure many of you have done
this. Can you tell me precisely what steps I need to take to configure this
properly?
thanks
Subject:Re: (usr-tc) Re: your mail From: Ricky Beam <jfbeam@interpath.net> Date: 1998-11-19 18:07:09
Brian K McIntire was heard to say:
>Thanks for the info Dominic. I upgraded to the code you specified. This
>time it doesn't work at all. Instead of seeing the error message incorrect
>passcode the RADIUS isn't even communicating with Secure ID. Any other
>Ideas. Debug was useless. All it showed was requests going out that were
>un- answered. Again, when I down graded back to 5.5.4 it worked like a
>charm.
Ok, so try this...
copy the bin/session from 5.5.4 to bin/saworker for 6.0.whatever and see
if that makes any difference at all. That's what the server runs to handle
authenticating with ACE. As far as I'm aware, other than the name change,
there is no difference between the two?
--Ricky
"Brian K McIntire" <bmcintire@commnet.com> writes:
> The difference I'm referring to is not just in performance but reliability
> and sometimes, on ease of turn up.
Oh, I thought the topic was "V.90 connect problems", but the points
are pretty much the same for reliability too.
> Fact is, in
> my experience, there are considerably more things the telco could screw up
> on a T1 than on a PRI.
All I've offered is some counterpoints from my own experience (with
tens of thousands of circuits all over the world of all types). It
may not agree with your experience, and that's fine, but would you
disagree that it should be just as valid a set of experience for the
purposes of discussion?
I agree that telcos can and do foul things up and it's generally more
likely to occur during initial setup than any other stage of the
process. With that said, I can't honestly say that we get more foul
ups with CT1 configurations than with PRI - ignoring for the moment
ground/loop start CT1, which I don't recommend people order (and I
don't consider such a caveat beyond the average ISP engineer).
BTW, it's fairly common for misconfigurations during initial PRI
setups when it comes to configuring the switch type. This is
primarily because we require custom configuration for service message
to work (not NI-2), but I have to believe that most providers want
service messages on PRI spans - how many know they must not get the
"national infrastructure" standard configuration if they want to be
able to take individual B channels out of service?
> >But an E&M configuration is almost always the opposite - digital
> >straight into the switch
>
> Again, assuming your telco knows what the hell they are doing.
Off the top of my head, I can't recall _ever_ getting an E&M
configuration with a channel bank and copper pairs. Interestingly
enough, we have occasionally seen a ground start going straight into a
switch (no channel bank), but not the reverse. This is with telcos
(LECs, and CAPs) all over the country.
> >Pedantic point - unless you are at an international site using E1's, a
> >"PRI" _is_ a "T1" fundamentally. Same framing. (...)
> >
> See above notes. Again, you are assuming I'm discussing nothing but
> throughput or connect speeds.
No, you missed my point. A PRI (domestically) _is_ a T1. Simple as
that. Now, you may be meaning "channelized T1 with xxx signaling"
when you write "T1", but I don't expect everyone to infer that.
A PRI circuit is simply a channelized T1 (or I suppose more
accurately, DS1) circuit that will have full 64Kbps clear channels and
use one of those channels as a D channel for OOB signaling. But the
physical circuit is still a T1 (and that's all the "term" T1 stands
for - an overall capacity of a digital span using a standard framing
mechanism).
As I said - pedantic - but I believe in being thorough in such
discussions.
> How often do you have to troubleshoot d-channel signaling?
If you include service messages (which are part of D channel
signaling, and which I've never had a problem with their equivalent on
E&M T1 circuits), then probably with a similar frequenty as we
troubleshoot signaling problems on our channelized T1 E&M circuits.
That is, not very much at all.
> However, you must remember, when you are posting to this list you are
> talking to a tremendous number of people no where near as informed as you.
Strange, and here I thought I was just offering a balanced opinion,
specifically in that vein :-)
> They experience different kinds of problems from sheer lack of experience
> and understanding. Again, I say, stop assuming your situation applies to
> everyone else. It doesn't!
Where did I? It seems that you offered some information based on your
experience, and I offered some based on mine. Just a balanced
discussion, which can only help readers of this list, no? Or are you
saying that your experience/situation is more applicable to everyone
else than mine?
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
Subject:RE: (usr-tc) using round robin & fixed assignment From: David Bolen <db3l@ans.net> Date: 1998-11-19 18:26:06
"Brian K McIntire" <bmcintire@commnet.com> writes:
> The "fix" may not be to wait for software features.
Well, the "fix" is definitely to fix the bug. But as you note there
are other "workarounds" that may be applicable.
> Although the suggestion
> you mention above sounds like it would work better than what I'm going to
> suggest. The amount of time it takes to the telco to reset their side can
> be tailored to your specifications depending on the equipment the telco
> uses. Try experimenting with it by having the time to reset changed to
> 250ms. I believe the telco can adjust up to 1200ms. I may be wrong.
It depends on the switch and software configuration. For our UK
problem, the maximum available by the telco was 250ms and we still
needed a 4-modem buffer to keep up adequately. But I agree that it is
something worth trying if you are on the margin in terms of timing.
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
Subject:Re: (usr-tc) padding on trunk groups From: David Bolen <db3l@ans.net> Date: 1998-11-19 18:30:40
"Brian K McIntire" <bmcintire@commnet.com> writes:
> Of course, connect speeds are always better when you compare
> padded against non-padded.
In general, only if it's analog padding. Digital padding is
automatically detected and accounted for in the 3Com x2/V.90
implementation, and I believe I've been told that holds true for V.90
for Rockwell/Lucent as well.
If it's analog though, then you're out of luck since the modem has to
treat it as equivalent to loss on the final leg. I don't know if you
could convince Ameritech to move from one type to the other or not.
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
Subject:RE: (usr-tc) v.90 connect problems From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-19 18:44:25
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of David Bolen
>Sent: Thursday, November 19, 1998 5:24 PM
>To: usr-tc@lists.xmission.com
>Subject: RE: (usr-tc) v.90 connect problems
>
>
>"Brian K McIntire" <bmcintire@commnet.com> writes:
>
>> The difference I'm referring to is not just in performance but
>reliability
>> and sometimes, on ease of turn up.
>
>Oh, I thought the topic was "V.90 connect problems", but the points
>are pretty much the same for reliability too.
>
>>
>Fact is, in
>> my experience, there are considerably more things the telco
>could screw up
>> on a T1 than on a PRI.
>
>All I've offered is some counterpoints from my own experience (with
>tens of thousands of circuits all over the world of all types). It
>may not agree with your experience, and that's fine, but would you
>disagree that it should be just as valid a set of experience for the
>purposes of discussion?
I agree
>
>I agree that telcos can and do foul things up and it's generally more
>likely to occur during initial setup than any other stage of the
>process. With that said, I can't honestly say that we get more foul
>ups with CT1 configurations than with PRI - ignoring for the moment
>ground/loop start CT1, which I don't recommend people order (and I
>don't consider such a caveat beyond the average ISP engineer).
>
>BTW, it's fairly common for misconfigurations during initial PRI
>setups when it comes to configuring the switch type. This is
>primarily because we require custom configuration for service message
>to work (not NI-2), but I have to believe that most providers want
>service messages on PRI spans - how many know they must not get the
>"national infrastructure" standard configuration if they want to be
>able to take individual B channels out of service?
Most do not
>
>> >But an E&M configuration is almost always the opposite - digital
>> >straight into the switch
>>
>> Again, assuming your telco knows what the hell they are doing.
>
>Off the top of my head, I can't recall _ever_ getting an E&M
>configuration with a channel bank and copper pairs. Interestingly
>enough, we have occasionally seen a ground start going straight into a
>switch (no channel bank), but not the reverse. This is with telcos
>(LECs, and CAPs) all over the country.
>
>> >Pedantic point - unless you are at an international site using E1's, a
>> >"PRI" _is_ a "T1" fundamentally. Same framing. (...)
>> >
>> See above notes. Again, you are assuming I'm discussing nothing but
>> throughput or connect speeds.
>
>No, you missed my point. A PRI (domestically) _is_ a T1. Simple as
>that. Now, you may be meaning "channelized T1 with xxx signaling"
>when you write "T1", but I don't expect everyone to infer that.
>
>A PRI circuit is simply a channelized T1 (or I suppose more
>accurately, DS1) circuit that will have full 64Kbps clear channels and
>use one of those channels as a D channel for OOB signaling. But the
>physical circuit is still a T1 (and that's all the "term" T1 stands
>for - an overall capacity of a digital span using a standard framing
>mechanism).
>
>As I said - pedantic - but I believe in being thorough in such
>discussions.
>
>> How often do you have to troubleshoot d-channel signaling?
>
>If you include service messages (which are part of D channel
>signaling, and which I've never had a problem with their equivalent on
>E&M T1 circuits), then probably with a similar frequenty as we
>troubleshoot signaling problems on our channelized T1 E&M circuits.
>That is, not very much at all.
>
>> However, you must remember, when you are posting to this list you are
>> talking to a tremendous number of people no where near as
>informed as you.
>
>Strange, and here I thought I was just offering a balanced opinion,
>specifically in that vein :-)
;)
>
>> They experience different kinds of problems from sheer lack of experience
>> and understanding. Again, I say, stop assuming your situation applies to
>> everyone else. It doesn't!
>
>Where did I? It seems that you offered some information based on your
>experience, and I offered some based on mine. Just a balanced
>discussion, which can only help readers of this list, no? Or are you
>saying that your experience/situation is more applicable to everyone
>else than mine?
Definitely not. In fact, based on what I know of you and have seen from you
on this list I think the people on this list have more to learn from you
than me.
I do agree, all things being equal, a T1 should be able to perform almost or
as well as a PRI.
I also agree, PRI's are a little more involved when it comes to making sure
it is configured the way you need it configured. ie. signaling.
In my experience PRI's are easier to work with telco's to resolve problems
on. Just because you request E&M type II doesn't mean the guy at the telco
who set up your line is running the damn thing through a channel bank. I
can't tell you how many times I had to work with a telco to resolve problems
with multiple codecs in the line. Never had to do that with a PRI.
I respect your opinion very much. My experience has proven "to me" my
original opinion is sound.
thanks again
>
>-- David
>
>/-----------------------------------------------------------------------\
> \ David Bolen \ Internet: db3l@ans.net /
> | ANS Communications, Inc. \ Phone: (914) 701-5327 |
> / 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
>\-----------------------------------------------------------------------/
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) using round robin & fixed assignment From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-19 18:48:40
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of David Bolen
>Sent: Thursday, November 19, 1998 5:26 PM
>To: usr-tc@lists.xmission.com
>Subject: RE: (usr-tc) using round robin & fixed assignment
>
>
>"Brian K McIntire" <bmcintire@commnet.com> writes:
>
>> The "fix" may not be to wait for software features.
>
>Well, the "fix" is definitely to fix the bug. But as you note there
>are other "workarounds" that may be applicable.
>
Unless the software was originally designed to do what you asked it can not
be a bug. It can only be a feature to be added later.
>> Although
>the suggestion
>> you mention above sounds like it would work better than what I'm going to
>> suggest. The amount of time it takes to the telco to reset
>their side can
>> be tailored to your specifications depending on the equipment the telco
>> uses. Try experimenting with it by having the time to reset changed to
>> 250ms. I believe the telco can adjust up to 1200ms. I may be wrong.
>
>It depends on the switch and software configuration. For our UK
>problem, the maximum available by the telco was 250ms and we still
>needed a 4-modem buffer to keep up adequately. But I agree that it is
>something worth trying if you are on the margin in terms of timing.
>
>-- David
>
>/-----------------------------------------------------------------------\
> \ David Bolen \ Internet: db3l@ans.net /
> | ANS Communications, Inc. \ Phone: (914) 701-5327 |
> / 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
>\-----------------------------------------------------------------------/
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
I am assuming hd102682.zip is effected as well? (must have been a .68
variant)
On Mon, 19 Nov 2018, John Powell wrote:
> We are looking into issues with 1.2.68 (and .67, .66, .65, and .64)
> causing problems on some T3 muxes. There is not an issue on 1.2.5 or
> engineering releases numbered above 1.2.68. It also does not appear to be
> an issue in E1 codebases.
>
> The issue seems contained to certain muxes, but is serious enough when it
> does happen, that I highly recommend not using those releases on any
> chassis that don't already have it until we completely nail this issue and
> issue a patch.
>
> If you are already using one of them, and are not experiencing
> problems, there is
> no need to back them off, this is not a problem that creeps up. Your
> spans either have a problem with it, or they don't.
>
> So, if you have 1.2.68 (that one is out there in large quantities), please
> use great care and I do not suggest you expand its' use at all.
>
> Regards,
>
> JP
>
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:RE: (usr-tc) using round robin & fixed assignment From: David Bolen <db3l@ans.net> Date: 1998-11-19 19:00:11
"Brian K McIntire" <bmcintire@commnet.com> writes:
> Unless the software was originally designed to do what you asked it can not
> be a bug. It can only be a feature to be added later.
Yeah, but that's a real reach in this case - it implies that the
original design of the software was intended to fail to accept calls
in what would be considered a normal operating mode. It's hard to
look at that as a sane design by any standard.
It's a bug. Me customer, it bug. I don't buy the "it's not a bug,
it's a feature" line. :-)
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
On Thu, 19 Nov 1998, Mike Wronski wrote:
> |With 4.1.72 on the hiperarc, theres an improvement with the NAS-Port-Id
> |output (used to all be set to 1), but I think theres still a glitch:
> |
> | NAS-Port-Id = 1539
> | NAS-Port-Id = 1028
> | NAS-Port-Id = 1793
> | NAS-Port-Id = 1026
> | NAS-Port-Id = 514
> | NAS-Port-Id = 1793
> | NAS-Port-Id = 1794
> | NAS-Port-Id = 912469510
> | NAS-Port-Id = 1795
> | NAS-Port-Id = 1796
> | NAS-Port-Id = 258
> | NAS-Port-Id = 2049
>
> Can you show the entire record the Odd entry came from??
> GNU GREP with -A 3 -B 3 ???
>
I did some more greps and found that the oddball entries always have
Acct-Multi-Session-Id entries, as the following full example shows, its
complete except for the IP and username etc info that I've stripped out.
--
Thu Nov 19 18:18:18 1998
Acct-Status-Type = Start
Acct-Session-Id = "3967"
Acct-Delay-Time = 0
Acct-Authentic = RADIUS
Service-Type = Framed-User
NAS-Port-Type = Async
NAS-Port-Id = 1537
USR-Modem-Training-Time = 23
USR-Interface-Index = 2793
USR-Chassis-Call-Slot = 7
USR-Chassis-Call-Span = 1
USR-Chassis-Call-Channel = 1
USR-Unauthenticated-Time = 9
USR-Modulation-Type = v34
USR-Simplified-MNP-Levels = 13
USR-Simplified-V42bis-Usage = ccittV42bis
USR-Connect-Speed = 26400_BPS
Framed-Protocol = PPP
USR-MP-MRRU = 0
Acct-Link-Count = 1
Acct-Multi-Session-Id = 875639045
NAS-Port-Id = 825242721
Timestamp = 911517498
Request-Authenticator = Unverified
--
Notice the NAS-Port-Id is listed twice! Up near the top it looks correct,
down at the bottom it seems to contain an Acct-Multi-Session-Id number by
mistake.
This is affecting cistron radius in particular, the radwho program.. which
I suppose could be recoded to use the first NAS-Port-Id, but shouldn't
there only be one anyway?
-L
John Powell writes...
>We are looking into issues with 1.2.68 (and .67, .66, .65, and .64)
>causing problems on some T3 muxes. There is not an issue on 1.2.5 or
>engineering releases numbered above 1.2.68. It also does not appear to be
>an issue in E1 codebases.
>
>The issue seems contained to certain muxes, but is serious enough when it
>does happen, that I highly recommend not using those releases on any
>chassis that don't already have it until we completely nail this issue and
>issue a patch.
Certain muxes as in particular muxes at certain customer sites, or certain
muxes as in Telco Systems, Nortel, AT&T, CAC, etc?
I'd like to reiterate that 1.2.68 was the first release to solve some
intractible T1 error problems for me. I hope these two aren't
related.
-a
David Bolen writes...
>Aaron Nabil <nabil@spiritone.com> writes:
>
>> The probability of losing a D channel is miniscule compared to losing the
>> span to something like a looped repeater. When this happens having a backup
>> D channel will make you very, very sorry as your calls dissapear down
>> a black hole.
>
>I think it depends on the number of spans you are discussing.
>
>When concerned with a single span, your odds tend to be greater of
>losing the entire span (D channel included) than just the D channel.
>Very little point to having a backup.
>
>However, once you start involving multiple spans sharing a D channel
>with NFAS, then it's very important to have at least one backup D
>channel since even if the odds favor losing the entire span containing
>your D channel, you don't want to affect all of the other spans
>depending on it.
If you are running NFAS, by all means use a backup D channel. A much,
much better solution is never to use NFAS.
Fate sharing good. NFAS bad.
-a
Subject:RE: (usr-tc) I'm about to pull my hair out From: David OBrien <growler@ac.net> Date: 1998-11-19 20:00:47
Okay i have set up ummm lets say 24 modems (6 quad digital cards 2059
chassis) got all the settings just like they should be to accept calls from
the CT1 saved it to NVRAM saved it again for the heck of it reset the
modems call is accepted when call is done the modem reverts back to the
settings before i saved to nvram (partiuclarly nic instead of t1 and mf
tones instead of dtmf tone) and the modem will not accept another call. so
I have been sitting here switching them back after each call. How can I
make it stick. I also have the reset method via the atNVram thingy.
can maybe someone help me?
-Dave
Subject:(usr-tc) I'm about to pull my hair out From: David OBrien <growler@ac.net> Date: 1998-11-19 20:25:52
sheesh where'd the RE: come from
anyways
Okay i have set up ummm lets say 24 modems (6 quad digital cards 2059
chassis) got all the settings just like they should be to accept calls from
the CT1 saved it to NVRAM saved it again for the heck of it reset the
modems call is accepted when call is done the modem reverts back to the
settings before i saved to nvram (partiuclarly nic instead of t1 and mf
tones instead of dtmf tone) and the modem will not accept another call. so
I have been sitting here switching them back after each call. How can I
make it stick. I also have the reset method via the atNVram thingy.
can maybe someone help me?
-Dave
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Aaron Nabil <nabil@spiritone.com> writes:
> The probability of losing a D channel is miniscule compared to losing the
> span to something like a looped repeater. When this happens having a backup
> D channel will make you very, very sorry as your calls dissapear down
> a black hole.
I think it depends on the number of spans you are discussing.
When concerned with a single span, your odds tend to be greater of
losing the entire span (D channel included) than just the D channel.
Very little point to having a backup.
However, once you start involving multiple spans sharing a D channel
with NFAS, then it's very important to have at least one backup D
channel since even if the odds favor losing the entire span containing
your D channel, you don't want to affect all of the other spans
depending on it.
It's probably not worth it at 2 spans (where doing so negates the
benefit of NFAS in the first place of getting you back an extra
channel for a user), but I'd say it's clearly necessary as part of a
good design at higher numbers of spans.
Yes, there is some risk that the failure mode of the span containing
your primary D channel will be such that the telco still perceives it
as viable for calls (although if service messages are active you
should be able to proactively request it be taken out of service, even
if with some delay after the failure), there are other failure modes
where that will not be the case.
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
Subject:Re: (usr-tc) I'm about to pull my hair out From: David Bolen <db3l@ans.net> Date: 1998-11-19 20:41:27
David OBrien <growler@ac.net> writes:
> Okay i have set up ummm lets say 24 modems (6 quad digital cards 2059
> chassis) got all the settings just like they should be to accept calls from
> the CT1 saved it to NVRAM saved it again for the heck of it reset the
You say saved "it", but in this configuration there are 24 "its" (the
modems) so is this just a grammar burp, or are you not actually saving
the modems themselves?
> I have been sitting here switching them back after each call. How can I
> make it stick. I also have the reset method via the atNVram thingy.
> can maybe someone help me?
The modem normally performs a reset at call completion, so the
reversion of the settings would seem to imply that they really aren't
being properly saved. It sounds like what you're doing should be ok
though.
You might try resetting all your modems to factory defaults just to be
on the safe side (perhaps they accidentally are set to reference a
different NVRAM profile than the default which is being saved to?),
then sending your changes, and then saving each modem to NVRAM. If
you are changing the line source settings, you normally have to reset
the modem (or card) as well to have it take effect.
You might also also check that the NMC isn't configured to
auto-configure equipment in the chassis. Although it should track
your changes to the modem settings (unless you're doing it through the
console rather than the NMC), if that gets out of whack, it might
reconfigure your modem on each reset to it's perception of the setting
values.
In terms of testing things, rather than waiting for a call, just reset
the modem yourself and ensure that the setting sticks.
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
John Powell writes...
>We are looking into issues with 1.2.68 (and .67, .66, .65, and .64)
>causing problems on some T3 muxes. There is not an issue on 1.2.5 or
>engineering releases numbered above 1.2.68. It also does not appear to be
>an issue in E1 codebases.
>
>The issue seems contained to certain muxes, but is serious enough when it
>does happen, that I highly recommend not using those releases on any
>chassis that don't already have it until we completely nail this issue and
>issue a patch.
I bet it's a timing problem, you should have your customer check that
he's set for loop timing (on the hiperdsp, not the mux).
I haven't run a pulse mask test on the output of a hiperdsp, but it's
also possible that the mux sees it as too hot. You could try increasing
the transmitter attenuation. On the Hiperdsp, the nic seems configuarble
on the fly (unlike the dual T1 card) for short haul, you could set
it to short if it's plugged directly into a mux.
Also note that the default "Jitter Attenuation" setting for the hiperdsp
is wrong, for loop-timed applications you want the jitter attenuation in
the receive path, not that transmit path.
-a
Subject:Re: (usr-tc) Interested in Line on the Netserver? From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-19 22:01:39
Thus spake Brian K McIntire
>The fact that the 2e supports only 30 may partially explain how it can
>handle things a little better. It doesn't have to do the work the USR
>NetServer does
Oh, come now...
A 386 with 30 ports, or a 486 with 48? You're saying a 486 is less than
double the power of a 386? shyeah right...and I've got some swampland
in florida to sell you.
A 486 with 48 ports should *clobber* a 386 with 30 ports in
performance...*particularly* when you consider the offloading of ppp
framing into the quad cards! This shouldn't even be *close*!
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
David Bolen writes...
>Aaron Nabil <nabil@spiritone.com> writes:
>
>> If you are running NFAS, by all means use a backup D channel. A much,
>> much better solution is never to use NFAS.
>
>Depends - NFAS can be helpful if you're exceeding telco capacities for
>hunt groups (too many D channels for the switch to handle - we run
>into that at around 64 spans on some switches and software releases),
>or if you just don't want to give up so many channels to D channel
>signaling on large hunt groups. For example, a fully populated HDM
>rack might have 56 spans in 4 chassis - using a D+backup setup per
>chassis would get back 48 channels, or two full spans worth, which can
>be worth a lot of money.
I'm just going to take a stab at an estimate of what a 56 span
system costs, say $250,000. You are talking about wasting, say,
$10,000 of that on the d-channels? That's miniscule compared to
the trouble NFAS causes in troubleshooting, provisioning, and
black-hole problems I've already mentioned when you lose the span
in a way that doesn't alarm at the CO.
One factor that often gets missed is that the CO often provisions
the backup D channel on the same switchmod. The single NFAS
installation I've worked with was this way. If that switchmod
fails, you've just idled your $250,000 investment.
--
Aaron Nabil
You are correct, that is 1.2.68.
I don't want to create any panic, as this has only been seen in two sites.
The problem is serious enough though that I would be negligent to not post
a warning a recommend not installing it.
JP
On Thu, 19 Nov 1998, Brian wrote:
>
> I am assuming hd102682.zip is effected as well? (must have been a .68
> variant)
>
>
> On Mon, 19 Nov 2018, John Powell wrote:
>
> > We are looking into issues with 1.2.68 (and .67, .66, .65, and .64)
> > causing problems on some T3 muxes. There is not an issue on 1.2.5 or
> > engineering releases numbered above 1.2.68. It also does not appear to be
> > an issue in E1 codebases.
> >
> > The issue seems contained to certain muxes, but is serious enough when it
> > does happen, that I highly recommend not using those releases on any
> > chassis that don't already have it until we completely nail this issue and
> > issue a patch.
> >
> > If you are already using one of them, and are not experiencing
> > problems, there is
> > no need to back them off, this is not a problem that creeps up. Your
> > spans either have a problem with it, or they don't.
> >
> > So, if you have 1.2.68 (that one is out there in large quantities), please
> > use great care and I do not suggest you expand its' use at all.
> >
> > Regards,
> >
> > JP
> >
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
> --------------------------------------------------------------------------
> Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Aaron Nabil <nabil@spiritone.com> writes:
> If you are running NFAS, by all means use a backup D channel. A much,
> much better solution is never to use NFAS.
Depends - NFAS can be helpful if you're exceeding telco capacities for
hunt groups (too many D channels for the switch to handle - we run
into that at around 64 spans on some switches and software releases),
or if you just don't want to give up so many channels to D channel
signaling on large hunt groups. For example, a fully populated HDM
rack might have 56 spans in 4 chassis - using a D+backup setup per
chassis would get back 48 channels, or two full spans worth, which can
be worth a lot of money.
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
Subject:Re: (usr-tc) padding on trunk groups From: John Powell <jp@packet.ae.usr.com> Date: 1998-11-20 06:33:20
Brian,
Your note exemplifies why we don't publish the info on pads. That leads
to driving the telcos nuts on a subject most users, and certainly most
telco people, do not fully understand. A very important note, is that the
modems do not detect analog pads, they look like loop loss and are not
discernable. The modems can only detect digital pads, and those are
really the only ones they are concerned about anyway (as they need to be
mathematically compensated for).
So, with that in mind, all the analog lines have pads, you just don't see
the ones applied to one of the lines as it is an analog pad (done on the
analog side of the codec).
Also, in common practice, the pads are not applied to the PRIs xmit
direction, they are applied to the codecs xmit direction (analog modem's
receive). This may seem like a semantics argument, but it is critical to
understand as that means it is not a configurable item in the PRIs
translations. You asked David to comment on this, but I don't see why he
would know or care, the pad is inserted by the CO serving the analog user.
It is possible to have pads provisioned on the PRI side in its' xmit
direction. That is not often done, and is bad practice. In that case you
would see two pads in the V.90 direction, the one that belongs there on
the CO serving the analog user, and the one that does not belong there on
the PRI side. I have seen this, but it is very rare.
On your "quirk", it is probably not a quirk at all. I assume you are
dialing your T1 from an analog line on the same CO. The specs recommend
that no pad be inserted on intra-CO calls. Sometimes you will see a 3db
pad, and that is OK, but often no pad at all. You see the 6db pad on the
AOL call, as it is in a different CO, quite likely on a CLEC span which is
treated like an LD call.
Pads are related to echo cancellation, they aid in reducing echo so the EC
has less to do. There is not a direct relationship though, as the
"triggers" are quite a bit different. Pads are triggered based on dial
plans and where you are calling/called from. ECs are generally not
installed on local networks, but are only on LD/IXC networks. Even then,
they are generally triggered and converged based on Round trip delay.
Bottom line is I am not sure you are on the right track looking at the
pads. They may be related to a speed click or so change, as digital pads
do affect "bit integrity" a hair. You will not be able to get them
changed and they are not affected by the provisioning of the T1/PRI in
most cases (and it does not appear your examples are one of the
exceptions). Also, there are plenty of conditions where digital pads are
actually a good thing for V.90 modems. Your single data point may show it
is the opposite...oh well.
I think you need to look more at analog line quality and let the big boys
handle the pads and stuff. I
don't see anything in your notes that
suggest pads are causing you any issues beyond a speedclick or so.
JP
On Thu, 19 Nov 1998, Brian K McIntire wrote:
> Please forgive the following "book"
>
> I'm faced with a problem. I'm working for a customer in an Ameritech
> region. When ever I place a call with my courier to a PRI I see a 6db pad
> being applied to my receive. (The PRI's transmit) I've been told there has
> to be a 6db pad applied because Ameritech has not installed echo
> cancellation in their telephone network. Makes some sense. I have further
> been told echo cancellation is not currently accepted as a standard but
> where ever it has been installed telephone companies remove the 6db pad my
> client modem see's on the receive.
>
> I understand there needs to be echo cancellation to go from a 4 to 2 wire.
> (PRI to analog) If not there needs to be a pad. Does anyone have a
> different opinion or know something different. Is there a way around this?
> (Probably not) If not does anyone know if using echo cancellation will
> become a standard?
>
> The bottom line is, I have some customer's that connect at higher speeds
> than others because of who their local telephone company is. The PRI I'm
> working on is set up to be a local # for several cities with different
> switches. One of the cities that dial this PRI applies no padding. The
> others do. The problem is my customer's main competition is centrally
> located in the local loop of the telco that does not pad. His customer's
> are being called by his competition and given a chance to call a non-padded
> PRI for free to prove their connect speeds will be better and steal
> customers. Of course, connect speeds are always better when you compare
> padded against non-padded.
>
> Any positive ideas?
>
>
>
>
> By the way, here's an interesting quirk. When I dial the PRI that goes to
> our internal lab chassis from a trunk off a non-padded T1 being provided by
> the same telco I see no pad and connect at 50666. When I dial AOL's local #
> (PRI's being provided by the same telco) I see 6db padding and connect at
> 49333 or 48000.
>
> When I dial either from home (from a different phone company "Ameritheft") I
> connect at 45333 or lower. Sometimes I don't even connect at v.90. I see
> 37333 quite often.
>
> David, do you have any comments on this. The AOL chassis here are yours.
>
> ========================================================================
> = Brian K. McIntire bmcintire@commnet.com =
> = Systems Engineer III 317-558-5050 x113 =
> = CommNetPlus, Indianapolis, IN http://www.commnet.com =
> = Your Remote Access and Security Experts! =
> = =
> = Our experienced technicians can design and install your high speed =
> = network for you. From Operating Systems and servers to routers and =
> = firewalls nobody does it better than CommNetPlus. Our Technical =
> = Support staff is available to you 24 hours a day to meet your needs.=
> = Offering remote management and monitoring packages for dedicated =
> = clients. (Specializing in ISP's). "Let us do the work for you!" =
> = =
> = Firm partnerships est. with current and emerging leaders of today's =
> = technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
> = Serving North America and parts of Canada. Reselling to the world. =
> ========================================================================
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) v.90 connect problems From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-20 08:37:10
Not sure where you get your info from but it doesn't sound informed. I too
have seen trouble on a span due to a looped repeater. However, we have seen
more problems with d-channels going down. To the tune of over a hundred a
month. I still love PRI's, and I think they are better and easier to
troubleshoot with telco, but I also believe having a backup d is very
important. It would have saved our customers allot of trouble. Anyway,
your ignorant "stupid" remark is noted. Hope you enjoy your minute in the
sun. Have a nice day
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Aaron Nabil
>Sent: Thursday, November 19, 1998 7:25 PM
>To: usr-tc@lists.xmission.com
>Subject: Re: (usr-tc) v.90 connect problems
>
>
>Brian K McIntire writes...
>>The difference I'm referring to is not just in performance but reliability
>>and sometimes, on ease of turn up. On PRI's you do run this risk
>of losing
>>an entire span whenever a d-cannel goes down. Just another argument for
>>Backup D.
>
>This is stupid.
>
>The probability of losing a D channel is miniscule compared to losing the
>span to something like a looped repeater. When this happens
>having a backup
>D channel will make you very, very sorry as your calls dissapear down
>a black hole.
>
>
>-a
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) FYI: 1.2.66 & PRI errors From: Jerry Kalligonis <jerryk@blazenet.net> Date: 1998-11-20 08:40:35
It was patch 1.2.66 (see subject) as see John Powell's warning post....
Jerry
At 05:31 PM 11/19/98 -0800, you wrote:
>Jerry Kalligonis writes...
>>At one particular location I have a good many PRI's. I was running the
>>1.2.5 code, but had heard good things about the newer patch from other
>>posts on this list.
>>
>> . . .
>>
>>Anyway, after hours of trouble shooting and no help from 3com, we found
>>that the new code puts out a certain level of errors onto the PRI line.
>>When enough of these PRIs are combined the error level is enough to bring
>>down a T3.
>>
>>Of course later that afternoon I got a call back from 3com saying, sorry I
>>couldn't get back to you this morning, but engineering says there may be a
>>problem with this code that puts out enough errors to take down a T3. Call
>>back to get DSP code 1.2.64
>
>You didn't mention what version the "newer patch" was that cause these
>problems for you.
>
>I was having trouble with false alarms coming from my DSP cards which
>version 1.2.68 fixed.
>
>
>--
>Aaron Nabil
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) Interested in Line on the NetServer? From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-20 08:44:47
Not everyone is using a NetServer for 48 ports. Many are using it for 96.
If many of these 96 calls are passing massive streams of small packets then
the 386 will perform better. makes sense. ;\
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jeff Mcadams
>Sent: Thursday, November 19, 1998 9:02 PM
>To: usr-tc@lists.xmission.com
>Subject: Re: (usr-tc) Interested in Line on the Netserver?
>
>
>Thus spake Brian K McIntire
>>The fact that the 2e supports only 30 may partially explain how it can
>>handle things a little better. It doesn't have to do the work the USR
>>NetServer does
>
>Oh, come now...
>
>A 386 with 30 ports, or a 486 with 48? You're saying a 486 is less than
>double the power of a 386? shyeah right...and I've got some swampland
>in florida to sell you.
>
>A 486 with 48 ports should *clobber* a 386 with 30 ports in
>performance...*particularly* when you consider the offloading of ppp
>framing into the quad cards! This shouldn't even be *close*!
>--
>Jeff McAdams Email: jeffm@iglou.com
>Head Network Administrator Voice: (502) 966-3848
>IgLou Internet Services (800) 436-4456
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) Interested in Line on the NetServer? From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-20 08:57:49
Thus spake Brian K McIntire
>Not everyone is using a NetServer for 48 ports. Many are using it for 96.
>If many of these 96 calls are passing massive streams of small packets then
>the 386 will perform better. makes sense. ;\
A NETServer handling 96 ports isn't the issue though....I'm not gonna
argue about that, because it is getting to the limit of what a 486 could
handle (it should be able to handle it, but its getting close)...
But "Quake Lag" has been a problem for a *long* time...long before the
HiPer equipment was available, meaning that its a problem even with only
48 ports in the box, so your argument that its a problem with 96 ports
and that its reasonable for 96 ports to bog down like that on a 486 is
moot. We're still talking 48 ports here where its bogging down.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:RE: (usr-tc) v.90 connect problems From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-20 08:57:50
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Aaron Nabil
>Sent: Friday, November 20, 1998 6:34 AM
>To: usr-tc@lists.xmission.com
>Subject: Re: (usr-tc) v.90 connect problems
>
>
>David Bolen writes...
>>Aaron Nabil <nabil@spiritone.com> writes:
>>
>>> If you are running NFAS, by all means use a backup D channel. A much,
>>> much better solution is never to use NFAS.
>>
>>Depends - NFAS can be helpful if you're exceeding telco capacities for
>>hunt groups (too many D channels for the switch to handle - we run
>>into that at around 64 spans on some switches and software releases),
>>or if you just don't want to give up so many channels to D channel
>>signaling on large hunt groups. For example, a fully populated HDM
>>rack might have 56 spans in 4 chassis - using a D+backup setup per
>>chassis would get back 48 channels, or two full spans worth, which can
>>be worth a lot of money.
>
>I'm just going to take a stab at an estimate of what a 56 span
>system costs, say $250,000. You are talking about wasting, say,
>$10,000 of that on the d-channels? That's miniscule compared to
>the trouble NFAS causes in troubleshooting, provisioning, and
>black-hole problems I've already mentioned when you lose the span
>in a way that doesn't alarm at the CO.
>
>One factor that often gets missed is that the CO often provisions
>the backup D channel on the same switchmod. The single NFAS
>installation I've worked with was this way. If that switchmod
>fails, you've just idled your $250,000 investment.
>
>
>--
>Aaron Nabil
This arguement is getting way off topic but i would have to agree it is
better to not use NFAS to beging with. However, most of my customer's like
it. They still like it even after I warn them about the possibilities of
what may happen. So, we use it. All in all, it can be riskier, but I can't
say I have had more trouble with it than not using it. Your experience may
be otherwise, but your lack of good experience with it does not negate mine.
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Thus spake David Bolen
>Aaron Nabil <nabil@spiritone.com> writes:
>> If you are running NFAS, by all means use a backup D channel. A much,
>> much better solution is never to use NFAS.
>For example, a fully populated HDM
>rack might have 56 spans in 4 chassis - using a D+backup setup per
>chassis would get back 48 channels, or two full spans worth, which can
>be worth a lot of money.
That, of course, is implying that the HDM's support NFAS, which they
don't to my knowledge, unless you're running a special build, but I
don't think that's likely. Agreed though, NFAS *can* save you a bundle
of channels, and it'd be a nice feature to have on the HDM's.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Thus spake Brian K McIntire
>Not sure where you get your info from but it doesn't sound informed. I too
>have seen trouble on a span due to a looped repeater. However, we have seen
>more problems with d-channels going down.
Lucky you...I've only had a d channel go down independantly of the span
once....I've had span's go down for other reasons myriads of times.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:RE: (usr-tc) Interested in Line on the NetServer? From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-20 09:07:23
I understand. Honestly, I haven't seen too many issues with 48 ports on the
486 NetServer but I have seen some
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jeff Mcadams
>Sent: Friday, November 20, 1998 7:58 AM
>To: usr-tc@lists.xmission.com
>Subject: Re: (usr-tc) Interested in Line on the NetServer?
>
>
>Thus spake Brian K McIntire
>>Not everyone is using a NetServer for 48 ports. Many are using it for 96.
>>If many of these 96 calls are passing massive streams of small
>packets then
>>the 386 will perform better. makes sense. ;\
>
>A NETServer handling 96 ports isn't the issue though....I'm not gonna
>argue about that, because it is getting to the limit of what a 486 could
>handle (it should be able to handle it, but its getting close)...
>
>But "Quake Lag" has been a problem for a *long* time...long before the
>HiPer equipment was available, meaning that its a problem even with only
>48 ports in the box, so your argument that its a problem with 96 ports
>and that its reasonable for 96 ports to bog down like that on a 486 is
>moot. We're still talking 48 ports here where its bogging down.
>--
>Jeff McAdams Email: jeffm@iglou.com
>Head Network Administrator Voice: (502) 966-3848
>IgLou Internet Services (800) 436-4456
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) v.90 connect problems From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-20 09:07:24
HDM's will support NFAS soon. I got the work from one of the Engineer's who
desogns the code. If it doesn't make TCS 3.5 it will make TCS 4.0
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jeff Mcadams
>Sent: Friday, November 20, 1998 8:01 AM
>To: usr-tc@lists.xmission.com
>Subject: Re: (usr-tc) v.90 connect problems
>
>
>Thus spake David Bolen
>>Aaron Nabil <nabil@spiritone.com> writes:
>>> If you are running NFAS, by all means use a backup D channel. A much,
>>> much better solution is never to use NFAS.
>
>>For example, a fully populated HDM
>>rack might have 56 spans in 4 chassis - using a D+backup setup per
>>chassis would get back 48 channels, or two full spans worth, which can
>>be worth a lot of money.
>
>That, of course, is implying that the HDM's support NFAS, which they
>don't to my knowledge, unless you're running a special build, but I
>don't think that's likely. Agreed though, NFAS *can* save you a bundle
>of channels, and it'd be a nice feature to have on the HDM's.
>--
>Jeff McAdams Email: jeffm@iglou.com
>Head Network Administrator Voice: (502) 966-3848
>IgLou Internet Services (800) 436-4456
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) v.90 connect problems From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-20 09:10:11
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jeff Mcadams
>Sent: Friday, November 20, 1998 8:03 AM
>To: usr-tc@lists.xmission.com
>Subject: Re: (usr-tc) v.90 connect problems
>
>
>Thus spake Brian K McIntire
>>Not sure where you get your info from but it doesn't sound
>informed. I too
>>have seen trouble on a span due to a looped repeater. However,
>we have seen
>>more problems with d-channels going down.
>
>Lucky you...I've only had a d channel go down independantly of the span
>once....I've had span's go down for other reasons myriads of times.
Out of curiosity, what problems have you experienced with PRI's going down.
Auto-busy outs at the switch; fibr cuts; loss of timing; etc..
>--
>Jeff McAdams Email: jeffm@iglou.com
>Head Network Administrator Voice: (502) 966-3848
>IgLou Internet Services (800) 436-4456
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Thus spake Brian K McIntire
>HDM's will support NFAS soon. I got the work from one of the Engineer's who
>desogns the code. If it doesn't make TCS 3.5 it will make TCS 4.0
Cool! That's good to know.
We might switch back to NFAS then if that's the case...hadn't done it to
this point because the benefit of gaining one channel per every other
span wasn't big enough to be significant with the dual-pri cards.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Thus spake Brian K McIntire
>>Thus spake Brian K McIntire
>>>Not sure where you get your info from but it doesn't sound
>>informed. I too
>>>have seen trouble on a span due to a looped repeater. However,
>>we have seen
>>>more problems with d-channels going down.
>>Lucky you...I've only had a d channel go down independantly of the span
>>once....I've had span's go down for other reasons myriads of times.
>Out of curiosity, what problems have you experienced with PRI's going down.
>Auto-busy outs at the switch; fibr cuts; loss of timing; etc..
Oh man...I don't think we've ever had the same problem twice... :)
These are just ones that I happen to remember...
fuse blowing on the DSX, wire at a punch-down block coming loose (both
66 blocks and 110 blocks), having an rj-45 come out of its connector,
dual-pri card loosing its config, probably others that I'm forgetting
about here, but that's a good start.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:RE: (usr-tc) Re: your mail From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-20 09:26:27
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Ricky Beam
>Sent: Thursday, November 19, 1998 5:07 PM
>To: usr-tc@lists.xmission.com
>Subject: Re: (usr-tc) Re: your mail
>
>
>Brian K McIntire was heard to say:
>>Thanks for the info Dominic. I upgraded to the code you specified. This
>>time it doesn't work at all. Instead of seeing the error message
>incorrect
>>passcode the RADIUS isn't even communicating with Secure ID. Any other
>>Ideas. Debug was useless. All it showed was requests going out that were
>>un- answered. Again, when I down graded back to 5.5.4 it worked like a
>>charm.
>
>Ok, so try this...
>
>copy the bin/session from 5.5.4 to bin/saworker for 6.0.whatever and see
>if that makes any difference at all. That's what the server runs to handle
>authenticating with ACE. As far as I'm aware, other than the name change,
>there is no difference between the two?
Thanks for the response. Unfortunately, that didn't make a difference
either. As I said before, with the same config, 5.5.4 works just fine.
Dominic, has that ER code 6.0.92 been thoroughly tested? I'm working with
another Engineer here who has installed with TCM S/A for Unix and Ace a few
dozen times. He indicates he configured 6.0.92 the same as he configured
5.5.4. Any ideas.
>
>--Ricky
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Laszlo Vecsey
|Sent: Thursday, November 19, 1998 6:11 PM
|To: usr-tc@lists.xmission.com
|Subject: RE: (usr-tc) hiperarc NAS-Port-Id
|
|
|On Thu, 19 Nov 1998, Mike Wronski wrote:
|
|> |With 4.1.72 on the hiperarc, theres an improvement with the NAS-Port-Id
|> |output (used to all be set to 1), but I think theres still a glitch:
|> |
|> | NAS-Port-Id = 1539
|> | NAS-Port-Id = 1028
|> | NAS-Port-Id = 1793
|> | NAS-Port-Id = 1026
|> | NAS-Port-Id = 514
|> | NAS-Port-Id = 1793
|> | NAS-Port-Id = 1794
|> | NAS-Port-Id = 912469510
|> | NAS-Port-Id = 1795
|> | NAS-Port-Id = 1796
|> | NAS-Port-Id = 258
|> | NAS-Port-Id = 2049
|>
|> Can you show the entire record the Odd entry came from??
|> GNU GREP with -A 3 -B 3 ???
|>
|
|I did some more greps and found that the oddball entries always have
|Acct-Multi-Session-Id entries, as the following full example shows, its
|complete except for the IP and username etc info that I've stripped out.
|
|--
|Thu Nov 19 18:18:18 1998
| Acct-Status-Type = Start
| Acct-Session-Id = "3967"
| Acct-Delay-Time = 0
| Acct-Authentic = RADIUS
| Service-Type = Framed-User
| NAS-Port-Type = Async
| NAS-Port-Id = 1537
| USR-Modem-Training-Time = 23
| USR-Interface-Index = 2793
| USR-Chassis-Call-Slot = 7
| USR-Chassis-Call-Span = 1
| USR-Chassis-Call-Channel = 1
| USR-Unauthenticated-Time = 9
| USR-Modulation-Type = v34
| USR-Simplified-MNP-Levels = 13
| USR-Simplified-V42bis-Usage = ccittV42bis
| USR-Connect-Speed = 26400_BPS
| Framed-Protocol = PPP
| USR-MP-MRRU = 0
| Acct-Link-Count = 1
| Acct-Multi-Session-Id = 875639045
| NAS-Port-Id = 825242721
| Timestamp = 911517498
| Request-Authenticator = Unverified
|--
|
|Notice the NAS-Port-Id is listed twice! Up near the top it looks correct,
|down at the bottom it seems to contain an Acct-Multi-Session-Id number by
|mistake.
|
|This is affecting cistron radius in particular, the radwho program.. which
|I suppose could be recoded to use the first NAS-Port-Id, but shouldn't
|there only be one anyway?
The dupe is strange.. Would it be possible to get a snoop/tcpdump trace of one of
these strange packets? Binary version is required.. (ie snoop -x 0 -o
acct.packets )..
Also is anyone else seeing this??
-M
Subject:RE: (usr-tc) Interested in Line on the Netserver? From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-20 09:56:32
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jeff Mcadams
|Sent: Thursday, November 19, 1998 9:02 PM
|To: usr-tc@lists.xmission.com
|Subject: Re: (usr-tc) Interested in Line on the Netserver?
|
|
|Thus spake Brian K McIntire
|>The fact that the 2e supports only 30 may partially explain how it can
|>handle things a little better. It doesn't have to do the work the USR
|>NetServer does
|
|Oh, come now...
|
|A 386 with 30 ports, or a 486 with 48? You're saying a 486 is less than
|double the power of a 386? shyeah right...and I've got some swampland
|in florida to sell you.
|
|A 486 with 48 ports should *clobber* a 386 with 30 ports in
|performance...*particularly* when you consider the offloading of ppp
|framing into the quad cards! This shouldn't even be *close*!
Of course this 486 doesnt have an external cache. Take a look at the board.
-M
Subject:RE: (usr-tc) Interested in Line on the NetServer? From: Laszlo Vecsey <master@internexus.net> Date: 1998-11-20 10:00:42
From what I've seen, quakelag would start kicking in when around 30+ports
are in use (on a 48port chasis). Probably less than 20 and its lag free, I
think thats also about the approximate figure for the number of ccp/stac
sessions the netserver can handle.
On Fri, 20 Nov 1998, Brian K McIntire wrote:
> I understand. Honestly, I haven't seen too many issues with 48 ports on the
> 486 NetServer but I have seen some
> >-----Original Message-----
> >From: owner-usr-tc@lists.xmission.com
> >[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jeff Mcadams
> >Sent: Friday, November 20, 1998 7:58 AM
> >To: usr-tc@lists.xmission.com
> >Subject: Re: (usr-tc) Interested in Line on the NetServer?
> >
> >
> >Thus spake Brian K McIntire
> >>Not everyone is using a NetServer for 48 ports. Many are using it for 96.
> >>If many of these 96 calls are passing massive streams of small
> >packets then
> >>the 386 will perform better. makes sense. ;\
> >
> >A NETServer handling 96 ports isn't the issue though....I'm not gonna
> >argue about that, because it is getting to the limit of what a 486 could
> >handle (it should be able to handle it, but its getting close)...
> >
> >But "Quake Lag" has been a problem for a *long* time...long before the
> >HiPer equipment was available, meaning that its a problem even with only
> >48 ports in the box, so your argument that its a problem with 96 ports
> >and that its reasonable for 96 ports to bog down like that on a 486 is
> >moot. We're still talking 48 ports here where its bogging down.
> >--
> >Jeff McAdams Email: jeffm@iglou.com
> >Head Network Administrator Voice: (502) 966-3848
> >IgLou Internet Services (800) 436-4456
> >
> >-
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) v.90 connect problems From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-20 10:04:03
Allot of the thing you mention sound like they would affect both T1 and PRI.
Been there done that
How about looped smartjacks?
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Jeff Mcadams
>Sent: Friday, November 20, 1998 8:25 AM
>To: usr-tc@lists.xmission.com
>Subject: Re: (usr-tc) v.90 connect problems
>
>
>Thus spake Brian K McIntire
>>>Thus spake Brian K McIntire
>>>>Not sure where you get your info from but it doesn't sound
>>>informed. I too
>>>>have seen trouble on a span due to a looped repeater. However,
>>>we have seen
>>>>more problems with d-channels going down.
>
>>>Lucky you...I've only had a d channel go down independantly of the span
>>>once....I've had span's go down for other reasons myriads of times.
>
>>Out of curiosity, what problems have you experienced with PRI's
>going down.
>>Auto-busy outs at the switch; fibr cuts; loss of timing; etc..
>
>Oh man...I don't think we've ever had the same problem twice... :)
>
>These are just ones that I happen to remember...
>fuse blowing on the DSX, wire at a punch-down block coming loose (both
>66 blocks and 110 blocks), having an rj-45 come out of its connector,
>dual-pri card loosing its config, probably others that I'm forgetting
>about here, but that's a good start.
>--
>Jeff McAdams Email: jeffm@iglou.com
>Head Network Administrator Voice: (502) 966-3848
>IgLou Internet Services (800) 436-4456
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) Interested in Line on the Netserver? From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-20 11:25:39
Thus spake Mike Wronski
>Of course this 486 doesnt have an external cache. Take a look at the board.
OK, just for yucks, I went downstairs and found a PM-25 that we had on a
shelf here (only 25 ports, but same architecture as the 2E from what I
understand)....guess what...it doesn't have an external cache either.
Can anyone confirm or deny if the 2E has an external cache? I don't
believe it does, but a confirmation would be nice.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:RE: (usr-tc) padding on trunk groups From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-20 13:09:04
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of John Powell
>Sent: Tuesday, November 20, 2018 6:33 AM
>To: usr tc
>Subject: Re: (usr-tc) padding on trunk groups
>
>
>Brian,
>
>Your note exemplifies why we don't publish the info on pads.
I have a line analyzer that reports the pads to me as well. Either way I
see pads.
>That leads
>to driving the telcos nuts on a subject most users, and certainly most
>telco people, do not fully understand. A very important note, is that the
>modems do not detect analog pads, they look like loop loss and are not
>discernable.
I know this
>The modems can only detect digital pads, and those are
>really the only ones they are concerned about anyway (as they need to be
>mathematically compensated for).
We've covered this
>
>So, with that in mind, all the analog lines have pads, you just don't see
>the ones applied to one of the lines as it is an analog pad (done on the
>analog side of the codec).
Again, we have covered this.
>
>Also, in common practice, the pads are not applied to the PRIs xmit
>direction, they are applied to the codecs xmit direction (analog modem's
>receive). This may seem like a semantics argument, but it is critical to
>understand as that means it is not a configurable item in the PRIs
>translations.
I know it isn't
>You asked David to comment on this, but I don't see why he
>would know or care, the pad is inserted by the CO serving the analog user.
>
It's not a matter of would he card, but more a matter of me asking him if he
has seen this before. Jeez, why is everyone getting excited over simple
questions! me customer you support
>It is possible to have pads provisioned on the PRI side in its' xmit
>direction. That is not often done, and is bad practice. In that case you
>would see two pads in the V.90 direction, the one that belongs there on
>the CO serving the analog user, and the one that does not belong there on
>the PRI side. I have seen this, but it is very rare.
>
>On your "quirk", it is probably not a quirk at all. I assume you are
>dialing your T1 from an analog line on the same CO. The specs recommend
>that no pad be inserted on intra-CO calls. Sometimes you will see a 3db
>pad, and that is OK, but often no pad at all. You see the 6db pad on the
>AOL call, as it is in a different CO, quite likely on a CLEC span which is
>treated like an LD call.
Incorrect. It is the same CO. The exact same co server my PRI and the T1
my analog test calls were made from.
>
>Pads are related to echo cancellation, they aid in reducing echo so the EC
>has less to do. There is not a direct relationship though, as the
>"triggers" are quite a bit different. Pads are triggered based on dial
>plans and where you are calling/called from. ECs are generally not
>installed on local networks, but are only on LD/IXC networks. Even then,
>they are generally triggered and converged based on Round trip delay.
>
>Bottom line is I am not sure you are on the right track looking at the
>pads. They may be related to a speed click or so change, as digital pads
>do affect "bit integrity" a hair. You will not be able to get them
>changed and they are not affected by the provisioning of the T1/PRI in
>most cases (and it does not appear your examples are one of the
>exceptions). Also, there are plenty of conditions where digital pads are
>actually a good thing for V.90 modems. Your single data point may show it
>is the opposite...oh well.
>
>I think you need to look more at analog line quality and let the big boys
>handle the pads and stuff. I
>don't see anything in your notes that
>suggest pads are causing you any issues beyond a speedclick or so.
>
John, you apparently do not remember anything we have talked about. I
appreciate your help, but if you are going to try to slam me at least know
what you are talking about. This is a continuation of the same problem we
have had here for quite awhile. I sent you copies of ati4i6i7i11y11 screens
as you requested. You never responded. I have made call after call after
call and my line tester shows no discernable differences other than padding
and/or connect speed. From several homes in the area we see connect speeds
range from 28800 to 49333. Each home capable of getting all speeds in
between. Honestly, I don't know what the next step is. I contacted this
list looking for a new direction and some information to expand my own
knowledge.
By the way, in your response you spent most of your time going over things
you and I have already gone over. (Very good stuff though) It would have
been more helpful if you actually tried answering the actual questions I was
asking.
Please do not respond to me again lecturing me about how padding works. We
have already been down this road. (I'm sure you know a great deal more
about it than I but that is besides the point) I have forgotten very little
of what you told me already.
One last thing. This list is present for people like me to be able to
communicate with people like yourself, get issues resolved and to answer
questions. (To find a means to an end) Re-read my original e-mail. I
think you will find you answered very little of what I asked (I was curious.
Never said I was on Man Chu trail of padding) and further; offered no way to
get me where I need to go. Not very helpful.
Again, this is not malice. I know you are a very smart man. You really
know your stuff, but if you need to vent your knowledge on someone in a
negative way please vent it somewhere else.
I do not want to open a ticket with 3COM on this. I'm sure it is telco
related. If anyone has more constructive criticism to offer then let me
here it.
Thank you
>JP
>
>On Thu, 19 Nov 1998, Brian K McIntire wrote:
>
>> Please forgive the following "book"
>>
>> I'm faced with a problem. I'm working for a customer in an Ameritech
>> region. When ever I place a call with my courier to a PRI I see
>a 6db pad
>> being applied to my receive. (The PRI's transmit) I've been
>told there has
>> to be a 6db pad applied because Ameritech has not installed echo
>> cancellation in their telephone network. Makes some sense. I
>have further
>> been told echo cancellation is not currently accepted as a standard but
>> where ever it has been installed telephone companies remove the
>6db pad my
>> client modem see's on the receive.
>>
>> I understand there needs to be echo cancellation to go from a 4
>to 2 wire.
>> (PRI to analog) If not there needs to be a pad. Does anyone have a
>> different opinion or know something different. Is there a way
>around this?
>> (Probably not) If not does anyone know if using echo cancellation will
>> become a standard?
>>
>> The bottom line is, I have some customer's that connect at higher speeds
>> than others because of who their local telephone company is. The PRI I'm
>> working on is set up to be a local # for several cities with different
>> switches. One of the cities that dial this PRI applies no padding. The
>> others do. The problem is my customer's main competition is centrally
>> located in the local loop of the telco that does not pad. His customer's
>> are being called by his competition and given a chance to call a
>non-padded
>> PRI for free to prove their connect speeds will be better and steal
>> customers. Of course, connect speeds are always better when you compare
>> padded against non-padded.
>>
>> Any positive ideas?
>>
>>
>>
>>
>> By the way, here's an interesting quirk. When I dial the PRI
>that goes to
>> our internal lab chassis from a trunk off a non-padded T1 being
>provided by
>> the same telco I see no pad and connect at 50666. When I dial
>AOL's local #
>> (PRI's being provided by the same telco) I see 6db padding and connect at
>> 49333 or 48000.
>>
>> When I dial either from home (from a different phone company
>"Ameritheft") I
>> connect at 45333 or lower. Sometimes I don't even connect at
>v.90. I see
>> 37333 quite often.
>>
>> David, do you have any comments on this. The AOL chassis here are yours.
>>
>> ========================================================================
>> = Brian K. McIntire bmcintire@commnet.com =
>> = Systems Engineer III 317-558-5050 x113 =
>> = CommNetPlus, Indianapolis, IN http://www.commnet.com =
>> = Your Remote Access and Security Experts! =
>> = =
>> = Our experienced technicians can design and install your high speed =
>> = network for you. From Operating Systems and servers to routers and =
>> = firewalls nobody does it better than CommNetPlus. Our Technical =
>> = Support staff is available to you 24 hours a day to meet your needs.=
>> = Offering remote management and monitoring packages for dedicated =
>> = clients. (Specializing in ISP's). "Let us do the work for you!" =
>> = =
>> = Firm partnerships est. with current and emerging leaders of today's =
>> = technologies. Sales 800-845-2981 x117 (Aggressive Pricing) =
>> = Serving North America and parts of Canada. Reselling to the world. =
>> ========================================================================
>>
>>
>> -
>> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
>> with "unsubscribe usr-tc" in the body of the message.
>> For information on digests or retrieving files and old messages send
>> "help" to the same address. Do not use quotes in your message.
>>
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) padding on trunk groups From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-20 14:14:19
The only good information I have seen or heard of is in the form of very
expensive books. An acquaintance of mine at a nearby telco said he spent
378,000 last year alone on books for his systems. I wish I knew of a book
that breaks things down into layman's terms too. I'm sure there's one out
there.
Subject:RE: (usr-tc) padding on trunk groups From: Sys Mgr - BrunNet <"stainforth, matthew (sys mgr - brunnet)"> Date: 1998-11-20 14:26:18
On Tuesday, November 20, 2018 8:33 AM, John Powell
[SMTP:jp@packet.ae.usr.com] wrote:
> Your note exemplifies why we don't publish the info on pads. That leads
> to driving the telcos nuts on a subject most users, and certainly most
> telco people, do not fully understand. A very important note, is that the
> modems do not detect analog pads, they look like loop loss and are not
> discernable. The modems can only detect digital pads, and those are
> really the only ones they are concerned about anyway (as they need to be
> mathematically compensated for).
Kind of off topic, but does anybody know of any on-line resources for
getting to know the line provisioning/telco side of things better? I've
looked for some time and I can't seem to find anything beyond a few
promotional articles on RBOC web sites and very few white papers.
Be Seeing You...
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Matthew Stainforth<>Technical Services Manager<>BrunNet Inc.<>(506)450-4562
Don't rush me, sonny. You rush a miracle maker and you get rotten miracles.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Subject:(usr-tc) Scripting the config of a HARC From: Eric <elorenzo@mediacity.com> Date: 1998-11-20 14:49:07
How can you script the configuration of a HARC? I was told that you can
TFTP the file onto the HARC and then type "do filename"
I'd like to be able to easily batch setup a HARC. With easy prompts,
like:
-What is my IP address?
-What is my netmask?
-What is my gateway?
etc.
Can someone send me some instructions on how to do this and maybe a sample
script?
Eric
---
Eric J. Lorenzo
Sr. Field Engineer
lorenzo@ispchannel.com
v:650.237.1465
http://www.ISPchannel.com
Subject:RE: (usr-tc) padding on trunk groups From: John Powell <jp@packet.ae.usr.com> Date: 1998-11-20 17:33:52
It is not online, actually far from it, but there is a fine book that
actually covers much of this. It is called "Bellcore's Notes on the
Network". Can be purchased from www.bellcore.com. Last I checked it was
"on-sale" for $596. Believe it or not, it is actually worth the money.
Outside of the LSSGR (~$13K), it is the best, most thorough, reference
that I have ever seen. It is very US/Canada centric though.
There is very little else written on the topic.
JP
On Fri, 20 Nov 1998, Stainforth, Matthew (Sys Mgr - BrunNet) wrote:
> On Tuesday, November 20, 2018 8:33 AM, John Powell
> [SMTP:jp@packet.ae.usr.com] wrote:
> > Your note exemplifies why we don't publish the info on pads. That leads
> > to driving the telcos nuts on a subject most users, and certainly most
> > telco people, do not fully understand. A very important note, is that the
> > modems do not detect analog pads, they look like loop loss and are not
> > discernable. The modems can only detect digital pads, and those are
> > really the only ones they are concerned about anyway (as they need to be
> > mathematically compensated for).
>
> Kind of off topic, but does anybody know of any on-line resources for
> getting to know the line provisioning/telco side of things better? I've
> looked for some time and I can't seem to find anything beyond a few
> promotional articles on RBOC web sites and very few white papers.
>
> Be Seeing You...
>
> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
> Matthew Stainforth<>Technical Services Manager<>BrunNet Inc.<>(506)450-4562
> Don't rush me, sonny. You rush a miracle maker and you get rotten miracles.
> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc)Y2K Problem in USR Chassis From: Syed Amiruddin <amir@cyber.net.pk> Date: 1998-11-20 18:44:02
Hi Everybody,
Can anyone confirm me, is Year 2000 problem can effect on USR Access
Servers or not?
Regards,
--
Syed Amiruddin
Sr.Systems Engineer
Network Operations
Cyber Internet Services (Pvt) Ltd.
E-mail: amir@cyber.net.pk
Phone : (92-21)111445566 Ext.232/201
Fax : (92-21)5686745
Subject:Re: (usr-tc) padding on trunk groups From: Aaron Nabil <nabil@spiritone.com> Date: 1998-11-20 20:54:09
John Powell writes...
>It is not online, actually far from it, but there is a fine book that
>actually covers much of this. It is called "Bellcore's Notes on the
>Network". Can be purchased from www.bellcore.com. Last I checked it was
>"on-sale" for $596. Believe it or not, it is actually worth the money.
>
>Outside of the LSSGR (~$13K), it is the best, most thorough, reference
>that I have ever seen. It is very US/Canada centric though.
In my opinion, the Notes chapter on loss plans is better than the
LSSGR "transmission" coverage of the topic. Really the only coherent
explanation of EML and TLP's I've ever seen, nice charts too.
>There is very little else written on the topic.
There is very little "practical" discussion in other texts, although
_any_ modern telecom/communication theory book has a chapter on loss
planning. It's one of those fundemental, building-block things.
--
Aaron Nabil
Subject:Re: (usr-tc)Y2K Problem in USR Chassis From: John Powell <jp@packet.ae.usr.com> Date: 1998-11-21 09:44:11
On Fri, 20 Nov 1998, Syed Amiruddin wrote:
> Hi Everybody,
> Can anyone confirm me, is Year 2000 problem can effect on USR Access
> Servers or not?
Syed,
3Com has a comprehensive Y2K site.
http://www.3com.com/products/yr2000.html
There is specific info on each TCH component, as well as just about every
3Com product, there.
Hope that helps.
JP
Do a show radius
and send me the info.
krish
\ T.S.V. Krishnan \
\ Network System Engineer \ ( : - : )
\ 3Com ............ \
----------------------------------------------/
tkrishna@bubba.ae.usr.com
----------------------------/ http://interproc.ae.usr.com ----/
The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
Any Sufficiently advanced bug is indistinguishable for a feature.
- Rick Kulawiec
On Sat, 21 Nov 1998, System Administrator wrote:
> I have what pretty much amounts to a production installation success story
> that I wanted to share with the list. Note that there are still a couple of
> small problems that I am curious about (see the end of the message), but
> nothing major. One reason for this "success story", is simply to inject some
> positive feedback into the mailing list, and let 3com know that while there
> are definitely "imperfections" (both technical and otherwise) with the Total
> Control systems, they also do seem to have a good core engineering team and
> have done many things _right_ (from my perspective) that no other vendor has
> been able to accomplish.
>
> Yesterday, at one of our remote POPs, a couple older Netserver based chassis
> were replaced with a brand new chassis, including six HDSP cards and one HARC.
> To 3com's benefit, the HARC interface is SIGNIFICANTLY better than the old
> pm-based Netserver one. Things went very smoothly, this was our first rollout
> of v.90 at this POP (we waited until we had the new chassis before rolling it
> out). We've put v.90 in on other POPs, and are quite experienced with the
> sort of trouble to expect (in terms of client modem issues). Everything went
> just about flawlessly. ;) We did have one problem with the second HDSP,
> serving the second PRI in this hunt group. For some reason, the majority of
> callers hitting the card were unable to connect (saw reason: 2 in the syslog),
> we never saw more than 50% of the card actually full. The card was pulled,
> swapped with one at the end of the hunt group, and almost immediately filled.
> I'm not sure what the problem with this card was, I have (1) reloaded template
> 1 from default, (2) reconfig'd, (3) saved to nvram, and (4) reloaded template
> 1 config channels. It's a bit hard to tell if it's still flakey, because now
> it is at the end of the hunt group and only takes calls at peak periods; never
> to full capacity (which is normal based on it's location in the hunt group).
>
> After some minor (non-HARC related) routing issues, everything seems to be
> running perfectly. Our level 3 techs, who have all had _significant_
> experience with PM3s and v.90, expressed the following to me this morning:
> "Boy, those usr racks are MUCH more forgiving with v.90 problems than the
> pm3s".
>
> At this time, my only remaining problem is with authentication (radius). For
> some reason, when I set BOTH the primary and secondary authentication server,
> the HARC begins using the secondary instead of the primary after approximately
> 30 minutes or so. Our secondary radius server is for emergency backup ONLY,
> it does not have all the features that the primary does (however, it does let
> users authenticate), and is only to be used in the event of a critical
> failure on the primary.
>
> a "show authentication" reveals the following (true ips removed to protect the
> innocent):
>
> RADIUS AUTHENTICATION SETTINGS
> Local Authentication is: ENABLED
> Remote Authentication is: ENABLED
> Hint Assigned is: DISABLED
> Primary Server is: aaa.bbb.ccc.ddd
> Primary Destination Port is: 1645
> Secondary Server is: 0.0.0.0
> Secondary Destination Port is: 1645
> Tertiary Server is: 0.0.0.0
> Tertiary Destination Port is: 1645
> Source Port is: 1645
> Retransmission Timeout: 3 seconds
> Max Retranmissions: 10
> Vendor Specific Attribute: DISABLED
> Active Authentication Server: aaa.bbb.ccc.ddd
>
> (note that I have disabled the secondary server for the time being, to avoid
> this problem)
>
> Anyone have any ideas on this?
>
> Thanks!
>
> --
> Jesse Sipprell
> Technical Operations Director
> Evolution Communications, Inc.
> 800-496-4736 (ext 106)
>
> * Finger sysadmin@evcom.net for my PGP Public Key *
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) New Chassis/HiperARC installation success story! (well, almost) From: System Administrator <sysadmin@evcom.net> Date: 1998-11-21 11:12:50
I have what pretty much amounts to a production installation success story
that I wanted to share with the list. Note that there are still a couple of
small problems that I am curious about (see the end of the message), but
nothing major. One reason for this "success story", is simply to inject some
positive feedback into the mailing list, and let 3com know that while there
are definitely "imperfections" (both technical and otherwise) with the Total
Control systems, they also do seem to have a good core engineering team and
have done many things _right_ (from my perspective) that no other vendor has
been able to accomplish.
Yesterday, at one of our remote POPs, a couple older Netserver based chassis
were replaced with a brand new chassis, including six HDSP cards and one HARC.
To 3com's benefit, the HARC interface is SIGNIFICANTLY better than the old
pm-based Netserver one. Things went very smoothly, this was our first rollout
of v.90 at this POP (we waited until we had the new chassis before rolling it
out). We've put v.90 in on other POPs, and are quite experienced with the
sort of trouble to expect (in terms of client modem issues). Everything went
just about flawlessly. ;) We did have one problem with the second HDSP,
serving the second PRI in this hunt group. For some reason, the majority of
callers hitting the card were unable to connect (saw reason: 2 in the syslog),
we never saw more than 50% of the card actually full. The card was pulled,
swapped with one at the end of the hunt group, and almost immediately filled.
I'm not sure what the problem with this card was, I have (1) reloaded template
1 from default, (2) reconfig'd, (3) saved to nvram, and (4) reloaded template
1 config channels. It's a bit hard to tell if it's still flakey, because now
it is at the end of the hunt group and only takes calls at peak periods; never
to full capacity (which is normal based on it's location in the hunt group).
After some minor (non-HARC related) routing issues, everything seems to be
running perfectly. Our level 3 techs, who have all had _significant_
experience with PM3s and v.90, expressed the following to me this morning:
"Boy, those usr racks are MUCH more forgiving with v.90 problems than the
pm3s".
At this time, my only remaining problem is with authentication (radius). For
some reason, when I set BOTH the primary and secondary authentication server,
the HARC begins using the secondary instead of the primary after approximately
30 minutes or so. Our secondary radius server is for emergency backup ONLY,
it does not have all the features that the primary does (however, it does let
users authenticate), and is only to be used in the event of a critical
failure on the primary.
a "show authentication" reveals the following (true ips removed to protect the
innocent):
RADIUS AUTHENTICATION SETTINGS
Local Authentication is: ENABLED
Remote Authentication is: ENABLED
Hint Assigned is: DISABLED
Primary Server is: aaa.bbb.ccc.ddd
Primary Destination Port is: 1645
Secondary Server is: 0.0.0.0
Secondary Destination Port is: 1645
Tertiary Server is: 0.0.0.0
Tertiary Destination Port is: 1645
Source Port is: 1645
Retransmission Timeout: 3 seconds
Max Retranmissions: 10
Vendor Specific Attribute: DISABLED
Active Authentication Server: aaa.bbb.ccc.ddd
(note that I have disabled the secondary server for the time being, to avoid
this problem)
Anyone have any ideas on this?
Thanks!
--
Jesse Sipprell
Technical Operations Director
Evolution Communications, Inc.
800-496-4736 (ext 106)
* Finger sysadmin@evcom.net for my PGP Public Key *
Subject:RE: (usr-tc) padding on trunk groups From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-21 14:27:02
I re-read my response to your e-mail. I realized it was much harsher than
it should have been.
Thanks John. I appreciate your help. I apologize for my chaffing response
earlier. Your e-mail to me was less than cordial and I'm afraid my return
was no better.
Your knowledge and help has been tremendous in the past. The problem I have
been troubleshooting here is a tough one. I never professed to be a telco
or v.90 expert as I think you thought I was trying to indicate. If I was I
would have never used this list for the problem. Also, I never expected the
padding to be doing anything more than take down the connect speed a couple
clicks. My questions about echo cancellation were curiosity questions only.
I never thought I was on the right track with that either.
I hope you will forgive my ignorance with this kind of a problem. It is
definitely way above my head. I firmly believe the problem is definitely
with the local telco and not the PRI. I simply do not know where to take it
from here. The telco needs guidance I can't give them.
I hope you will see your way to forgive my rash response. This entire thing
has been very frustrating. Getting nothing but negative remarks about my
questions instead of answers to anything I asked just kind of added to it.
I'm sorry. :(
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of John Powell
>Sent: Tuesday, November 20, 2018 5:34 PM
>To: 'usr-tc@lists.xmission.com'
>Subject: RE: (usr-tc) padding on trunk groups
>
>
>
>It is not online, actually far from it, but there is a fine book that
>actually covers much of this. It is called "Bellcore's Notes on the
>Network". Can be purchased from www.bellcore.com. Last I checked it was
>"on-sale" for $596. Believe it or not, it is actually worth the money.
>
>Outside of the LSSGR (~$13K), it is the best, most thorough, reference
>that I have ever seen. It is very US/Canada centric though.
>
>There is very little else written on the topic.
>
>JP
>
>On Fri, 20 Nov 1998, Stainforth, Matthew (Sys Mgr - BrunNet) wrote:
>
>> On Tuesday, November 20, 2018 8:33 AM, John Powell
>> [SMTP:jp@packet.ae.usr.com] wrote:
>> > Your note exemplifies why we don't publish the info on pads.
>That leads
>> > to driving the telcos nuts on a subject most users, and certainly most
>> > telco people, do not fully understand. A very important note,
>is that the
>> > modems do not detect analog pads, they look like loop loss and are not
>> > discernable. The modems can only detect digital pads, and those are
>> > really the only ones they are concerned about anyway (as they
>need to be
>> > mathematically compensated for).
>>
>> Kind of off topic, but does anybody know of any on-line resources for
>> getting to know the line provisioning/telco side of things better? I've
>> looked for some time and I can't seem to find anything beyond a few
>> promotional articles on RBOC web sites and very few white papers.
>>
>> Be Seeing You...
>>
>> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>> Matthew Stainforth<>Technical Services Manager<>BrunNet
>Inc.<>(506)450-4562
>> Don't rush me, sonny. You rush a miracle maker and you get
>rotten miracles.
>> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
>>
>>
>>
>> -
>> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
>> with "unsubscribe usr-tc" in the body of the message.
>> For information on digests or retrieving files and old messages send
>> "help" to the same address. Do not use quotes in your message.
>>
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) ospf From: Brian <signal@shreve.net> Date: 1998-11-22 03:07:08
Does anyone know if OSPF is slated for ARC v4.2? Just trying to get a
handle on the timeline.
Brian
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:RE: (usr-tc) ospf From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-22 06:02:35
It's slated for the next full featured release of HiPer ARC. Don't know the
timeline. Within the next 2 months i would think, but don't quote me.
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian
>Sent: Sunday, November 22, 1998 3:07 AM
>To: USRobotics TC Mailing List
>Subject: (usr-tc) ospf
>
>
>Does anyone know if OSPF is slated for ARC v4.2? Just trying to get a
>handle on the timeline.
>
>Brian
>
>
>--------------------------------------------------------------------------
>Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
>Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
>signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
>(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) HiPer ARC 4.1.72 - 7 From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-22 06:10:20
I just installed 4.1.72 - 7 on one of our chassis. The HiPer DSP's have
1.2.5. We have had NMC's crash before so we had NMC chassis awareness
disabled and statically configured the HiPer ARC for the DSP's we want it to
manage. Here's what happened. Once the HiPer ARC rebooted with the new
code all of our HiPer DSP's went into a continual reboot. They rebooted
several times before I pulled the HiPer ARC and they stopped. I waited
several minutes, put the HiPer ARC back in, and they all started rebooting
again. I connected to the console of the HiPer ARC and typed delete config.
As the HiPer ARC rebooted the DSP's stopped rebooting. I ran through quick
setup and kept all defaults. I kept NMC chassis awareness enabled this
time. The HiPer DSP's did not reboot this time. Calls started flowing in.
Just for the hell of it I disabled NMC chassis awareness and configured
things statically again. I saved all and rebooted. When the HiPer ARC came
up the HiPer DSP's went into a perpetual reboot state. I, of course, am
using NMC chassis awareness now. At least until we figure out what is
causing this. Has anyone else seen this problem?
Subject:Re: (usr-tc) Interested in Linux on the Netserver? From: MegaZone <megazone@megazone.org> Date: 1998-11-22 10:38:21
Once upon a time Jeff Mcadams shaped the electrons to say...
>>|I tell you what.
>>|I get better quake on a livingston portmaster 2e and an anolog modem
>>|attach than I do on a netserver 486 with pri lines. The portmaster has
>>|an AMD 386 DX40.
>>How many ports??
>A 2e has 30.
The PortMaster-2 family, the PM-25, the IRX family, and the ORs all run
AMD 386 CPUs. The fastest they are clocked is 33MHz, some older units
are 25MHz. The high port density would be a PM-2e-30 with 30 async serial
ports at 115.2Kbps (officially - 230.4Kbps works but is unofficial). Or
a PM-2ei with 2 5BRI cards for 15BRI ports, or 30 B channels.
The IRX-114 router has 2 T1/W1 ports, 2 64K sync ports, 1 async serial
port, an 1 Ethernet. However, you really can't use all sync ports at
once. It generally will handle the 2 T1/E1 ports as they are DMA. The
64K ports are interrupt driven so they chew up CPU when used along with
the faster ports.
The PM-3 has an AMD 486-66 as its CPU. With that it easily handles the
two T1/E1/PRI WAN ports. Which can be dial or leased lines. Additionally
there is a T1 WAN card which can be added to provide for a T1 leased line
for a POP-in-a-box approach. The rough estimate from Lucent is that the
CPU in the PM-3 is 6x that as the PM-2 - being a 486 and faster, AND having
less to do. In the PM-3 many of the tasks the CPU used to do are offloaded
to the HDLC controllers, DSPs, modem controllers, etc. PPP for example
is performed on the modem cards (for modem calls). For ISDN I think it
is done in the HDLC controller. Of course, the PM-3 also does OSPF and
BGP, and in the latest betas NAT, NFAS, L2TP, IPIP, etc. Stac compression
and IPSec encryption (the latter is in beta) are handled by daughter cards.
The PM-4 is a much more distributed system. Each card has a K5 or K6 I
believe, and there are many more CPUs dedicated to tasks. The main ethernet
daughter card has 2 DEC RISC processors dedicated just to driving the
Ethernet - one is inbound the other is outbound.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Subject:Re: (usr-tc) Interested in Line on the Netserver? From: MegaZone <megazone@megazone.org> Date: 1998-11-22 10:40:55
Once upon a time Jeff Mcadams shaped the electrons to say...
>OK, just for yucks, I went downstairs and found a PM-25 that we had on a
>shelf here (only 25 ports, but same architecture as the 2E from what I
Yes, it is pretty much just like the PM-2 with different serial port HW.
>Can anyone confirm or deny if the 2E has an external cache? I don't
>believe it does, but a confirmation would be nice.
I'm not 100% sure, but AFAIK it does not.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Subject:Re: (usr-tc) Interested in Line on the Netserver? From: MegaZone <megazone@megazone.org> Date: 1998-11-22 10:42:38
Once upon a time Brian K McIntire shaped the electrons to say...
>The fact that the 2e supports only 30 may partially explain how it can
>handle things a little better. It doesn't have to do the work the USR
>NetServer does
A PM-3 has an AMD 486-66 and it supports 2 T1/E1/PRI ports *AND* a T1 WAN
card. Oh - while doing OSPF and BGP.
Just to compare similar hardware.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Subject:Re: (usr-tc) Interested in Line on the Netserver? From: MegaZone <megazone@megazone.org> Date: 1998-11-22 10:50:45
Once upon a time Jeff Mcadams shaped the electrons to say...
>A 486 with 48 ports should *clobber* a 386 with 30 ports in
>performance...*particularly* when you consider the offloading of ppp
>framing into the quad cards! This shouldn't even be *close*!
Especially when you consider that this is a 386-33 (sometimes -25) with
30 ports vers a 486-100 with 48.
A 486 that is 3 (or 4) times as fast with 1.6x the number of ports.
I'm following this Linux-on-Netserver effort, it is intriguing to say
the least. If a HW engineer got involved perhaps they could logic-probe
the interfaces to get around not having a published spec.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Once upon a time Aaron Nabil shaped the electrons to say...
>If you are running NFAS, by all means use a backup D channel. A much,
>much better solution is never to use NFAS.
>
>Fate sharing good. NFAS bad.
This is myopic at best. NFAS is just fine - it just depends on where you
use it. NFAS with 2 lines? Stupid. NFAS with 20 lines? NOT so stupid.
With a primary D and backup D you just gained *18* ports. Ok, if they got
NFAS working on a HiPer ARC chassis you could gain 12 ports across the
chassis.
If multichassis NFAS enters the picture, as Lucent is testing, then you
could have more ports involved. The maximum varies by the switch being
used. 20-24 is the number I see most often.
And on a chassis like a PM-4 or CVX-1800 you can have enough lines in
one chassis to have *multiple* NFAS groups.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Subject:Re: (usr-tc) Interested in Line on the Netserver? From: Jeff Mcadams <jeffm@iglou.com> Date: 1998-11-22 17:00:13
Thus spake MegaZone
>Once upon a time Jeff Mcadams shaped the electrons to say...
>>Can anyone confirm or deny if the 2E has an external cache? I don't
>>believe it does, but a confirmation would be nice.
>I'm not 100% sure, but AFAIK it does not.
Which pretty much blows the theory out of the water of not having an
external cache being the problem with "Quake Lag."
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
On Sun, 22 Nov 1998, MegaZone wrote:
> And on a chassis like a PM-4 or CVX-1800 you can have enough lines in
> one chassis to have *multiple* NFAS groups.
How does the CVX-1800 compare to the PM-4? Ive never heard of Aptis.
-Dan
Subject:Re: (usr-tc) HiperDSP possible almost DOA From: Tatai SV Krishnan <tkrishna@bubba.ae.usr.com> Date: 1998-11-22 18:32:56
On Sun, 22 Nov 1998, Douglas Palmer wrote:
> I just went through the installation and configuration of our new HiperDSP
> and HiperARC cards. The HiperARC seems to work fine (though I cannot get it
> to display the connection banner file) and the HiperDSP seemed ok for the
> first 10 hours.
There are two banners files - The first one which says Welcome to 3com
....
That you will always see - you can also have a banner on connection - and
that is something like PPP starting from etc - Which one are you talking
about. The first one is default which should always be displyed - unless
you removed/changed it. The second one is something that you must setup
-
You must enable auth hint assigned and set the banner you want.
>
> After 10 hours, the HiperDSP started rebooting (dropping users and busying
> out all lines on the PRI for about 30 seconds) every two hours or so. After
> 10 hours of this, it stopped working all together. The NAC no longer
> responds to any stimulation beyond a red RUN light. It will not take a
> reflash or any other input at all (it shows up as a red question mark on
> the TCM diagram).
>
First check the code on the DSP - If you cannot then try reloading the
code throught the console. you can use zmodem to download the code. For
the most part you should see some type of error message on the cosole -
for all you may know it chould be a config problem in terms of ownership.
Check the code - if its 1.2.5 - then its a good stable code
if it is anything other then you are better of loading 1.2.5
Check the hiper arc and make sure that you either have static or nmc
chassis awarness enabled.
This should solve your problem
regards
krish
> I'll be calling in tomorrow for an RMA to get a replacement -- unless
> someone here has an idea what might be done. It seemed fine for the first
> 10 hours. Better connect rates for my v.90 users and good throughput. For
> now, I am back on the Digital Quads.
>
> Thanks in advance for any pointers.
>
> DCP
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) Interested in Line on the Netserver? From: Dan Hollis <goemon@sasami.anime.net> Date: 1998-11-22 18:58:29
On Sun, 22 Nov 1998, MegaZone wrote:
> I'm following this Linux-on-Netserver effort, it is intriguing to say
> the least. If a HW engineer got involved perhaps they could logic-probe
> the interfaces to get around not having a published spec.
How would Lucent feel about Linux-on-PM2 or Linux-on-PM3?
-Dan
Subject:(usr-tc) HiperDSP possible almost DOA From: Douglas Palmer <palmer@usdc-edny.com> Date: 1998-11-22 19:01:21
I just went through the installation and configuration of our new HiperDSP
and HiperARC cards. The HiperARC seems to work fine (though I cannot get it
to display the connection banner file) and the HiperDSP seemed ok for the
first 10 hours.
After 10 hours, the HiperDSP started rebooting (dropping users and busying
out all lines on the PRI for about 30 seconds) every two hours or so. After
10 hours of this, it stopped working all together. The NAC no longer
responds to any stimulation beyond a red RUN light. It will not take a
reflash or any other input at all (it shows up as a red question mark on
the TCM diagram).
I'll be calling in tomorrow for an RMA to get a replacement -- unless
someone here has an idea what might be done. It seemed fine for the first
10 hours. Better connect rates for my v.90 users and good throughput. For
now, I am back on the Digital Quads.
Thanks in advance for any pointers.
DCP
Once upon a time Dan Hollis shaped the electrons to say...
>On Sun, 22 Nov 1998, MegaZone wrote:
>> And on a chassis like a PM-4 or CVX-1800 you can have enough lines in
>> one chassis to have *multiple* NFAS groups.
>How does the CVX-1800 compare to the PM-4? Ive never heard of Aptis.
Have you heard of Nortel? ;-) Aptis was a start-up, the CVX-1800 was
their first product. Nortel bought them before it shipped, it is a Nortel
product now.
Overall it is similar. It is a midplane, the PM-4 is a backplane. To
generalize I'd say the Pm-4 was designed more for ISPs and the CVX more for
telcos. The Pm-4 has internal slots for AC power supplies if needed, or
it can run on native DC. The CVX is DC only, with a seperate AC-DC converter
chassis available if needed. That gives it a few more slots. Both have
similar bandwith per slot. And since the CVX is a bit taller they have
similar rack densities.
Now, at Interop I was told by the Nortel Networks folks that since they
now have Bay as part of thier group that the CVX-1800 will ONLY be supporting
dial access needs. And all of the router needs will be based on the Bay
Versalar routing platforms. Whereas Lucent has indicated they intend to
exploit the flexibility of the PM-4 to handle xDSL, ATM, SONET, etc - with
OC3 and OC12 capabilities that are designed into the chassis. So it looks
like the PM-4 will be a more versatile unit.
Right now it is a large access box, but that's what is out now just at
first customer ship. More interface types will be added in time.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Subject:Re: (usr-tc) Interested in Line on the Netserver? From: MegaZone <megazone@megazone.org> Date: 1998-11-22 20:22:49
Once upon a time Dan Hollis shaped the electrons to say...
>On Sun, 22 Nov 1998, MegaZone wrote:
>> I'm following this Linux-on-Netserver effort, it is intriguing to say
>> the least. If a HW engineer got involved perhaps they could logic-probe
>> the interfaces to get around not having a published spec.
>How would Lucent feel about Linux-on-PM2 or Linux-on-PM3?
I doubt they'd release the hardware specs for them (maybe the PM-2 when
it end of lifes, but then that's a pretty basic unit) but I don't think
they'd try to stop it.
Getting Linux to handle the proprietary systems like the backplane and
modem cards in the PM-3 would be the hard part.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Once upon a time Mark R. Lindsey shaped the electrons to say...
>`Dial-up servers' are routers had darned better act like it.
They route, but it is useful not to call them routers. It is generally
accepted that a 'router' is something that has at least on sync interface
and that is what it primarily is designed to drive. Normally T1/E1 or
faster, but it can be fractional thereof. The only exceptions really
are "ISDN routers" and the general "SOHO router" catagory.
I understood what they meant - if you're looking for T1/E1, DS-3c, OC-3,
or OC-12 for leased lines then you don't look at the CVX-1800. If you
want ATM, SONET, etc - same deal. You want to look at their other 'router'
platforms.
If you are looking for ISDN or modem access, and likely VOIP/FOIP in the
future, then the CVX is the high end platform.
I suspect the CVX won't be doing BGP either because of this - while the
PM-4 does because leased line routing is something it was designed to do.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Subject:Re: (usr-tc) v.90 connect problems From: Mark R. Lindsey <mark@vielle.datasys.net> Date: 1998-11-23 00:02:47
Quoth MZ:
: Now, at Interop I was told by the Nortel Networks folks that since they
: now have Bay as part of thier group that the CVX-1800 will ONLY be supporting
: dial access needs. And all of the router needs will be based on the Bay
: Versalar routing platforms.
<rant>
It still disappoints me to hear of manufacturers' spokesmen making this
broad, sweeping distinctions between routers and network-access equipment.
Router NAS
~~~~~~ ~~~
* Forwards packets * Forwards packets
* Has lots of interfaces * Has lots of interfaces
* Requires significant * Requires excruciating
diagnostics diagnostics
Somehow, the radically-different bit rates common on router interfaces
(e.g., 100 mbit) versus NAS interfaces (e.g., 64 kbit) turns some into
babbling dullards.
`Dial-up servers' are routers had darned better act like it.
</rant>
Subject:Re: (usr-tc) Interested in Line on the Netserver? From: Brian Elfert <brian@citilink.com> Date: 1998-11-23 09:10:14
On Sun, 22 Nov 1998, Dan Hollis wrote:
> On Sun, 22 Nov 1998, MegaZone wrote:
> > I'm following this Linux-on-Netserver effort, it is intriguing to say
> > the least. If a HW engineer got involved perhaps they could logic-probe
> > the interfaces to get around not having a published spec.
>
> How would Lucent feel about Linux-on-PM2 or Linux-on-PM3?
I'm sure Lucent wouldn't like it at all.
I think the PM2/3 and the Netserver card are quite different. The
Netserver is pretty much a PC on a card. The PM2/3 do use x86 chips, but
I don't think they are really a full PC. I think it'd be a lot harder to
load something besides ComOS on the PM2/3 whereas the Netserver has the
PCSDL thing.
People aren't clamoring for a new OS on the PM2/3. ComOS already has
OSPF, and it doesn't seem to have the Quake lag problems like the
Netserver. Lucent is still developing code for the PM2/3 while the
Netserver is dead after Dec 31st. (There are rumors of PM2 software
development ceasing after ComOS 3.9.)
Brian
Subject:RE: (usr-tc) New Chassis/HiperARC installation success story! (well, almost) From: Mike Wronski <mike@coredump.ae.usr.com> Date: 1998-11-23 09:51:49
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of System
|Administrator
|Sent: Saturday, November 21, 1998 10:13 AM
|To: usr-tc@lists.xmission.com
|Subject: (usr-tc) New Chassis/HiperARC installation success story!
|(well, almost)
|
|At this time, my only remaining problem is with authentication (radius). For
|some reason, when I set BOTH the primary and secondary authentication server,
|the HARC begins using the secondary instead of the primary after approximately
|30 minutes or so. Our secondary radius server is for emergency backup ONLY,
|it does not have all the features that the primary does (however, it does let
|users authenticate), and is only to be used in the event of a critical
|failure on the primary.
|
This is due to your "AUTHENTICATION_ALGORITHM". If you do a 'show radius' it is
probably set to "ROUND_ROBIN".
In this setting: If contact is lost with the primary, it will swap primary and
secondary, making the secondary the first attempt. Only when the secondary goes
down will the swap take place again..
For your requirements set it to "FALL_THROUGH". This way the primary is tried
every time and only tries the others if no response comes from the primary.
-M
Subject:Re: (usr-tc) [isp-services] FS:USR TOTAL CONTROL UNITS (fwd) From: Frank Basso <frank@got.net> Date: 1998-11-23 12:23:09
Yeah if you use ANALOG CO Trunks and NOT PRI's and you like Token ring :)
-----Original Message-----
>wow, 2400 dollars.
>
>
>--------------------------------------------------------------------------
>Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
>Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
>signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
>(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
>
>---------- Forwarded message ----------
>Date: Mon, 23 Nov 1998 13:06:30 -0500
>From: "Andrew:PC Global, Inc." <pcglobal@gate.net>
>To: isp-equipment@isp-equipment.com
>Cc: isp-services@ispc.org
>Subject: [isp-services] FS:USR TOTAL CONTROL UNITS
>
>Had to drop the price due to remain competitive. Please note that the
>prices on these will go back up. Now selling for $2500 for a limited time
>only!!
>
>>I Just received 10 more chassis. These may be the last ones I get for a
>>while. Hope they can save/make you all some $$$!!
>>
>>
>>60 PORT ANALOG/DIGITAL
>>
>>WE HAVE COMPLETE CHASSIS UNITS EACH INCLUDING:
>>
>>(1)CHASSIS
>>
>>(1)FAN TRAY
>>(1) NET SERVER CARD REV.5 486DX/4 16MB USR#001369-00
>>(1) NETWORK MANAGEMENT CARD REV.3 USR#000978-00
>>(15) QUAD V.34 ANALOG/DIGITAL MODEM REV.2 USR#000793-04 (60 Analog Ports)
>>(15) NIC ANALOG ADAPTERS USR#000385-00
>>(2) AC PSU 45A CARD
>>
>>USR BUILD DATE 10/28/96
>>
>>MORE INFO:
>>
>>-THESE UNITS ARE TOKEN RING
>>
>>-MODEMS ARE UPGRADABLE TO V.90 56K VIA USR/3COM
>>-UNITS ARE ANALOG/DIGITAL, A PRI-T1 CARD IS NEEDED FOR DIGITAL
>>
>>-UNITS GUARANTEED WORKING/JUST TAKEN OUT OF SERVICE. NATIONWIDE ISP WENT
>OUT
>>OF BUSINESS.
>>-EQUIPMENT IS FREE AND CLEAR OF ANY LEINS OR INCUMBRANCES
>>
>>
>>WE HAVE SOME INDIVIDUAL PARTS AVAILABLE.
>>-NET SERVER CARDS
>>-NET MGMT CARDS
>>-ANALOG/DIGITAL MODEM CARDS
>>-POWER SUPPLIES
>>-FAN TRAYS
>>----NO PRI/T1 CARDS LEFT
>>
>>PRICE:
>>>>>$2,400 EACH
>>QUANTITY PRICING AVAILABLE (3 OR MORE)
>>
>>PAYMENT: CREDIT CARD (VS/MC/AMX/DISC), COD cert funds
>>
>>Thank you for your time and have a great day!
>>
>> Andrew Shlensky
>> ****************************
>> PC Global, Incorporated
>> (305) 667-2111 TEL
>> (305) 667-3636 FAX
>> URL: http://www.pcglobal.net
>> E-MAIL: andrew@pcglobal.net
>>
>>
>>
>>
>
>
>---------------------------------------------------------------------
>To unsubscribe, e-mail: isp-services-unsubscribe@ispc.org
>For additional commands, e-mail: isp-services-help@ispc.org
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) Interested in Line on the Netserver? From: MegaZone <megazone@megazone.org> Date: 1998-11-23 14:05:07
Once upon a time Brian Elfert shaped the electrons to say...
>Netserver is dead after Dec 31st. (There are rumors of PM2 software
>development ceasing after ComOS 3.9.)
That wouldn't surprise me. ComOS 3.9 has lots of nifty things like NAT
and L2TP. Really, the only things not in 3.9 I can think of off hand that
might be wanted on an old async unit like the PM-2 would be mulitcast
support and extended SNMP support (they seem to notch SNMP up quite a bit
in each new release, but I don't think 3.9 will be quite 'complete').
If they can work out the few applicable bugs the current code - the oddball
BRI issues that a handful of people can't seem to shake, trouble with the
new OSPF on demand links some have reported, etc - it looks like it would
be a nice high note to end on. And the PM-2 family has been around a LONG
time - the last new HW in that line were the ISDN BRI pieces which came
out in late 1995 and early 1996. The basic design itself is rather older
than that. Still works though. I think they might freeze software (save
for bug fixes and security patches, if any) but keep the HW line alive since
it does what it was designed to do well enough.
I will say that ComOS 3.9b and PMVision 1.3b are lots of fun to play with
together. :-)
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Subject:(usr-tc) [isp-services] FS:USR TOTAL CONTROL UNITS (fwd) From: Brian <signal@shreve.net> Date: 1998-11-23 14:06:16
wow, 2400 dollars.
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
---------- Forwarded message ----------
Cc: isp-services@ispc.org
Had to drop the price due to remain competitive. Please note that the
prices on these will go back up. Now selling for $2500 for a limited time
only!!
>I Just received 10 more chassis. These may be the last ones I get for a
>while. Hope they can save/make you all some $$$!!
>
>
>60 PORT ANALOG/DIGITAL
>
>WE HAVE COMPLETE CHASSIS UNITS EACH INCLUDING:
>
>(1)CHASSIS
>
>(1)FAN TRAY
>(1) NET SERVER CARD REV.5 486DX/4 16MB USR#001369-00
>(1) NETWORK MANAGEMENT CARD REV.3 USR#000978-00
>(15) QUAD V.34 ANALOG/DIGITAL MODEM REV.2 USR#000793-04 (60 Analog Ports)
>(15) NIC ANALOG ADAPTERS USR#000385-00
>(2) AC PSU 45A CARD
>
>USR BUILD DATE 10/28/96
>
>MORE INFO:
>
>-THESE UNITS ARE TOKEN RING
>
>-MODEMS ARE UPGRADABLE TO V.90 56K VIA USR/3COM
>-UNITS ARE ANALOG/DIGITAL, A PRI-T1 CARD IS NEEDED FOR DIGITAL
>
>-UNITS GUARANTEED WORKING/JUST TAKEN OUT OF SERVICE. NATIONWIDE ISP WENT
OUT
>OF BUSINESS.
>-EQUIPMENT IS FREE AND CLEAR OF ANY LEINS OR INCUMBRANCES
>
>
>WE HAVE SOME INDIVIDUAL PARTS AVAILABLE.
>-NET SERVER CARDS
>-NET MGMT CARDS
>-ANALOG/DIGITAL MODEM CARDS
>-POWER SUPPLIES
>-FAN TRAYS
>----NO PRI/T1 CARDS LEFT
>
>PRICE:
>>>>$2,400 EACH
>QUANTITY PRICING AVAILABLE (3 OR MORE)
>
>PAYMENT: CREDIT CARD (VS/MC/AMX/DISC), COD cert funds
>
>Thank you for your time and have a great day!
>
> Andrew Shlensky
> ****************************
> PC Global, Incorporated
> (305) 667-2111 TEL
> (305) 667-3636 FAX
> URL: http://www.pcglobal.net
> E-MAIL: andrew@pcglobal.net
>
>
>
>
To unsubscribe, e-mail: isp-services-unsubscribe@ispc.org
For additional commands, e-mail: isp-services-help@ispc.org
Subject:(usr-tc) connecting a cisco router to total control ISDN From: Steve Lalonde <steve@enta.net> Date: 1998-11-23 17:41:48
Hi All
I have a strange one
Cisco 1603 router trying to connect to a totalcontrol chassis
we can get the cisco to connect and authenticate using pap
but we never see any accounting data in the radius logs so the cisco just
hangs up
is there anything strange about a cisco connecting to usr kit?
there are 2 netserver based and 1 hyper based racks here and all do
the same thing.
TIA
Steve Lalonde
Systems Manager
ENTANET International Ltd
Subject:Re: (usr-tc) HiperDSP possible almost DOA From: Douglas Palmer <palmer@usdc-edny.com> Date: 1998-11-23 17:43:11
At 06:32 PM 11/22/1998 -0600, you wrote:
>> After 10 hours, the HiperDSP started rebooting (dropping users and busying
>> out all lines on the PRI for about 30 seconds) every two hours or so. After
>> 10 hours of this, it stopped working all together. The NAC no longer
>> responds to any stimulation beyond a red RUN light. It will not take a
>> reflash or any other input at all (it shows up as a red question mark on
>> the TCM diagram).
>>
>
>First check the code on the DSP - If you cannot then try reloading the
>code throught the console. you can use zmodem to download the code. For
>the most part you should see some type of error message on the cosole -
>for all you may know it chould be a config problem in terms of ownership.
The code is 1.2.5 -- it came with a slightly older rev and I updated it
when the board arrived. It will no longer take a flash reload from the console.
>Check the hiper arc and make sure that you either have static or nmc
>chassis awarness enabled.
Chassis awareness is enabled and the ownership is good. Like I said -- it
worked fine for 10 hours, then started to go down hill. Now, it will not
even go through the boot sequence when the rack is powered on. I'll try
reflashing again tomorrow.
DCP
Subject:RE: (usr-tc) v.90 connect problems From: Robert von Bismarck <rvb@petrel.ch> Date: 1998-11-23 18:15:58
For what it's worth, here's the worst and the trickiest that happened at
our site(s)
- Cable impendance not right (ethernet cable is no good, you need 120ohm
STP)
- DSP hangs (dreaded sight : yellow board in TCM, one solution :
hardware reset)
- Switch sends fluctuating signals down the PRI lines... this does weird
things to the boards
- Wrong Fibermux plugged in (channelized E1 instead of E1 PRI, funny
thing it's the same box the telco gives us, it's just a dip switch to
set...)
That's all for the DSP's
- Too many SNMPget on the NMC will reboot the board (version 5.5.5) (is
this a registered bug/feature ?)
That's all for the NMC's
No weird stuff on the ARC
Hope this helps,
Robert
> -----Original Message-----
> From: Jeff Mcadams [SMTP:jeffm@iglou.com]
> Sent: vendredi, 20. novembre 1998 15:25
> To: usr-tc@lists.xmission.com
> Subject: Re: (usr-tc) v.90 connect problems
>
> Thus spake Brian K McIntire
> >>Thus spake Brian K McIntire
> >>>Not sure where you get your info from but it doesn't sound
> >>informed. I too
> >>>have seen trouble on a span due to a looped repeater. However,
> >>we have seen
> >>>more problems with d-channels going down.
>
> >>Lucky you...I've only had a d channel go down independantly of the
> span
> >>once....I've had span's go down for other reasons myriads of times.
>
> >Out of curiosity, what problems have you experienced with PRI's going
> down.
> >Auto-busy outs at the switch; fibr cuts; loss of timing; etc..
>
> Oh man...I don't think we've ever had the same problem twice... :)
>
> These are just ones that I happen to remember...
> fuse blowing on the DSX, wire at a punch-down block coming loose (both
> 66 blocks and 110 blocks), having an rj-45 come out of its connector,
> dual-pri card loosing its config, probably others that I'm forgetting
> about here, but that's a good start.
> --
> Jeff McAdams Email: jeffm@iglou.com
> Head Network Administrator Voice: (502) 966-3848
> IgLou Internet Services (800) 436-4456
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
Subject:(usr-tc) multiple default routes From: Brian <signal@shreve.net> Date: 1998-11-24 15:26:54
Is their a way on the ARC to declare multiple default routes for
redundancy? something like, so that it uses the first default route, but
if that fails it tries the second?
Brian
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:Re: (usr-tc) Seeking a way to 'hide' users from "list connections" From: Tatai SV Krishnan <tkrishna@bubba.ae.usr.com> Date: 1998-11-24 16:38:58
On Tue, 24 Nov 1998, T.J. Weber wrote:
> Okee, last night I posted this to one of 3com's internal newsgroups:
>
> ---- begin ----
>
> This must seem a little odd, but I would like to know if there is a way
> I can hide users from showing up in the "list connections". I use
> RADIUS authentication, and I am wondering if there is a configuration
> value I can set for the specific users in my RADIUS 'users' file that
> will be sent back to my TC telling the HiPer not to list that user in
> *ANY* records (specifically the "list connections" output).
NO. There is no such options to hide users.
If you want however you can set ppp passthrough meaning you can disable
authentication for users and use the default user for all calls for say
ppp. This way the only user who will be seen in the hIper arc is always
the default user
krish
\ T.S.V. Krishnan \
\ Network System Engineer \ ( : - : )
\ 3Com ............ \
----------------------------------------------/
tkrishna@bubba.ae.usr.com
----------------------------/ http://interproc.ae.usr.com ----/
The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
Any Sufficiently advanced bug is indistinguishable for a feature.
- Rick Kulawiec
>
> Using RADIUS would be my best guess in this case, unless there is a way
> I can specify users to "hide" set directly via the HiPer config prompt.
>
> ---- end ----
>
> No responce from 3com quite yet on this matter, three cents to the dollar
> odds that they won't respond with anything at all.
>
> I setup a Total Control box for my client in Chicago. They are a fairly
> big law firm that is providing internet access to their employees. I am
> not the only one with access to the total control box, there is someone
> from their IS department that monitors connections on the box itself.
> However, some of the partners in the firm don't want to be known that they
> are online at a specific time (strange request if you ask me, i think they
> don't want the guys in the IS dept. to know they are online).
>
> I also have another question regarding the TC boxes. In my configuration
> for the total control box, i set aside a pool of 24 IP addresses for
> dynamic assignment (we only have one operational DSP card). However,
> there are about 10 other users that I would like to assign a DYNAMIC IP
> from another pool (NOT the pool assigned to the total control). Once
> again, I think that this is something I have to setup in RADIUS, but I'm
> not sure how to do it. Please advise!
>
> So if anyone knows a simple solution to these problems that would be
> great!
>
> Thanks in advance,
> --t.j.
>
>
> \|||/
> (o o)
> ooO-(_)-Ooo----------------------------------------------------------
> T.J. Weber | Oxymoron: Microsoft Works
> Interplanetary Media |------------------------------------
> phone: 847.205.5200 | WARNING: objects in mirror are
> fax: 847.205.5201 | dumber than they appear.
> e-mail: admin@ipmedia.net |------------------------------------
> web: http://www.ipmedia.net | Out of my mind. Back in 5 minutes.
>
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> Two things are infinite: the universe and human stupidity;
> and I'm not so sure about the universe. (Albert Einstein)
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) new user log in problem with v.90 From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-24 16:45:18
I saw this happen when the time on my server was set incorrectly and then
set correctly. All users that had been added before the correction had
different info in them. It wouldn't hurt to check time on everything
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Gorkem Yuksel
>Sent: Tuesday, November 24, 1998 3:45 PM
>To: 'usr-tc@lists.xmission.com'
>Subject: (usr-tc) new user log in problem with v.90
>
>
>Hi!
>Everybody.
>
>I am a new guy on the block.(small ISP)
>
>I am having some problems with new user's logging in.
>
>It does not allow to logging in right away when it comes to new users.
>
>Sometimes it started working 7 hours later, 1.5 hours later depending on
>the Hiper DSP's mood, I guess.
>
>But all the old users connecting OK. All the new users you key in today
>will be ok by tommorrow.
>
>I have TCM, Hiper DSP cards, NMC card, Hiper ARC card, about 6 month old.
>We upgraded to v.90 a month ago, it worked fine I guess.
>(or I did not have customers who wanted an instant setup???)
>
>I have system release 3.3.
>Hiper ARC 1.1.11
>Hiper DSP T1/PRI 1.2.5
>NMC 16Meg 486 5.5.5
>TCM Windows 5.5.1
>
>I e-mailed for help to 3Com but no answer for 3 days...
>
>I do not have tech support contract...too much money..
>
>It started happening last Sat and Sun...Mon and Tue.. which is today.
>I had one new customer yesterday.. it did not rscognize his username and
>password right away, but started working about one and half hour later.
>
>I have another new one keyed in 14:33 pm today, it is not working yet
>(16:37).
>I just keep typing in his user name and password by using dial up every 15
>to 30 mins to see if it is working. I did not call this customer
>to set up
>his internet account yet. I may have to wait till tomorrow.
>
>Anybody have any good sugestions, please write me, I will appreciate it
>very much.
>
>Please write all the detailed instructions, since I am not that good..
>
>
>Ken. 905-826-0737 ken@gncom.com
>
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) new user log in problem with v.90 From: Gorkem Yuksel <gorkem@gncom.com> Date: 1998-11-24 16:45:21
Hi!
Everybody.
I am a new guy on the block.(small ISP)
I am having some problems with new user's logging in.
It does not allow to logging in right away when it comes to new users.
Sometimes it started working 7 hours later, 1.5 hours later depending on
the Hiper DSP's mood, I guess.
But all the old users connecting OK. All the new users you key in today
will be ok by tommorrow.
I have TCM, Hiper DSP cards, NMC card, Hiper ARC card, about 6 month old.
We upgraded to v.90 a month ago, it worked fine I guess.
(or I did not have customers who wanted an instant setup???)
I have system release 3.3.
Hiper ARC 1.1.11
Hiper DSP T1/PRI 1.2.5
NMC 16Meg 486 5.5.5
TCM Windows 5.5.1
I e-mailed for help to 3Com but no answer for 3 days...
I do not have tech support contract...too much money..
It started happening last Sat and Sun...Mon and Tue.. which is today.
I had one new customer yesterday.. it did not rscognize his username and
password right away, but started working about one and half hour later.
I have another new one keyed in 14:33 pm today, it is not working yet
(16:37).
I just keep typing in his user name and password by using dial up every 15
to 30 mins to see if it is working. I did not call this customer to set up
his internet account yet. I may have to wait till tomorrow.
Anybody have any good sugestions, please write me, I will appreciate it
very much.
Please write all the detailed instructions, since I am not that good..
Ken. 905-826-0737 ken@gncom.com
Subject:RE: (usr-tc) multiple default routes From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-24 16:47:58
You may be able to add additional default routes by specifying a metric of 2
for the second and 3 for the 3rd. I set this up on one of ours. Although I
could not verify the setting took place I was able to see one of my backup
default routes take over when I deleted the original one. Try it Maybe you
could even set up a second eth and make the default route entry a metric of
2 for it and unplug eth 1 to see if the default route for eth 2 shows up and
takes over.
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian
>Sent: Tuesday, November 24, 1998 3:27 PM
>To: USRobotics TC Mailing List
>Subject: (usr-tc) multiple default routes
>
>
>Is their a way on the ARC to declare multiple default routes for
>redundancy? something like, so that it uses the first default route, but
>if that fails it tries the second?
>
>Brian
>
>
>--------------------------------------------------------------------------
>Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
>Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
>signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
>(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) new user log in problem with v.90 From: K Mitchell <mitch@keyconn.net> Date: 1998-11-24 16:52:25
At 04:45 PM 11/24/98 -0500, Gorkem Yuksel <gorkem@gncom.com> wrote:
>I am having some problems with new user's logging in.
>
>It does not allow to logging in right away when it comes to new users.
>
>Sometimes it started working 7 hours later, 1.5 hours later depending on
>the Hiper DSP's mood, I guess.
>
>But all the old users connecting OK. All the new users you key in today
>will be ok by tommorrow.
Without more detailed information, it's hard to diagnose what might be
going on.
>I have system release 3.3.
>Hiper ARC 1.1.11
>Hiper DSP T1/PRI 1.2.5
>NMC 16Meg 486 5.5.5
>TCM Windows 5.5.1
Current v90 ARC code is 4.1.72, I believe 4.0.30 is the current x2 ARC code
>It started happening last Sat and Sun...Mon and Tue.. which is today.
>I had one new customer yesterday.. it did not rscognize his username and
>password right away, but started working about one and half hour later.
>
>I have another new one keyed in 14:33 pm today, it is not working yet
>(16:37).
>I just keep typing in his user name and password by using dial up every 15
>to 30 mins to see if it is working. I did not call this customer to set up
>his internet account yet. I may have to wait till tomorrow.
What are you using for radius(authentication) and how are you entering the
users?
Kirk Mitchell-General Manager sysadmin@keyconn.net
Keystone Connect http://www.keyconn.net
Altoona, PA 814-941-5000 We Unlock the World
Subject:(usr-tc) Seeking a way to 'hide' users from "list connections" From: T.J. Weber <lists@ipmedia.net> Date: 1998-11-24 17:02:13
Okee, last night I posted this to one of 3com's internal newsgroups:
---- begin ----
This must seem a little odd, but I would like to know if there is a way
I can hide users from showing up in the "list connections". I use
RADIUS authentication, and I am wondering if there is a configuration
value I can set for the specific users in my RADIUS 'users' file that
will be sent back to my TC telling the HiPer not to list that user in
*ANY* records (specifically the "list connections" output).
Using RADIUS would be my best guess in this case, unless there is a way
I can specify users to "hide" set directly via the HiPer config prompt.
---- end ----
No responce from 3com quite yet on this matter, three cents to the dollar
odds that they won't respond with anything at all.
I setup a Total Control box for my client in Chicago. They are a fairly
big law firm that is providing internet access to their employees. I am
not the only one with access to the total control box, there is someone
from their IS department that monitors connections on the box itself.
However, some of the partners in the firm don't want to be known that they
are online at a specific time (strange request if you ask me, i think they
don't want the guys in the IS dept. to know they are online).
I also have another question regarding the TC boxes. In my configuration
for the total control box, i set aside a pool of 24 IP addresses for
dynamic assignment (we only have one operational DSP card). However,
there are about 10 other users that I would like to assign a DYNAMIC IP
from another pool (NOT the pool assigned to the total control). Once
again, I think that this is something I have to setup in RADIUS, but I'm
not sure how to do it. Please advise!
So if anyone knows a simple solution to these problems that would be
great!
Thanks in advance,
--t.j.
\|||/
(o o)
ooO-(_)-Ooo----------------------------------------------------------
T.J. Weber | Oxymoron: Microsoft Works
Interplanetary Media |------------------------------------
phone: 847.205.5200 | WARNING: objects in mirror are
fax: 847.205.5201 | dumber than they appear.
e-mail: admin@ipmedia.net |------------------------------------
web: http://www.ipmedia.net | Out of my mind. Back in 5 minutes.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Two things are infinite: the universe and human stupidity;
and I'm not so sure about the universe. (Albert Einstein)
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Subject:(usr-tc) default routes From: Brian <signal@shreve.net> Date: 1998-11-24 17:16:49
How do you view the default routes on an arc that are set?
Brian
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
list rtab pre
will show you the route table and the routes that is stored in flash
krish
\ T.S.V. Krishnan \
\ Network System Engineer \ ( : - : )
\ 3Com ............ \
----------------------------------------------/
tkrishna@bubba.ae.usr.com
----------------------------/ http://interproc.ae.usr.com ----/
The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
Any Sufficiently advanced bug is indistinguishable for a feature.
- Rick Kulawiec
On Tue, 24 Nov 1998, Brian wrote:
>
> How do you view the default routes on an arc that are set?
>
> Brian
>
>
> --------------------------------------------------------------------------
> Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) default routes From: Brian <signal@shreve.net> Date: 1998-11-24 18:33:05
So does anyone know if this will work?
add ip defaultroute gateway 208.206.76.1 metric 1
add ip defaultroute gateway 208.206.76.2 metric 2
I want it so that if 208.206.76.1 is down, or fails, it will use the
second route. Will that work?
Brian
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:Re: (usr-tc) default routes From: Brian <signal@shreve.net> Date: 1998-11-24 18:59:36
It doesnt show you all of them, only the one it is using. Watch, I start
off with 208.206.76.1 as my default route. Then add 208.206.76.44, which
then becomes the preffered route, then I drop 208.206.76.44 and it goes
back to 208.206.76.1. Where in the arc can I get a list of all the
currently added defaultroutes?
Brian
HiPer>> list rtab pre
ROUTING TABLE PREFERRED ROUTES
Destination Prot Age NextHop Metric Interface
0.0.0.0/0 REMOTE 25 208.206.76.1 1 eth:1
.
.
.
HiPer>> add ip deFAULTROUTE gatEWAY 208.206.76.44 metric 1
HiPer>> list rtab pre
ROUTING TABLE PREFERRED ROUTES
Destination Prot Age NextHop Metric Interface
0.0.0.0/0 REMOTE 25 208.206.76.44 1 eth:1
.
.
.
HiPer>> del ip deFAULTROUTE gatEWAY 208.206.76.44
HiPer>> list rtab pre
ROUTING TABLE PREFERRED ROUTES
Destination Prot Age NextHop Metric Interface
0.0.0.0/0 REMOTE 25 208.206.76.1 1 eth:1
.
.
.
On Tue, 24 Nov 1998, Tatai SV Krishnan wrote:
> list rtab pre
>
> will show you the route table and the routes that is stored in flash
>
> krish
>
> -----------------------------------------
> \ T.S.V. Krishnan \
> \ Network System Engineer \ ( : - : )
> \ 3Com ............ \
> ----------------------------------------------/
> tkrishna@bubba.ae.usr.com
> ----------------------------/ http://interproc.ae.usr.com ----/
> The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
> -------------------------------------------------------------------------\
> Any Sufficiently advanced bug is indistinguishable for a feature.
> - Rick Kulawiec
> -------------------------------------------------------------------------/
>
> On Tue, 24 Nov 1998, Brian wrote:
>
> >
> > How do you view the default routes on an arc that are set?
> >
> > Brian
> >
> >
> > --------------------------------------------------------------------------
> > Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> > Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> > signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> > (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
'list ip defaultroute' should do the trick.
At 05:16 PM 11/24/98 -0600, you wrote:
>
>How do you view the default routes on an arc that are set?
>
>Brian
>
>
>--------------------------------------------------------------------------
>Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
>Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
>signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
>(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
John Rockwell
e-mail: jrockwel@clarityconnect.com
Network Engineer
Clarityconnect, Inc.
Ithaca Area: (607)257-8268
Outside Ithaca Area: (888)322-4900
Try us: http://www.clarityconnect.com
Subject:Re: (usr-tc) default routes From: Brian <signal@shreve.net> Date: 1998-11-24 20:42:57
how do you redistribute the default route? I mean the default route of my
border router is different than what I would want the arc to
hold.........I mean where does the redistributing start with?
On my router I do:
router rip
version 2
timers basic 30 30 2 60 300
passive-interface Ethernet0/0
passive-interface Serial0/0.1
passive-interface Ethernet1/0
passive-interface Serial1/0.1
passive-interface FastEthernet3/0
network 208.206.76.0
no auto-summary
how would I tell it to distribute 208.206.76.1 as the default gw for my
arcs?
> Depends...how is 208.206.76.1 connected? If its connected via ethernet,
> how does the ARC *know* that it's down? Periodic ICMP echo packets to
> it? Otherwise I don't know of any way of detecting that gateway is down
> on an ethernet network (if its off a serial port, that's obviously
> easily determined). Cisco has this same setup...if you're default
> gateway is across an ethernet, the cisco has no way of knowing if the
> gateway is down to switch to a floating default. The "solution" here is
> to distribute your default route in your routing protocol. I know
> that's not the *best* solution, but its about the only way to handle
> this on an ethernet network that I'm aware of. If you distribute
> default in your routing protocol, that's also one less config option
> that you have to set on your new boxes when you set them up. :)
> --
> Jeff McAdams Email: jeffm@iglou.com
> Head Network Administrator Voice: (502) 966-3848
> IgLou Internet Services (800) 436-4456
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Thus spake Brian
>So does anyone know if this will work?
>add ip defaultroute gateway 208.206.76.1 metric 1
>add ip defaultroute gateway 208.206.76.2 metric 2
>I want it so that if 208.206.76.1 is down, or fails, it will use the
>second route. Will that work?
Depends...how is 208.206.76.1 connected? If its connected via ethernet,
how does the ARC *know* that it's down? Periodic ICMP echo packets to
it? Otherwise I don't know of any way of detecting that gateway is down
on an ethernet network (if its off a serial port, that's obviously
easily determined). Cisco has this same setup...if you're default
gateway is across an ethernet, the cisco has no way of knowing if the
gateway is down to switch to a floating default. The "solution" here is
to distribute your default route in your routing protocol. I know
that's not the *best* solution, but its about the only way to handle
this on an ethernet network that I'm aware of. If you distribute
default in your routing protocol, that's also one less config option
that you have to set on your new boxes when you set them up. :)
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:(usr-tc) For Sale New NMC Card From: Brian Gordon <brian@westelcom.com> Date: 1998-11-24 22:44:59
One NMC full memory configuration (came with a trade up bundle)
Still in sealed static bag -
999.99 or best offer.
Please email Brian -
flash@highlogic.com
Subject:Re: (usr-tc) Seeking a way to 'hide' users from "list connections" From: Ricky Beam <jfbeam@interpath.net> Date: 1998-11-25 07:24:51
Tatai SV Krishnan was heard to say:
>> This must seem a little odd, but I would like to know if there is a way
>> I can hide users from showing up in the "list connections". I use
>> RADIUS authentication, and I am wondering if there is a configuration
>> value I can set for the specific users in my RADIUS 'users' file that
>> will be sent back to my TC telling the HiPer not to list that user in
>> *ANY* records (specifically the "list connections" output).
...
>If you want however you can set ppp passthrough meaning you can disable
>authentication for users and use the default user for all calls for say
>ppp. This way the only user who will be seen in the hIper arc is always
>the default user
Hmm, this makes me wonder what happens if the RADIUS server rewrites the
username in the access-accept.
--Ricky
Thus spake Brian
>how do you redistribute the default route? I mean the default route of my
>border router is different than what I would want the arc to
>hold.........I mean where does the redistributing start with?
Check out "default-information originate" or something along those
lines, I think that will tell your routing protocol (RIP in this case)
to announce the 0.0.0.0 route, which it typically doesn't do. You can
check on CCO for the specifics of this announcement. You can do all
sorts of cool stuff with it, like only announce it if a route-map is
satisfied and stuff...I use a variation of this a bit within my OSPF
domain to provide redundancy for Internet connectivity. T3 comes into
one router, our T1's into another, one of the routers on that network
isn't beefy enough to handle full BGP routes, so I gave it a static
default route with a high administrative distance (above the admin
distance of OSPF), then my cisco which has the T1's has
default-information originate with a route-map on it in OSPF. The
route-map specifies that if it *doesn't* see the /30 for the IP
addresses on the T3 that it *should* advertise the default out in OSPF.
This means that the wimpy router that can't do OSPF will switch over its
default to the T1 router if the T3 or the T3 router dies.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
|o| how do you redistribute the default route? I mean the default route of my
|o| border router is different than what I would want the arc to
|o| hold.........I mean where does the redistributing start with?
default-information originate in the router process with a route
map attached should do it
take care
|o|----------------------------------------------------------------------|o|
|o| bryan s. blank (203)-351-1178 voice |o|
|o| senior systems analyst (203)-351-1186 fax |o|
|o| discovernet, incorporated (203)-979-5126 emerg |o|
Subject:Re: (usr-tc) Seeking a way to 'hide' users from "list connections" From: Tatai SV Krishnan <tkrishna@bubba.ae.usr.com> Date: 1998-11-25 09:24:52
On Wed, 25 Nov 1998, Ricky Beam wrote:
> Tatai SV Krishnan was heard to say:
> >> This must seem a little odd, but I would like to know if there is a way
> >> I can hide users from showing up in the "list connections". I use
> >> RADIUS authentication, and I am wondering if there is a configuration
> >> value I can set for the specific users in my RADIUS 'users' file that
> >> will be sent back to my TC telling the HiPer not to list that user in
> >> *ANY* records (specifically the "list connections" output).
> ...
> >If you want however you can set ppp passthrough meaning you can disable
> >authentication for users and use the default user for all calls for say
> >ppp. This way the only user who will be seen in the hIper arc is always
> >the default user
>
> Hmm, this makes me wonder what happens if the RADIUS server rewrites the
> username in the access-accept.
>
Radius does not get a access-request here - Just the accounting packet is
sent.
krish
> --Ricky
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
_list rtab en
krish
\ T.S.V. Krishnan \
\ Network System Engineer \ ( : - : )
\ 3Com ............ \
----------------------------------------------/
tkrishna@bubba.ae.usr.com
----------------------------/ http://interproc.ae.usr.com ----/
The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
Any Sufficiently advanced bug is indistinguishable for a feature.
- Rick Kulawiec
On Wed, 25 Nov 1998, Brian K McIntire wrote:
> >-----Original Message-----
> >From: owner-usr-tc@lists.xmission.com
> >[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Tatai SV Krishnan
> >Sent: Tuesday, November 24, 1998 5:32 PM
> >To: Brian
> >Cc: USRobotics TC Mailing List
> >Subject: Re: (usr-tc) default routes
> >
> >
> >list rtab pre
> >
> >will show you the route table and the routes that is stored in flash
>
> I tried this. list rtab preferred did not show me the second default route
> i added. However, if i delete the primary default route the one I had
> enetered before with a metric of 2 showed up under list rtab preferred and
> list ip routes.
> >
> >krish
> >
> >-----------------------------------------
> > \ T.S.V. Krishnan \
> > \ Network System Engineer \ ( : - : )
> > \ 3Com ............ \
> > ----------------------------------------------/
> >tkrishna@bubba.ae.usr.com
> >----------------------------/ http://interproc.ae.usr.com ----/
> >The Yadda Yadda Search - for simple anwers -
> http://interproc.ae.usr.com/tkb.html
> -------------------------------------------------------------------------\
> Any Sufficiently advanced bug is indistinguishable for a feature.
> - Rick Kulawiec
> -------------------------------------------------------------------------/
>
> On Tue, 24 Nov 1998, Brian wrote:
>
> >
> > How do you view the default routes on an arc that are set?
> >
> > Brian
> >
> >
> > --------------------------------------------------------------------------
> > Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> > Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> > signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> > (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) default routes From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-25 09:37:54
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Tatai SV Krishnan
>Sent: Tuesday, November 24, 1998 5:32 PM
>To: Brian
>Cc: USRobotics TC Mailing List
>Subject: Re: (usr-tc) default routes
>
>
>list rtab pre
>
>will show you the route table and the routes that is stored in flash
I tried this. list rtab preferred did not show me the second default route
i added. However, if i delete the primary default route the one I had
enetered before with a metric of 2 showed up under list rtab preferred and
list ip routes.
>
>krish
>
>-----------------------------------------
> \ T.S.V. Krishnan \
> \ Network System Engineer \ ( : - : )
> \ 3Com ............ \
> ----------------------------------------------/
>tkrishna@bubba.ae.usr.com
>----------------------------/ http://interproc.ae.usr.com ----/
>The Yadda Yadda Search - for simple anwers -
http://interproc.ae.usr.com/tkb.html
Any Sufficiently advanced bug is indistinguishable for a feature.
- Rick Kulawiec
On Tue, 24 Nov 1998, Brian wrote:
>
> How do you view the default routes on an arc that are set?
>
> Brian
>
>
> --------------------------------------------------------------------------
> Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:RE: (usr-tc) default routes From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-25 09:40:29
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian
>Sent: Tuesday, November 24, 1998 6:33 PM
>To: USRobotics TC Mailing List
>Subject: (usr-tc) default routes
>
>
>
>So does anyone know if this will work?
>
>add ip defaultroute gateway 208.206.76.1 metric 1
>add ip defaultroute gateway 208.206.76.2 metric 2
>
>I want it so that if 208.206.76.1 is down, or fails, it will use the
>second route. Will that work?
I tried this here. I did get it to work, but there seemed to be a little
lag on how fast the second route took over when I made the first one
un-available.
Seemed to be roughly 60 seconds.
>
>Brian
>
>
>--------------------------------------------------------------------------
>Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
>Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
>signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
>(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) default routes From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-25 09:40:30
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian
>Sent: Tuesday, November 24, 1998 7:00 PM
>To: Tatai SV Krishnan
>Cc: USRobotics TC Mailing List
>Subject: Re: (usr-tc) default routes
>
>
>
>It doesnt show you all of them, only the one it is using. Watch, I start
>off with 208.206.76.1 as my default route. Then add 208.206.76.44, which
>then becomes the preffered route, then I drop 208.206.76.44 and it goes
>back to 208.206.76.1.
Did you specify a metric of 2 for the second entry? When i did it did not
over ride the primary until the primary went down
Where in the arc can I get a list of all the
>currently added defaultroutes?
>
>Brian
>
>
>HiPer>> list rtab pre
>ROUTING TABLE PREFERRED ROUTES
>Destination Prot Age NextHop Metric Interface
> 0.0.0.0/0 REMOTE 25 208.206.76.1 1 eth:1
>.
>.
>.
>
>HiPer>> add ip deFAULTROUTE gatEWAY 208.206.76.44 metric 1
>HiPer>> list rtab pre
>ROUTING TABLE PREFERRED ROUTES
>Destination Prot Age NextHop Metric Interface
> 0.0.0.0/0 REMOTE 25 208.206.76.44 1 eth:1
>.
>.
>.
>HiPer>> del ip deFAULTROUTE gatEWAY 208.206.76.44
>HiPer>> list rtab pre
>ROUTING TABLE PREFERRED ROUTES
>Destination Prot Age NextHop Metric Interface
> 0.0.0.0/0 REMOTE 25 208.206.76.1 1 eth:1
>.
>.
>.
>
>
>
>
>
>
>On Tue, 24 Nov 1998, Tatai SV Krishnan wrote:
>
>> list rtab pre
>>
>> will show you the route table and the routes that is stored in flash
>>
>> krish
>>
>> -----------------------------------------
>> \ T.S.V. Krishnan \
>> \ Network System Engineer \ ( : - : )
>> \ 3Com ............ \
>> ----------------------------------------------/
>> tkrishna@bubba.ae.usr.com
>> ----------------------------/ http://interproc.ae.usr.com ----/
>> The Yadda Yadda Search - for simple anwers -
http://interproc.ae.usr.com/tkb.html
> -------------------------------------------------------------------------\
> Any Sufficiently advanced bug is indistinguishable for a feature.
> - Rick Kulawiec
> -------------------------------------------------------------------------/
>
> On Tue, 24 Nov 1998, Brian wrote:
>
> >
> > How do you view the default routes on an arc that are set?
> >
> > Brian
> >
> >
>
> --------------------------------------------------------------------------
> > Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service
Provider
> > Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> > signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> > (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
|-----Original Message-----
|From: owner-usr-tc@lists.xmission.com
|[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian K McIntire
|Sent: Tuesday, November 24, 1998 4:48 PM
|To: usr-tc@lists.xmission.com
|Subject: RE: (usr-tc) multiple default routes
|
|
|You may be able to add additional default routes by specifying a metric of 2
|for the second and 3 for the 3rd. I set this up on one of ours. Although I
|could not verify the setting took place I was able to see one of my backup
|default routes take over when I deleted the original one. Try it Maybe you
|could even set up a second eth and make the default route entry a metric of
|2 for it and unplug eth 1 to see if the default route for eth 2 shows up and
|takes over.
|
This is the correct way to do it.. You can see the routes by typing "_li rt en"
It will not show up in your routing table untill your default route becomes
unreachable. A request has already been make to make this show in the table
without a special command.
-M
Subject:Re: (usr-tc) default routes From: Peter D. Mayer <dmayer@netwalk.com> Date: 1998-11-25 10:39:53
I think "list ip routes" will list all the routes regardless of preference.
Peter D. Mayer
NetWalk System Administrator
dmayer@netwalk.com
-----Original Message-----
Cc: usr-tc@lists.xmission.com <usr-tc@lists.xmission.com>
_list rtab en
krish
\ T.S.V. Krishnan \
\ Network System Engineer \ ( : - : )
\ 3Com ............ \
tkrishna@bubba.ae.usr.com
----------------------------/ http://interproc.ae.usr.com ----/
The Yadda Yadda Search - for simple anwers -
http://interproc.ae.usr.com/tkb.html
Any Sufficiently advanced bug is indistinguishable for a feature.
- Rick Kulawiec
On Wed, 25 Nov 1998, Brian K McIntire wrote:
> >-----Original Message-----
> >From: owner-usr-tc@lists.xmission.com
> >[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Tatai SV Krishnan
> >Sent: Tuesday, November 24, 1998 5:32 PM
> >To: Brian
> >Cc: USRobotics TC Mailing List
> >Subject: Re: (usr-tc) default routes
> >
> >
> >list rtab pre
> >
> >will show you the route table and the routes that is stored in flash
>
> I tried this. list rtab preferred did not show me the second default
route
> i added. However, if i delete the primary default route the one I had
> enetered before with a metric of 2 showed up under list rtab preferred and
> list ip routes.
> >
> >krish
Subject:RE: (usr-tc) default routes From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-25 10:46:12
No, it won't
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Peter D. Mayer
>Sent: Wednesday, November 25, 1998 9:40 AM
>To: usr-tc@lists.xmission.com
>Subject: Re: (usr-tc) default routes
>
>
>I think "list ip routes" will list all the routes regardless of preference.
>
>Peter D. Mayer
>NetWalk System Administrator
>dmayer@netwalk.com
>
>-----Original Message-----
>From: Tatai SV Krishnan <tkrishna@bubba.ae.usr.com>
>To: Brian K McIntire <bmcintire@commnet.com>
>Cc: usr-tc@lists.xmission.com <usr-tc@lists.xmission.com>
>Date: Wednesday, November 25, 1998 10:20 AM
>Subject: RE: (usr-tc) default routes
>
>
>_list rtab en
>
>krish
>
>-----------------------------------------
>\ T.S.V. Krishnan \
>\ Network System Engineer \ ( : - : )
> \ 3Com ............ \
>----------------------------------------------/
>tkrishna@bubba.ae.usr.com
>----------------------------/ http://interproc.ae.usr.com ----/
>The Yadda Yadda Search - for simple anwers -
>http://interproc.ae.usr.com/tkb.html
>-------------------------------------------------------------------------\
>Any Sufficiently advanced bug is indistinguishable for a feature.
>- Rick Kulawiec
>-------------------------------------------------------------------------/
>
>On Wed, 25 Nov 1998, Brian K McIntire wrote:
>
>> >-----Original Message-----
>> >From: owner-usr-tc@lists.xmission.com
>> >[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Tatai SV Krishnan
>> >Sent: Tuesday, November 24, 1998 5:32 PM
>> >To: Brian
>> >Cc: USRobotics TC Mailing List
>> >Subject: Re: (usr-tc) default routes
>> >
>> >
>> >list rtab pre
>> >
>> >will show you the route table and the routes that is stored in flash
>>
>> I tried this. list rtab preferred did not show me the second default
>route
>> i added. However, if i delete the primary default route the one I had
>> enetered before with a metric of 2 showed up under list rtab
>preferred and
>> list ip routes.
>> >
>> >krish
>
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) Global Villiage Modems? From: Brian Gordon <administrator@westelcom.com> Date: 1998-11-25 10:52:25
Whats the Deal with Global Villiage modems and the Hiper DSP, we are running
the latest code and it seems like people are getting really crappy
connections speeds etc.
Help Needed here.
Brian Gordon, MCP
Network Administrator
Westelcom Internet / Westelcom Computer
518-566-8376 Voice
518-566-8348 Fax
administrator@westelcom.com
http://home.westelcom.com
Subject:(usr-tc) Telepath modem running 1.41.017 From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-25 10:56:16
Does anyone know of issues with this particular modem. I'm working with a
customer running 1.2.5 on his DSP's and 4.1.72 - 7 on his HiPer ARC. With
the same user account I can connect 20 times out of 20 with my courier. The
client that has a Telepath connects 1 out of every 3-7 times.
Subject:Re: (usr-tc) connecting a cisco router to total control ISDN From: Tatai SV Krishnan <tkrishna@bubba.ae.usr.com> Date: 1998-11-25 11:07:58
> Hi All
>
> I have a strange one
>
> Cisco 1603 router trying to connect to a totalcontrol chassis
>
> we can get the cisco to connect and authenticate using pap
> but we never see any accounting data in the radius logs so the cisco just
> hangs up
Why would the cisco care about accounting data in radius?
>
> is there anything strange about a cisco connecting to usr kit?
>
Check yadda yadda http://interproc.ae.usr.com/tkb.html
search for lan-to-lan
you will the config for hiper/netserver/cisco
krish
> there are 2 netserver based and 1 hyper based racks here and all do
> the same thing.
>
> TIA
>
> Steve Lalonde
> Systems Manager
> ENTANET International Ltd
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) A rather nasty problem with NMCs From: Gilles Melanson <gilles@vianet.on.ca> Date: 1998-11-25 12:04:20
Here's a fun one:
NMC code 5.5.5:
I started adding authorized hosts (via TCM) to one of my chassis.
I came to add two IPs:
192.168.1.1 255.255.255.255 description1
192.168.1.2 255.255.255.255 description2
as soon as I put in the second one, boom, I lost the NMC.
'lo and behold, the two community strings on the card reset themselves to
public/private. Wtf?
It happened on two different chassis' (hiper boxes), where I put in two
consecutive IPs from the same /24 (as above).
If put IPs in a different order (Ie: mix what /24 each host is in), it
works fine.
This behavior is still rather alarming. If someone @ 3Com can test this
out a little more thoroughly..
It's probably not critical, though. You can't set authorized hosts w/o
having the original/good read/write strings to begin with, but the fact
that they were reset to public/private w/o actually being set manually..
*shrug*
--
Gilles Melanson ViaNet Internet Solutions
System Administrator 128 Larch St. Suite 301
gilles@vianet.on.ca Sudbury, ON Canada P3E 5J8
There isn't any reason why Linux can't be implemented as an enterprise
computing solution. Find out what you've been missing while you've
been rebooting Windows NT. -- Eric Hammond
Subject:RE: (usr-tc) #2 new user log in problem with v.90 From: Gorkem Yuksel <gorkem@gncom.com> Date: 1998-11-25 12:58:57
Thank you for responding my help call.
It happened several times last few days, but for example..
I keyed in a new user( ie. kim225) yesterday 14:33pm.
Before I call my customer and start setting up, I tried to log in using
this new customers user name(normally I don't do this), but rejected.
The reason is user name and password is not valid.
I am using my other computer(windows 95) in my office to dial. So it
dials and makes handshake and start verifying user name and pass word.
And it gives me an error messge window and ask me enter password again
and again..... and disconnects.
When I try to make connection with any other old usernames, it connects
right away.
So I go to my UNIX(BSDI 2.0) machine, open etc/raddb directory and check
users file, surely new user kim225 is there on top of other files.
I go to etc , check password file and see kim225 also.
When I had this same problems about a month ago, about 2 weeks after
installing new v.90 code, I called and begged to connect to 3Com's tech
support(since I do not have service contract). After several tries, they
helped me out. But the guy connected to our server and corrected and did
not tell me exactly how it wsa done. I asked for manuals and he tried to
send by e-mail but it was too big, he had problems sending it, so I never
got the manuals for Hiper Chassis. I can open up hiper chassis and type in
"list connection" and it gives me who is connected.... that is all I
know...
He said two things, not 100% sure though...
1) reboot the whole chassis ... from Total Control manager
(I can open this total control manager and see their connection
speed and things, and tha is about it)
Can you tell me how to reboot the whole chassis....(it will
disconnect everybody, that is how much I know)
And it did not solve the problem, he said he is going to reset the 2nd
card(I have only two DSP cards) seperately. and it worked. But he did not
tell me how.
I guess I should reboot from Hiper>> prompt, so I tried
Hyper>> reset interface slot:2/mod:[1-23]
but it gave me error message.... should I type in modems? or
authentication instead fo interface?
I key in users in BATS accounting system in UNIX.
How do I reset/reload radius daemon in UNIX machine? or is it in TCM?
Thanks in advance.
Ken.
Just thought I would add that I experienced the same problem with this
version of code. Using mon ppp it seemed like it was experiencing the LCP
problem that web tv and imac's had. It did not get up to the
authentication stage. I'm still running 4.0.30 on our production chassis'
this being one of the reasons...
Jerry
At 10:56 AM 11/25/98 -0600, you wrote:
>Does anyone know of issues with this particular modem. I'm working with a
>customer running 1.2.5 on his DSP's and 4.1.72 - 7 on his HiPer ARC. With
>the same user account I can connect 20 times out of 20 with my courier. The
>client that has a Telepath connects 1 out of every 3-7 times.
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) Telepath modem running 1.41.017 From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-25 13:30:53
>Just thought I would add that I experienced the same problem with this
>version of code. Using mon ppp it seemed like it was experiencing the LCP
>problem that web TV and imac's had. It did not get up to the
>authentication stage. I'm still running 4.0.30 on our production chassis'
>this being one of the reasons...
>
Thanks, that does help. We are seeing exactly the same thing on ours. The
user attempts to connect, handshakes and goes no further. Monitor radius
shows nothing. It just seems to be hanging. I guess I'll call 3COM and
open a tkt with them. Thought I might have some bad code on that modem but
if you are seeing it too then maybe it's a bug with something else.
Thanks again
>Jerry
>
>At 10:56 AM 11/25/98 -0600, you wrote:
>>Does anyone know of issues with this particular modem. I'm working with a
>>customer running 1.2.5 on his DSP's and 4.1.72 - 7 on his HiPer ARC. With
>>the same user account I can connect 20 times out of 20 with my
>courier. The
>>client that has a Telepath connects 1 out of every 3-7 times.
Subject:(usr-tc) Edgeserver call-back only at 28.8 ?? From: Robert von Bismarck <rvb@petrel.ch> Date: 1998-11-25 13:31:34
Hi,
We have deployed a TC at a customers site which he will use with
call-back feature. This rack is laid out the following way :
USR-TC 70amp chassis w/ fantray
1 DUAL E1-PRI card
4 Quad modem single-sided (software version 5.10.9)
1 Edgeserver (486 based) version 1.5.0
1 NMC (software version 5.5.5)
We use win NT RAS for dial-up (due to SecurID) and would like to be able
top set up a callback system.
This works fine as long as the call is V.90. With ISDN, the call-in is
ok, but the call-back is V.34, which is not good at all...
Is there something to configure in the edgeserver so that the call out
will be made with the same protocol (ISDN or analog) as the call in ?
RADIUS is NOT used, we log into an NT domain through the edgeserver,
using NT domain accounts, and using RADIUS with this configuration is
not an option, as we do not have either Netserver or HiperARC.
One option available is that all calls out be ISDN, I know I could set
that up by selecting stuff in the quad modem config, bu I'd rather not
experiment with this, has anyone ever done this, or has a configuration
template for the quads ?
I'm sorry if this FAQ stuff, but 'till now there is no FAQ ;-) and my
experience is mostly with HiPer chassis...
Thanks for any help / pointers,
Robert
Subject:Re: (usr-tc) A rather nasty problem with NMCs From: Gilles Melanson <gilles@vianet.on.ca> Date: 1998-11-25 13:46:18
Ignore it.
I'm officially a dumbass.
I locked myself out of my own NMCs. *laf*
It's unfortunate that you have to add each host one at a time, instead of
an entire list of IPs and THEN setting the whole lot of them. My brain is
in Cisco-mode this morning. =o
--
Gilles Melanson ViaNet Internet Solutions
System Administrator 128 Larch St. Suite 301
gilles@vianet.on.ca Sudbury, ON Canada P3E 5J8
There isn't any reason why Linux can't be implemented as an enterprise
computing solution. Find out what you've been missing while you've
been rebooting Windows NT. -- Eric Hammond
Subject:(usr-tc) FW: (CSO TotalService) RE: (CSO TotalService) Dial out test from HiPer DSP (T1 channel) From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-25 13:52:13
-----Original Message-----
[mailto:"Nair, Shibu (MED, Wipro Systems, Inc)"@totalservice.usr.com]
Sent: Wednesday, November 25, 1998 10:53 AM
HiPer DSP (T1 channel)
Hi
Since i have not get any reply iam sending this again with more
details...
I have tried dialing from the mdm prompt of Hiper DSP card.
As soon as i tried i checked the at status on the tslot device on HiPER
DSP.
Initially it gave dialout and then it connection out which is shown
below,
(This is the case of when i dial out to a normal telephone number)
span1/tslot1> displ atst
Tslot Status Modem Status Call ID Action Q931
Connect Srvc State Queued Ref
01 Conn Out 001 IS 0x00040001 None
0x00000000
Still when i try dialing to a unix host iam not getting the unix host
prompt.
and mdm> prompt is returning back...
Hope somebody can help me on this,
My span card configuration is shown below, correct me if iam wrong....
span1> displ ccrcfig
Span1 Configured Signal Mode is (sigmode): ROBBED BIT
Span1 Signal Mode Active is: ROBBED BIT
Span1 DNIS Enable is (dnisena): DNIS ADDRESS
Span1 Dial In Out Trunk Start (diotrst): WINK
Span1 Dial In Address ACK Wink (daackwnk): ACK WINK DISABLED
Span1 Dial Out Address Delay (doadrdly): 70 milliseconds
Span1 Dial In Out Trunk Type (dtrnktyp): E&M TYPE II
Span1 Configured Switch Type is (swtype): 5ESS
Span1 Switch Type Active is: 5ESS
Span1 Idle Byte is (idlebyte): 0xFE
Span1 Ana Calls Blocked Err Code (ancbec): 58
Span1 Digi Calls Blocked Err Code (dcbec): 58
Span1 No IGWS Avail Err Code (noigwsav): 58
Span1 Chan Blocked Err Code (chanblck): 58
Span1 Block Call Type is (blcaltyp): BLOCK NONE
Span1 Tone Type is (tonetype): DTMF TONE
Span1 Number Of DTMF Tones is (numdtmft): 5
Span1 Dial Out Select Direction (dseldir): DOWN
Span1 Dial Out Next Timeslot (dntslot): 24
Span1 Channelized T1 Profile (cprofile): E&M TYPE II GENERIC PROFILE
Regards
Shibu
----------
From: "Nair, Shibu (MED, Wipro Systems,
Inc)"@totalservice.usr.com[SMTP:"Nair, Shibu (MED, Wipro Systems,
Inc)"@totalservice.usr.com]
Reply To: user-forum-totalcontrol@totalservice.usr.com
Sent: Tuesday, November 24, 1998 2:01 PM
Subject: (CSO TOTALservice) Dial out test from HiPer DSP
(T1 channel) card
Hi
I was testing my T1 ckt using atdt command on HiPer DSP
card
console.
First i diald to my telephone number(to confirm) using
atdt
command
from T1 card prompt, I did the following for the same...
>chdev mdm
mdm1> atdt32853
My telephone ring for a while and disconnect...
Now i tried to dial to a portmaster(which is configured
for
dialup) from HiPer DSP card prompt. I have followed the above
steps with
appropriate dialup number. But i did not get the loging prompt
of the
machine....
Did i do anything wrong....?
How do i see the status of the dilaup channel..?
Pls help me on this ....
(Pls note that, i have tested using console which is
connected
to HiPer DSP card).
Regards
Shibu
The 4.1.72-7 HARC code fixes document mentions a couple of fixes wrt
MPIP. Can someone at USR/3Com check out trouble ticket 58316 and let me
know if these fixes address this problem? If these fixes address this
problem, then one of the fixes mentions the requirement of a new version
of code on the NETServer (which I understand is going to be required to
fix the issues in 58316), if this fix relates to this trouble ticket,
when will the corresponding NETServer code be available? I am assuming
it will be a ER/SR for 3.8.1....and I'm assuming that it will be later
than 3.8.90 (ie, 3.8.89 or lower number) as I have 3.8.90 and it didn't
fix the problem.
Thanks
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) Seeking a way to 'hide' users from "list connections" From: MegaZone <megazone@megazone.org> Date: 1998-11-25 14:13:26
Once upon a time David Bolen shaped the electrons to say...
>Hmm, since a Username attribute isn't even permitted in an
>Access-Accept (see RFC2138, although I think I saw some e-mail
>comments about changing that to permit it fly by recently), other than
Yes, I believe it will be permitted in the final version of the RFC. If
I'm recalling my ietf-radius threads properly.
-MZ
--
<URL:mailto:megazone@megazone.org> Gweep, Discordian, Author, Engineer, me..
Join ISP/C Internet Service Providers' Consortium <URL:http://www.ispc.org/>
"A little nonsense now and then, is relished by the wisest men" 781-788-0130
<URL:http://www.megazone.org/> <URL:http://www.gweep.net/> Hail Discordia!
Subject:(usr-tc) HiPer ARC 4.0.30 From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-25 14:28:23
I have a customer with 1.2.5 on his DSP's and 4.0.30 on his HiPer ARC. NMC
is running 5.5.5. It seems, each day, 7-8 modems change from up up to down
up. When I reset the DSP the problem goes away just to come back another
day. :( I don't remember if there was a bug relating to something like
this in 4.0.30. If there was, I can use that as a reason to make the
upgrade happen. I would appreciate any help on this. Thank you.
Subject:RE: (usr-tc) Scripting the config of a HARC From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-25 14:57:52
See chapter 10-150 of the 4.1 HiPer ARC user manual. There is a good sized
example in there
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Eric
>Sent: Friday, November 20, 1998 4:49 PM
>To: TC List
>Subject: (usr-tc) Scripting the config of a HARC
>
>
>How can you script the configuration of a HARC? I was told that you can
>TFTP the file onto the HARC and then type "do filename"
>
>I'd like to be able to easily batch setup a HARC. With easy prompts,
>like:
>
>-What is my IP address?
>-What is my netmask?
>-What is my gateway?
>etc.
>
>Can someone send me some instructions on how to do this and maybe a sample
>script?
>
>Eric
>---
>Eric J. Lorenzo
>Sr. Field Engineer
>lorenzo@ispchannel.com
>v:650.237.1465
>http://www.ISPchannel.com
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) rcvdGatewayDiscCmd From: Charles Sprickman <spork@inch.com> Date: 1998-11-25 14:59:06
Hi,
From what I understand "rcvdGatewayDiscCmd" means that the ARC/Netserver
has essentially dropped DTR on a modem to cause a hangup. What are the
various reasons a Netserver might do this besides idle-timeout and
session-timeout? I'm all of a sudden seeing lots of these on all my
chassis.
While it's implied that this is the Netserver dropping the call for a
reason, is it possible that a line drop could result in this cause code
(ie: Netserver realizes call is gone and terminates it)??
Thanks,
Charles
=-----------------= =
| Charles Sprickman Internet Channel |
| INCH System Administration Team (212)243-5200 |
| spork@inch.com access@inch.com |
= =----------------=
On Wed, 25 Nov 1998, Jeff Mcadams wrote:
> The 4.1.72-7 HARC code fixes document mentions a couple of fixes wrt
> MPIP. Can someone at USR/3Com check out trouble ticket 58316 and let me
> know if these fixes address this problem? If these fixes address this
> problem, then one of the fixes mentions the requirement of a new version
> of code on the NETServer (which I understand is going to be required to
> fix the issues in 58316), if this fix relates to this trouble ticket,
> when will the corresponding NETServer code be available? I am assuming
> it will be a ER/SR for 3.8.1....and I'm assuming that it will be later
> than 3.8.90 (ie, 3.8.89 or lower number) as I have 3.8.90 and it didn't
> fix the problem.
The problem that HiPer arc fixes with MPIP is
1. VTP problem
2. Deregistration problem
In the first case the VTP problem - the hiper arc would not send correct
data over the tunnel - meaning users after sometime would not be able to
make a MPIP multilink call
The second problem is is how to deregister a MPIP link. Some clients
would not send deregistration properlly - meaning the NeTServer code
changed its way of MPIP operation and that cause the HiPer arc not to be
able to deregister the link properlly.
The NeTServer code you have does have the fixes but there was some
additional fixes in a subsequent ER you need that code.
krish
>
> Thanks
> --
> Jeff McAdams Email: jeffm@iglou.com
> Head Network Administrator Voice: (502) 966-3848
> IgLou Internet Services (800) 436-4456
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) Seeking a way to 'hide' users from "list connections" From: David Bolen <db3l@ans.net> Date: 1998-11-25 15:26:25
Ricky Beam <jfbeam@Interpath.net> writes:
> Hmm, this makes me wonder what happens if the RADIUS server rewrites the
> username in the access-accept.
Hmm, since a Username attribute isn't even permitted in an
Access-Accept (see RFC2138, although I think I saw some e-mail
comments about changing that to permit it fly by recently), other than
ignoring it, I'm guessing that nothing would happen :-)
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
Subject:(usr-tc) Netserver reboot From: Martin Lathoud <nytral@endirect.qc.ca> Date: 1998-11-25 15:29:32
Hi,
I am new to this list and since web archives don't seem to be up, I will
ask here my question:
I am experiencing self-reboot from my Netserver card (3.8.1) used only for
dialin PPP, from once a day to once a week. Sometimes it hangs, but more
generaly I only see that lights turn red then blinking green..
It is running with 20MB. I would like to know if someone already met this
kind of problem and if I have a way to be sure that it's an hardware
failure, not sofware (support didn't find anything so they suggest
I replace the card... they are soo helpful.. not! ).
TIA,
Martin
Hi
Iam new to Total Control Enterprise Network Hub.
I have configured Channelised T1 ckt for the HiPer DSP card.
Since i do no have Total Control Manager software, iam testing using the
console connection.
I did the following test to ensure the dialout from the HiPerDSP ...
1) I have dialed to my local telephone and received the dial tone..
>mdm <telephone number>
2) I have dialed to a machine which has dialup setup
At this time it is trying for dialing out and comes back to the mdm
prompt.
Iam not getting machine access when i dial....
mdm2> atdt<machine dialup number>
mdm2>
HiPer DSP Rev. 1.0.7
Pls help on this problem....
Regards
Shibu
Subject:RE: (usr-tc) rcvdGatewayDiscCmd From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-25 15:45:21
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Charles Sprickman
>Sent: Wednesday, November 25, 1998 1:59 PM
>To: usr-tc@xmission.com
>Subject: (usr-tc) rcvdGatewayDiscCmd
>
>
>Hi,
>
>>From what I understand "rcvdGatewayDiscCmd" means that the ARC/Netserver
>has essentially dropped DTR on a modem to cause a hangup.
Not exactly. Remember the NMC is the one reporting to you a "received
gateway disconnect command" What this means is the HiPer ARC told the NMC
the user has disconnected before the NMC could get the info from somewhere
else.
>What are the
>various reasons a NetServer might do this besides idle-timeout and
>session-timeout? I'm all of a sudden seeing lots of these on all my
>chassis.
>
>While it's implied that this is the NetServer dropping the call for a
>reason, is it possible that a line drop could result in this cause code
>(ie: Netserver realizes call is gone and terminates it)??
Doubt it. In that case the NMC would probably hear from the modem the call
disconnected from before it heard from the Hiper ARC. I'm definately not an
expert on this but that is the way it was explained to me by NetWork
Engineering so if my answer is wrong it is because I was mis-informed.
>
>Thanks,
>
>Charles
>
>=-----------------= =
>| Charles Sprickman Internet Channel |
>| INCH System Administration Team (212)243-5200 |
>| spork@inch.com access@inch.com |
>= =----------------=
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) Dial out from HiPer DSP card From: Tatai SV Krishnan <tkrishna@bubba.ae.usr.com> Date: 1998-11-25 16:04:14
On Wed, 25 Nov 1998, Nair, Shibu (MED, Wipro Systems, Inc) wrote:
> Hi
>
> Iam new to Total Control Enterprise Network Hub.
>
> I have configured Channelised T1 ckt for the HiPer DSP card.
>
> Since i do no have Total Control Manager software, iam testing using the
> console connection.
>
> I did the following test to ensure the dialout from the HiPerDSP ...
>
> 1) I have dialed to my local telephone and received the dial tone..
> >mdm <telephone number>
>
> 2) I have dialed to a machine which has dialup setup
>
> At this time it is trying for dialing out and comes back to the mdm
> prompt.
>
> Iam not getting machine access when i dial....
The DSP cards also needs a packetbus card in the chassis like the hiper
arc or the NETServer - Do you have those?
krish
>
> mdm2> atdt<machine dialup number>
> mdm2>
>
> HiPer DSP Rev. 1.0.7
>
> Pls help on this problem....
>
> Regards
> Shibu
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) Scripting the config of a HARC From: norm_miller@3com.com Date: 1998-11-25 17:07:07
FYI,
Just some things I use.
As you may, or may not know, the ARC has a very handy little command called
DO. With the DO command you can execute a text file of commands, any ARC
commands. It also has TFTP capability which most 3COM products have.
What this allows you to do is TFTP several versions of the ARC code onto
the ARC and swap back and forth between versions by copying the file on the
ARC to netserve.dmf then rebooting the ARC. When the ARC comes up it sees
the netserve.dmf file and loads it. To swap back, copy the original file
to netserve.dmf, reboot etc. You should be able to keep 2 to 3 different
code files on the ARC at a time if needed. Depends on space.
To get the code file to the ARC you will need a TFTP client package. TFTP
is included in UNIX, however very few of us carry around a laptop with
Solaris or Linux on it. To get around this minor inconvenience 3COM in its
infinite wisdom (in the person of Dan Gill) wrote a little Windows TFTP
application that we give away to all of our customers that request it. It
is called 3CServer. You can pull it off the 3Com web site Go to
www.3com.com and search on 3CServer.
Getting back to the DO command, If you want to, you can create multiple ARC
configuration files. Thou shalt not use Notepad, Edit, or WordPad. You can
use Vi or Word. Just make sure that if you use Word that you save them
using the format Text Only. Otherwise, Word will add a carriage return line
feed that the ARC does not like.
What I use this for is that in the lab, I use a variety of different
configs, depending on what particular feature I am trying to test. I create
the config files in Word, and save them with the Text Only format and then,
using 3CServer, TFTP the files down to the ARC. When ready to run a
particular test do a DEL CONFIG on the ARC and it deletes all .cfg files
and reboots. Key point: DO NOT save the config files with a .cfg extension.
If you do, then when you issue the delete config command, you also lose
your text files with the config commands. Save it with a .txt or anything
as long as it is not .cfg. When the system comes up say No to the quick
config and do a DO textfilename.txt
The system is now configured.
I always have a SAVE ALL as the last command in the file and I reboot
before running any tests. Below is a sample config file:
This may be helpful in creating canned configs for installs. If someone was
good at writing Word forms to populate the site and ip variables this may
save quite a bit of time on installs.
regards,
/norm
Below is a sample config file:
Cut here =======================
set system name "SysName" transmit_authentication_name "SysName"
set system location "SysLocation"
set system contact "NotMe"
add snmp community public address 0.0.0.0 access RW
enable security_option remote_user_administration telnet
add user "!root" password "WhatEver" type login,manage
add ip network ip interface eth:1 address 220.10.10.2/C frame ethernet_ii
enable no
enable ip network ip
add ip defaultroute gateway 220.10.10.1 metric 1
add tftp client 0.0.0.0
set authentication primary_server xxx.xxx.xxx.xxx
set accounting primary_server xxx.xxx.xxx.xxx primary_secret MySecret
set accounting attribute_style netserver
set ntp primary_server xxx.xxx.xxx.xxx
enable ntp
reconfigure ip network ip interface eth:1
set ip network ip routing_protocol ripv1
add ip pool ippool initial_pool_address 130.20.20.2 size 128
set authentication primary_secret MySecret
disable authentication local
enable authentication hint_assigned
set modem_group all connection_type no_prompt message "\b"
set ppp session_start_message
enable service_loss_busyout radius
disable telnet trying_message
delete user adm
set interface eth:1 filter_access on
enable ip proxy_arp_all_dialin
set command history 50
save all
reboot after the save.
Cut here =======================
"Eric" <elorenzo@mediacity.com> on 11/20/98 05:49:07 PM
Please respond to usr-tc@lists.xmission.com
cc: (Norm Miller/US/3Com)
How can you script the configuration of a HARC? I was told that you can
TFTP the file onto the HARC and then type "do filename"
I'd like to be able to easily batch setup a HARC. With easy prompts,
like:
-What is my IP address?
-What is my netmask?
-What is my gateway?
etc.
Can someone send me some instructions on how to do this and maybe a sample
script?
Eric
---
Eric J. Lorenzo
Sr. Field Engineer
lorenzo@ispchannel.com
v:650.237.1465
http://www.ISPchannel.com
-
To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages send
"help" to the same address. Do not use quotes in your message.
Subject:(usr-tc) Trade HiperARC for HiperDSP From: billt@dscga.com Date: 1998-11-25 17:14:26
I have an extra Hiperarc card set I'd love to trade for a HiperDSP
card. Anyone interested?
Subject:Re: (usr-tc) new user log in problem with v.90 From: Tatai SV Krishnan <tkrishna@bubba.ae.usr.com> Date: 1998-11-26 10:59:06
On Tue, 24 Nov 1998, Gorkem Yuksel wrote:
> Hi!
> Everybody.
>
> I am a new guy on the block.(small ISP)
>
> I am having some problems with new user's logging in.
>
> It does not allow to logging in right away when it comes to new users.
>
> Sometimes it started working 7 hours later, 1.5 hours later depending on
> the Hiper DSP's mood, I guess.
>
> But all the old users connecting OK. All the new users you key in today
> will be ok by tommorrow.
>
> I have TCM, Hiper DSP cards, NMC card, Hiper ARC card, about 6 month old.
> We upgraded to v.90 a month ago, it worked fine I guess.
> (or I did not have customers who wanted an instant setup???)
>
> I have system release 3.3.
> Hiper ARC 1.1.11
> Hiper DSP T1/PRI 1.2.5
> NMC 16Meg 486 5.5.5
> TCM Windows 5.5.1
Hiper arc version 1.1.11 does not sound right. Guess you have 4.1.11.
When you see this problem can you get a trace - you can either do mon ppp
or mon radius from the hiper arc. That will tell us what the problem is.
krish
>
> I e-mailed for help to 3Com but no answer for 3 days...
>
> I do not have tech support contract...too much money..
>
> It started happening last Sat and Sun...Mon and Tue.. which is today.
> I had one new customer yesterday.. it did not rscognize his username and
> password right away, but started working about one and half hour later.
>
> I have another new one keyed in 14:33 pm today, it is not working yet
> (16:37).
> I just keep typing in his user name and password by using dial up every 15
> to 30 mins to see if it is working. I did not call this customer to set up
> his internet account yet. I may have to wait till tomorrow.
>
> Anybody have any good sugestions, please write me, I will appreciate it
> very much.
>
> Please write all the detailed instructions, since I am not that good..
>
>
> Ken. 905-826-0737 ken@gncom.com
>
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) new user log in problem with v.90 From: Tatai SV Krishnan <tkrishna@bubba.ae.usr.com> Date: 1998-11-26 12:02:10
On Thu, 26 Nov 1998, Gorkem Yuksel wrote:
> How do I do mom ppp and mom radius???
>
> just type "mom ppp" at the hiper>> prompt??
When you recreate the problem, logon to the HiPer arc as admin user.
Type mon ppp
It will give you a menu - you can monitor based on user name
choose that and capture the data it sends.
Then make another call and this type on the hiper arc type mon radius
that also provides you with a menu. Choose the user option and capture
the same.
Once you get this capture, send it to me and also let me know the version
number of the hiper arc, DSP and also what radius you are using.
regards
krish
>
> Ken.
>
> ----------
> From: Tatai SV Krishnan[SMTP:tkrishna@bubba.ae.usr.com]
> Reply To: usr-tc@lists.xmission.com
> Sent: Thursday, November 26, 1998 11:59 AM
> To: Gorkem Yuksel
> Cc: 'usr-tc@lists.xmission.com'
> Subject: Re: (usr-tc) new user log in problem with v.90
>
> On Tue, 24 Nov 1998, Gorkem Yuksel wrote:
>
> > Hi!
> > Everybody.
> >
> > I am a new guy on the block.(small ISP)
> >
> > I am having some problems with new user's logging in.
> >
> > It does not allow to logging in right away when it comes to new users.
> >
> > Sometimes it started working 7 hours later, 1.5 hours later depending on
> > the Hiper DSP's mood, I guess.
> >
> > But all the old users connecting OK. All the new users you key in today
> > will be ok by tommorrow.
> >
> > I have TCM, Hiper DSP cards, NMC card, Hiper ARC card, about 6 month old.
> > We upgraded to v.90 a month ago, it worked fine I guess.
> > (or I did not have customers who wanted an instant setup???)
> >
> > I have system release 3.3.
> > Hiper ARC 1.1.11
> > Hiper DSP T1/PRI 1.2.5
> > NMC 16Meg 486 5.5.5
> > TCM Windows 5.5.1
>
> Hiper arc version 1.1.11 does not sound right. Guess you have 4.1.11.
> When you see this problem can you get a trace - you can either do mon ppp
> or mon radius from the hiper arc. That will tell us what the problem is.
>
> krish
>
> >
> > I e-mailed for help to 3Com but no answer for 3 days...
> >
> > I do not have tech support contract...too much money..
> >
> > It started happening last Sat and Sun...Mon and Tue.. which is today.
> > I had one new customer yesterday.. it did not rscognize his username and
> > password right away, but started working about one and half hour later.
> >
> > I have another new one keyed in 14:33 pm today, it is not working yet
> > (16:37).
> > I just keep typing in his user name and password by using dial up every 15
> > to 30 mins to see if it is working. I did not call this customer to set up
> > his internet account yet. I may have to wait till tomorrow.
> >
> > Anybody have any good sugestions, please write me, I will appreciate it
> > very much.
> >
> > Please write all the detailed instructions, since I am not that good..
> >
> >
> > Ken. 905-826-0737 ken@gncom.com
> >
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) new user log in problem with v.90 From: Gorkem Yuksel <gorkem@gncom.com> Date: 1998-11-26 12:18:51
How do I do mom ppp and mom radius???
just type "mom ppp" at the hiper>> prompt??
Ken.
Subject:RE: (usr-tc) rcvdGatewayDiscCmd From: John Powell <jp@packet.ae.usr.com> Date: 1998-11-26 13:26:21
On Wed, 25 Nov 1998, Brian K McIntire wrote:
> >>From what I understand "rcvdGatewayDiscCmd" means that the ARC/Netserver
> >has essentially dropped DTR on a modem to cause a hangup.
>
> Not exactly. Remember the NMC is the one reporting to you a "received
> gateway disconnect command" What this means is the HiPer ARC told the NMC
> the user has disconnected before the NMC could get the info from somewhere
> else.
Not correct. Charles' original statement was correct.
This cause is stored in the modem and only transferred via the NMC, the
NMC's role is passive not active. You could get the same data by
attaching directly to the modem and executing an ATI6. Whatever the root
cause, it is definitely the gateway (ARC/NS/Edgeserver) instructing the
modem to hangup.
Common causes are idle-timeout, session-limit, failure to login in the
specified period, failure to authenticate, etc. I guess there could be
some session failures that might cause the gateway to hangup also, not
sure (sorry, just a modem geek here).
I am not sure, but I guess a user initiated disconnect, such as a telnet
session (non-PPP) terminated with an "exit", or some sort of PPP
disconnect could also cause this term cause. Normally the latter is done
with the more terse, V.42 disc or NormalUserCallClear (remote modem saying
adios in one way or another).
Just for kicks, I took a walk through one of our corp telecommuting
chassis (pretty typical ISP-like production environment) and found almost
all of the disconnects were V42 disc or normalUSerCallClear, the only
gateway initiated discs were session limit (3 hours in our case).
JP
Subject:RE: (usr-tc) rcvdGatewayDiscCmd From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-26 15:52:43
>> >>From what I understand "rcvdGatewayDiscCmd" means that the
>ARC/Netserver
>> >has essentially dropped DTR on a modem to cause a hangup.
>>
>> Not exactly. Remember the NMC is the one reporting to you a "received
>> gateway disconnect command" What this means is the HiPer ARC told the NMC
>> the user has disconnected before the NMC could get the info from
>somewhere
>> else.
>
>Not correct. Charles' original statement was correct.
>
>This cause is stored in the modem and only transferred via the NMC, the
>NMC's role is passive not active.
Makes more sense than what I was told over a year ago by a nameless 3COM
employee. I have always thought of the NMC as strictly a UI to the chassis.
Just trying to help. Sorry, I'll stop randomly quoting what I have been
told without thinking about it first.
In the interest of pulling at least one foot out of my mouth and Since no
one from 3COM has commented about the NAS side of this to help answer your
questions let me put my neck out again. :)
I have seen a bad card in the chassis causing "interference or garbage" on
the packet bus resulting in random disconnects where the modem reports
"Received Gateway Disconnect command"
So basically, anything ahead of the modem connection is going to show up
with that error. Anything on the modem, or initiated by the client modem,
or between the client modem and the chassis modem will show up with
something else. (Duh. Again; sorry)
Also, with some versions of quad modem software I have seen calls drop when
I rebooted another card on the chassis. Normally, I would say that means
there is a hardware malfunction but once that card came back up again and we
upgraded it we could not duplicate the problem again so I guess it was
software related.
Oh, I also saw it one time when we had a version of NetServer software with
a memory leak. All calls were eventually dropping or would not even
connect. Received Gateway disconnect command was always the reason. Sh mem
on the NetServer showed only a few bytes of memory remaining.
Of course, I've also seen the error when the HiPer ARC crashed.
I'll stick my neck out again and tell you something else I was told. I hope
3COM will forgive the reference. No-one expects them to be perfect.
It sounds as vague and strange now as when he first told me but.....When a
user initiates something where a response of some kind is expected but never
received a timeout occurs and the call may sometimes drop. Sounds technical
doesn't it. Makes no sense to me though. The NAS should handle all
requests and retransmit when necessary. Why would the user get dropped? To
me, one thing should have nothing to do with the other. If the user gets
dropped it should be a bug, but that's my opinion.
This is not a random quote. I've got a file a few megabytes in size with
all things I have ever been told by high level 3COM & former USR engineers
neatly entered the day they told me in as close to the original wording as
when they told me. I don't pretend to know everything so I keep it and use
it often. (I'm sure someone will try to flame me for this. Oh well.)
I hope starting this argument was at least helpful. Sometimes starting one
and putting your neck out is the only way to get anyone who really knows the
answer to comment on it. (They can smell the blood.) :) Anyway, e-mails
seen by the people who have the answers go unanswered all the time. At
least John commented on it and helped us out.
>You could get the same data by attaching directly to the modem and
executing an ATI6. Whatever >the root cause, it is definitely the gateway
(ARC/NS/EdgeServer) instructing the
>modem to hang-up.
>
>Common causes are idle-timeout, session-limit, failure to login in the
>specified period, failure to authenticate, etc. I guess there could be
>some session failures that might cause the gateway to hang-up also, not
>sure (sorry, just a modem geek here).
>
>I am not sure, but I guess a user initiated disconnect, such as a telnet
>session (non-PPP) terminated with an "exit", or some sort of PPP
>disconnect could also cause this term cause. Normally the latter is done
>with the more terse, V.42 disc or NormalUserCallClear (remote modem saying
>adios in one way or another).
>
>Just for kicks, I took a walk through one of our Corp telecommuting
>chassis (pretty typical ISP-like production environment) and found almost
>all of the disconnects were V42 disc or NormalUserCallClear, the only
>gateway initiated discs were session limit (3 hours in our case).
>
>JP
On Thu, 26 Nov 1998, Brian K McIntire wrote:
> >> >>From what I understand "rcvdGatewayDiscCmd" means that the
> >ARC/Netserver
> >> >has essentially dropped DTR on a modem to cause a hangup.
> >>
> >> Not exactly. Remember the NMC is the one reporting to you a "received
> >> gateway disconnect command" What this means is the HiPer ARC told the NMC
> >> the user has disconnected before the NMC could get the info from
> >somewhere
> >> else.
> >
> >Not correct. Charles' original statement was correct.
> >
> >This cause is stored in the modem and only transferred via the NMC, the
> >NMC's role is passive not active.
>
> Makes more sense than what I was told over a year ago by a nameless 3COM
> employee. I have always thought of the NMC as strictly a UI to the chassis.
> Just trying to help. Sorry, I'll stop randomly quoting what I have been
> told without thinking about it first.
It does not make sense. Don't say that you have been told by someone in
3com and you have a record - rather a journal of everything that was said
to you with date and time and with their signature to prove it etc etc.
Everyone in 3com who worked in the past ( USR ) and who is working now
and those people who have access to yadda yadda search can see what this
problem means.
This problem is simple and stright forward. The modem disconnected the
call because the modem was told to do so by the Gateway card. Once this
is done - the modem sends this info to the NMC or to the Gateway card and
in some cases to both to send the same to Trap destinations or to Radius
accounting servers.
A call can be disconnected by a Gateway card (NETServer, EdgeServer,
Hiper arc, X.25 Card, App Card ... etc) for the simple reason that it did
not like the call. Say for example during a ppp session you the hiper
arc receives lot of errors and this continues to happen to a certain
level - the the hiper arc can tell the modem to disconnect.
This method is a very normal way of clearing a call. If you feel that
there is a abnormal disconnects and you feel that for some reaon your
NETServer/HiPer arc is just disconnecting users - you need to go around
and find it why it is happening.
You need radius accounting logs, syslogs and then you must corelate them
with the gateway disconnect you see.
>
> In the interest of pulling at least one foot out of my mouth and Since no
> one from 3COM has commented about the NAS side of this to help answer your
> questions let me put my neck out again. :)
>
> I have seen a bad card in the chassis causing "interference or garbage" on
> the packet bus resulting in random disconnects where the modem reports
> "Received Gateway Disconnect command"
> So basically, anything ahead of the modem connection is going to show up
> with that error. Anything on the modem, or initiated by the client modem,
> or between the client modem and the chassis modem will show up with
> something else. (Duh. Again; sorry)
>
Not True, if you have a bad card in the chassis you will see the
connection to that port causing problems or that card is also capable of
disconnecting all the cards from the chassis - meaning you will have
modems get off the packet bus. But no Modem card is intelligent enough
to send out gateway disconnect reason without receiving them.
> Also, with some versions of quad modem software I have seen calls drop when
> I rebooted another card on the chassis. Normally, I would say that means
> there is a hardware malfunction but once that card came back up again and we
> upgraded it we could not duplicate the problem again so I guess it was
> software related.
You have to take every problem case by case. You cannot assume that
two problems can be uniquely the same - just because it looks the same.
This problem that you may have seen could be due to chassis awarness in
the quad 4.0 code but then again this was fixed long long ago.
>
> Oh, I also saw it one time when we had a version of NetServer software with
> a memory leak. All calls were eventually dropping or would not even
> connect. Received Gateway disconnect command was always the reason. Sh mem
> on the NetServer showed only a few bytes of memory remaining.
>
Memory leak --- If you run into this then you will have problems with
calls droped but again you will not be able to make new calls or even
telnet to the NETServer.
Yes in this case the gateway card has no memory so the calls were told be
droped.
> Of course, I've also seen the error when the HiPer ARC crashed.
>
Yes - this is possible again the modem did get a disconnect from the
hiper arc.
> I'll stick my neck out again and tell you something else I was told. I hope
> 3COM will forgive the reference. No-one expects them to be perfect.
> It sounds as vague and strange now as when he first told me but.....When a
> user initiates something where a response of some kind is expected but never
> received a timeout occurs and the call may sometimes drop. Sounds technical
> doesn't it. Makes no sense to me though. The NAS should handle all
> requests and retransmit when necessary. Why would the user get dropped? To
Again I have to ask you to take a look back and see in what situation
this particular statement was made.
Go back and take a look at your notes and see why this statement was made
and in what refrence this was made. There can be a situation which could
be true. I am not sure nor can I think of a situation right now but
there could be something like that. Just because it was told you does
not mean that you were told a lie. It all depends upon when and whom you
asked the question.
> me, one thing should have nothing to do with the other. If the user gets
> dropped it should be a bug, but that's my opinion.
>
A bug is when a user gets droped without any reason. If there is a valid
reason then you have to take a look why this happened and if the reason
is right its not a BUG.
> This is not a random quote. I've got a file a few megabytes in size with
> all things I have ever been told by high level 3COM & former USR engineers
> neatly entered the day they told me in as close to the original wording as
> when they told me. I don't pretend to know everything so I keep it and use
> it often. (I'm sure someone will try to flame me for this. Oh well.)
>
krish
> I hope starting this argument was at least helpful. Sometimes starting one
> and putting your neck out is the only way to get anyone who really knows the
> answer to comment on it. (They can smell the blood.) :) Anyway, e-mails
> seen by the people who have the answers go unanswered all the time. At
> least John commented on it and helped us out.
>
> >You could get the same data by attaching directly to the modem and
> executing an ATI6. Whatever >the root cause, it is definitely the gateway
> (ARC/NS/EdgeServer) instructing the
> >modem to hang-up.
> >
> >Common causes are idle-timeout, session-limit, failure to login in the
> >specified period, failure to authenticate, etc. I guess there could be
> >some session failures that might cause the gateway to hang-up also, not
> >sure (sorry, just a modem geek here).
> >
> >I am not sure, but I guess a user initiated disconnect, such as a telnet
> >session (non-PPP) terminated with an "exit", or some sort of PPP
> >disconnect could also cause this term cause. Normally the latter is done
> >with the more terse, V.42 disc or NormalUserCallClear (remote modem saying
> >adios in one way or another).
> >
> >Just for kicks, I took a walk through one of our Corp telecommuting
> >chassis (pretty typical ISP-like production environment) and found almost
> >all of the disconnects were V42 disc or NormalUserCallClear, the only
> >gateway initiated discs were session limit (3 hours in our case).
> >
> >JP
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
The last few Telepath user I tried to debug were connecting without v.42
or MNP 4 error correction -- both to Quads and DSP's. That may be the
real problem -- things often don't work too well when you can't correct
line errors. Disabling v.42 (S15=128) didn't seem to help... it didn't
connect with MNP 4 either.
I didn't research this too much farther... so there's likely a lot I'm
missing here. I suspect upgrading to v.90 may have helped -- I think they
were x2-only modems. But maybe it'll give you something else to look at?
Mike Andrews (MA12) icq 6602506 -------------- mandrews@termfrost.org
VP 'n' Systems/Network Administrator --------------- mandrews@dcr.net
Digital Crescent, Frankfort, KY ----------- http://www.termfrost.org/
# view;touch;unzip;finger;mount;mv;mv;mv;yes;mv;yes;mv;yes;umount;sleep
On Wed, 25 Nov 1998, Brian K McIntire wrote:
> >Just thought I would add that I experienced the same problem with this
> >version of code. Using mon ppp it seemed like it was experiencing the LCP
> >problem that web TV and imac's had. It did not get up to the
> >authentication stage. I'm still running 4.0.30 on our production chassis'
> >this being one of the reasons...
> >
> Thanks, that does help. We are seeing exactly the same thing on ours. The
> user attempts to connect, handshakes and goes no further. Monitor radius
> shows nothing. It just seems to be hanging. I guess I'll call 3COM and
> open a tkt with them. Thought I might have some bad code on that modem but
> if you are seeing it too then maybe it's a bug with something else.
>
> Thanks again
>
> >Jerry
> >
> >At 10:56 AM 11/25/98 -0600, you wrote:
> >>Does anyone know of issues with this particular modem. I'm working with a
> >>customer running 1.2.5 on his DSP's and 4.1.72 - 7 on his HiPer ARC. With
> >>the same user account I can connect 20 times out of 20 with my
> >courier. The
> >>client that has a Telepath connects 1 out of every 3-7 times.
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) Re: mpip From: C Thompson <cthompson@wingnet.net> Date: 1998-11-27 10:36:50
I have a question on setting up multiple MPIP servers. We have
three chasses (a, b, and c with address 1, 2, and 3 respectively).
If I want to set up 'a' as a server, I do the following:
Chassis Command
Thus spake C Thompson
>I have a question on setting up multiple MPIP servers. We have
>three chasses (a, b, and c with address 1, 2, and 3 respectively).
>If I want to set up 'a' as a server, I do the following:
[snip]
That should pretty much cover it.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
We have been using pmmon to limit multiple logins now for a few month and
have had great success with it. When I first installed it I heard from a
few angry customers who insisted that they were not connected more than
once. However, on every occassion I was able to go through our radius
logs and find records of multiple logins. Logging the
"Calling-Station-Id" really helped with this. Customers quiet down when
you can give them the telephone number the connections were made from.
Just my $.02
Bob
On Fri, 27 Nov 1998, Ralph Helfenberger wrote:
> Hi
> I've been follwing the thread "Solution for idle time limits and dual
> logins". It was very interesting for me. As far as I could follow there
> are two main solutions to prevent dual logins:
>
> 1) Every login runs through a single point RADIUS authentication. This
> point has a table with users already logged in and it disconnects users,
> if they already have an active session. Obvious drawbacks: If the RADIUS
> server doesn't get the notification of an "end of session" this user can
> login in again (exept if there is some kind of timeout mechanism). Other
> drawback: The RADIUS servers can't do load sharing. Other drawback:
> There is a single point of failure.
>
> 2) There is an independent server wich regularly polls all NAS for
> doubled logins. It disconnects all users wich are logged in twice.
> Drawback: users acutally get the access but will be disconnected later.
> This can cause a lot of support calls.
>
> I'd like to hear from you who is using wich mechanism and what the
> experiences are? How reliable does it run? I should implement something
> like that in an environment with about 10 TotalControl chassis
> (Netserver mainly).
>
> Thanks for your comments.
>
> Ralph
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:RE: (usr-tc) rcvdGatewayDiscCmd From: Brian K McIntire <bmcintire@commnet.com> Date: 1998-11-27 13:36:00
>-----Original Message-----
>From: owner-usr-tc@lists.xmission.com
>[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Tatai SV Krishnan
>Sent: Thursday, November 26, 1998 4:33 PM
>To: Brian K McIntire
>Cc: usr-tc@lists.xmission.com
>Subject: RE: (usr-tc) rcvdGatewayDiscCmd
>
>
>On Thu, 26 Nov 1998, Brian K McIntire wrote:
>
>> >> >>From what I understand "rcvdGatewayDiscCmd" means that the
>> >ARC/Netserver
>> >> >has essentially dropped DTR on a modem to cause a hangup.
>> >>
>> >> Not exactly. Remember the NMC is the one reporting to you a "received
>> >> gateway disconnect command" What this means is the HiPer ARC
>told the NMC
>> >> the user has disconnected before the NMC could get the info from
>> >somewhere
>> >> else.
>> >
>> >Not correct. Charles' original statement was correct.
>> >
>> >This cause is stored in the modem and only transferred via the NMC, the
>> >NMC's role is passive not active.
>>
>> Makes more sense than what I was told over a year ago by a nameless 3COM
>> employee. I have always thought of the NMC as strictly a UI to
>the chassis.
>> Just trying to help. Sorry, I'll stop randomly quoting what I have been
>> told without thinking about it first.
>
>It does not make sense. Don't say that you have been told by someone in
>3com and you have a record - rather a journal of everything that was said
>to you with date and time and with their signature to prove it etc etc.
>Everyone in 3com who worked in the past ( USR ) and who is working now
>and those people who have access to yadda yadda search can see what this
>problem means.
>
>This problem is simple and stright forward. The modem disconnected the
>call because the modem was told to do so by the Gateway card. Once this
>is done - the modem sends this info to the NMC or to the Gateway card and
>in some cases to both to send the same to Trap destinations or to Radius
>accounting servers.
>
>A call can be disconnected by a Gateway card (NETServer, EdgeServer,
>Hiper arc, X.25 Card, App Card ... etc) for the simple reason that it did
>not like the call. Say for example during a ppp session you the hiper
>arc receives lot of errors and this continues to happen to a certain
>level - the the hiper arc can tell the modem to disconnect.
>
>This method is a very normal way of clearing a call. If you feel that
>there is a abnormal disconnects and you feel that for some reaon your
>NETServer/HiPer arc is just disconnecting users - you need to go around
>and find it why it is happening.
>
>You need radius accounting logs, syslogs and then you must corelate them
>with the gateway disconnect you see.
I told you we could get someone to comment on this once we started arguing
about it
>>
>> In the interest of pulling at least one foot out of my mouth and Since no
>> one from 3COM has commented about the NAS side of this to help
>answer your
>> questions let me put my neck out again. :)
>>
>> I have seen a bad card in the chassis causing "interference or
>garbage" on
>> the packet bus resulting in random disconnects where the modem reports
>> "Received Gateway Disconnect command"
>> So basically, anything ahead of the modem connection is going to show up
>> with that error. Anything on the modem, or initiated by the
>client modem,
>> or between the client modem and the chassis modem will show up with
>> something else. (Duh. Again; sorry)
>>
>Not True, if you have a bad card in the chassis you will see the
>connection to that port causing problems or that card is also capable of
>disconnecting all the cards from the chassis - meaning you will have
>modems get off the packet bus. But no Modem card is intelligent enough
>to send out gateway disconnect reason without receiving them.
I have escalated a few tickets that were resolved by replacing bad quad
modem cards. The original symptom "Received gateway Disconnect Command".
Perhaps the tech that worked on each of the escalated tickets did something
else to resolve the problem and didn't document it. I don't know. I only
log information not ticket #'s (Unless they are my tickets). Wish i had
some tickets #'s to give you.
>
>
>> Also, with some versions of quad modem software I have seen
>calls drop when
>> I rebooted another card on the chassis. Normally, I would say that means
>> there is a hardware malfunction but once that card came back up
>again and we
>> upgraded it we could not duplicate the problem again so I guess it was
>> software related.
>
>You have to take every problem case by case. You cannot assume that
>two problems can be uniquely the same - just because it looks the same.
>This problem that you may have seen could be due to chassis awarness in
>the quad 4.0 code but then again this was fixed long long ago.
>
>>
>> Oh, I also saw it one time when we had a version of NetServer
>software with
>> a memory leak. All calls were eventually dropping or would not even
>> connect. Received Gateway disconnect command was always the
>reason. Sh mem
>> on the NetServer showed only a few bytes of memory remaining.
>>
>Memory leak --- If you run into this then you will have problems with
>calls droped but again you will not be able to make new calls or even
>telnet to the NETServer.
>Yes in this case the gateway card has no memory so the calls were told be
>droped.
>
>> Of course, I've also seen the error when the HiPer ARC crashed.
>>
>Yes - this is possible again the modem did get a disconnect from the
>hiper arc.
>
>
>> I'll stick my neck out again and tell you something else I was
>told. I hope
>> 3COM will forgive the reference. No-one expects them to be perfect.
>> It sounds as vague and strange now as when he first told me
>but.....When a
>> user initiates something where a response of some kind is
>expected but never
>> received a timeout occurs and the call may sometimes drop.
>Sounds technical
>> doesn't it. Makes no sense to me though. The NAS should handle all
>> requests and retransmit when necessary. Why would the user get
>dropped? To
>Again I have to ask you to take a look back and see in what situation
>this particular statement was made.
>Go back and take a look at your notes and see why this statement was made
>and in what refrence this was made. There can be a situation which could
>be true. I am not sure nor can I think of a situation right now but
>there could be something like that. Just because it was told you does
>not mean that you were told a lie. It all depends upon when and whom you
>asked the question.
>
>> me, one thing should have nothing to do with the other. If the user gets
>> dropped it should be a bug, but that's my opinion.
>>
>A bug is when a user gets droped without any reason. If there is a valid
>reason then you have to take a look why this happened and if the reason
>is right its not a BUG.
>
>> This is not a random quote. I've got a file a few megabytes in size with
>> all things I have ever been told by high level 3COM & former USR
>engineers
>> neatly entered the day they told me in as close to the original
>wording as
>> when they told me. I don't pretend to know everything so I keep
>it and use
>> it often. (I'm sure someone will try to flame me for this. Oh well.)
>>
>
>krish
>
>
>> I hope starting this argument was at least helpful. Sometimes
>starting one
>> and putting your neck out is the only way to get anyone who
>really knows the
>> answer to comment on it. (They can smell the blood.) :) Anyway, e-mails
>> seen by the people who have the answers go unanswered all the time. At
>> least John commented on it and helped us out.
>>
>> >You could get the same data by attaching directly to the modem and
>> executing an ATI6. Whatever >the root cause, it is definitely
>the gateway
>> (ARC/NS/EdgeServer) instructing the
>> >modem to hang-up.
>> >
>> >Common causes are idle-timeout, session-limit, failure to login in the
>> >specified period, failure to authenticate, etc. I guess there could be
>> >some session failures that might cause the gateway to hang-up also, not
>> >sure (sorry, just a modem geek here).
>> >
>> >I am not sure, but I guess a user initiated disconnect, such as a telnet
>> >session (non-PPP) terminated with an "exit", or some sort of PPP
>> >disconnect could also cause this term cause. Normally the
>latter is done
>> >with the more terse, V.42 disc or NormalUserCallClear (remote
>modem saying
>> >adios in one way or another).
>> >
>> >Just for kicks, I took a walk through one of our Corp telecommuting
>> >chassis (pretty typical ISP-like production environment) and
>found almost
>> >all of the disconnects were V42 disc or NormalUserCallClear, the only
>> >gateway initiated discs were session limit (3 hours in our case).
>> >
>> >JP
>>
>>
>> -
>> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
>> with "unsubscribe usr-tc" in the body of the message.
>> For information on digests or retrieving files and old messages send
>> "help" to the same address. Do not use quotes in your message.
>>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) Dual logins From: Ralph Helfenberger <rhelfenberger@comlight.ch> Date: 1998-11-27 17:57:46
Hi
I've been follwing the thread "Solution for idle time limits and dual
logins". It was very interesting for me. As far as I could follow there
are two main solutions to prevent dual logins:
1) Every login runs through a single point RADIUS authentication. This
point has a table with users already logged in and it disconnects users,
if they already have an active session. Obvious drawbacks: If the RADIUS
server doesn't get the notification of an "end of session" this user can
login in again (exept if there is some kind of timeout mechanism). Other
drawback: The RADIUS servers can't do load sharing. Other drawback:
There is a single point of failure.
2) There is an independent server wich regularly polls all NAS for
doubled logins. It disconnects all users wich are logged in twice.
Drawback: users acutally get the access but will be disconnected later.
This can cause a lot of support calls.
I'd like to hear from you who is using wich mechanism and what the
experiences are? How reliable does it run? I should implement something
like that in an environment with about 10 TotalControl chassis
(Netserver mainly).
Thanks for your comments.
Ralph
Subject:Re: (usr-tc) Dual logins From: Mark R. Lindsey <mark@vielle.datasys.net> Date: 1998-11-27 21:21:53
Quoth Ralph Helfenberger <rhelfenberger@comlight.ch> --
: 1) Every login runs through a single point RADIUS authentication. This
: point has a table with users already logged in and it disconnects users,
: if they already have an active session. Obvious drawbacks: If the RADIUS
: server doesn't get the notification of an "end of session" this user can
: login in again (exept if there is some kind of timeout mechanism).
And this drawback classifies such a method as a kludge. To actually
solve the problem, the authentication system is going to need to verify
that its database of connections is accurate.
: Other drawback: The RADIUS servers can't do load sharing. Other
: drawback: There is a single point of failure.
It's nontrivial, but you *can* make your RADIUS server set use a common
database among all of the servers. There are fault-tolerant network
databases to be had, and this is a good application of one.
We should be careful to generalize our solutions: if we design strictly
for one case -- where any multiple concurrent session is intolerable --
then won't be able to gracefully handle probable future cases, where
each customer has access to a certain number of concurrent sessions
that's greater than one.
Thus spake Mark R. Lindsey
>Quoth Ralph Helfenberger <rhelfenberger@comlight.ch> --
>: 1) Every login runs through a single point RADIUS authentication. This
>: point has a table with users already logged in and it disconnects users,
>: if they already have an active session. Obvious drawbacks: If the RADIUS
>: server doesn't get the notification of an "end of session" this user can
>: login in again (exept if there is some kind of timeout mechanism).
>And this drawback classifies such a method as a kludge. To actually
>solve the problem, the authentication system is going to need to verify
>that its database of connections is accurate.
I disagree that its a kludge...at least in theory. This really would be
the clean way to do it, the problem is, is that it seems that noone can
make RADIUS client systems that can guarentee that the right records get
to the right places. It would also be nice if RADIUS had the support
for an incremental check or something, but in theory, even that isn't
needed.
Now, of course, in practice, this doesn't work, and since we're dealing
with the real world, this isn't the answer. :/
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) Dual logins From: Mark R. Lindsey <mark@vielle.datasys.net> Date: 1998-11-28 09:49:06
: Thus spake Mark R. Lindsey
: >Quoth Ralph Helfenberger <rhelfenberger@comlight.ch> --
: >: If the RADIUS server doesn't get the notification of an "end of
: >: session" this user can login in again (exept if there is some kind
: >: of timeout mechanism).
:
: >And this drawback classifies such a method as a kludge.
:
: I disagree that its a kludge...at least in theory. This really would be
: the clean way to do it, the problem is, is that it seems that noone can
: make RADIUS client systems that can guarentee that the right records get
: to the right places.
Perhaps, but only if your theory doesn't account for servers (NASsen)
that die and network links that never come back (where `never' refers to
a very long time).
As with the case of network design, your theory *does* need to account
for all of the circumstances; as in this example, if you design around
perfect operation conditions in a world where of imperfect operating
conditions, then you have a poor design.
Take another example from a related discipline: we started using RAID
systems when it became obvious that our assumption of perfect discs is
invalid. By designing such that Disc Failure is a normal operating
condition, we are able to operate through it.
Similarly, we should *plan* on things breaking, because Excrement
Occurs. Any design that ignores this (at the higher layers, anyway) is
doomed to be replaced by one that does.
Thus spake Mark R. Lindsey
>Similarly, we should *plan* on things breaking, because Excrement
>Occurs. Any design that ignores this (at the higher layers, anyway) is
>doomed to be replaced by one that does.
Agreed...thus my qualifications of things being "in theory." Now, how
to plan for operational breakdowns (ala a disk dieing in a RAID set).
Incremental RADIUS checks would be a start, even ICMP reachability
checks would be a start. Making sure reboot RADIUS requests go through
is another important point. These of course are mostly protocol issues
that require changes, additions to the RADIUS protocol, so they really
don't help at the moment, but I think its the way to go rather than
using a possibly platform specific kludge like pmmon to check various
systems. That we have a RADIUS server (Cistron I think?) that *does*
manage to handle this doesn't make it any less of a kludge. :)
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
Subject:Re: (usr-tc) Dual logins From: Mark R. Lindsey <mark@vielle.datasys.net> Date: 1998-11-28 10:17:34
Quoth Jeff McAdams:
: Now, how to plan for operational breakdowns (ala a disk dieing in
: a RAID set). Incremental RADIUS checks would be a start, even ICMP
: reachability checks would be a start. Making sure reboot RADIUS
: requests go through is another important point. These of course are
: mostly protocol issues that require changes, additions to the RADIUS
: protocol [...]
I like RADIUS the way it is. (Maybe I'm simplistic.)
I think this problem can be solved externally to the protocol. Consider:
Each NAS has a table available via SNMP that has a list of all current
sessions, and the indeces of that table are the Port numbers, as sent to
the RADIUS accounting server.
As such, a database of current sessions could be maintained, and
verified. As new authentication requests are received, that database is
consulted, and entries in it are verified using SNMP.
To do this, we don't need a new RADIUS -- all we need is NASsen with
the appropriate tables. I understand that the ARC has such a table, and
that's a good start.
Subject:(usr-tc) radius on usr-tc From: newbie@www.datatone.com Date: 1998-11-28 15:08:02
Hi,
I posted this to portmaster-radius and didnt get any reply..so I am trying this list now..any help is greatly appreicated..
TIA
harish
Greetings:
I have a problem with radius and USR Total Control( I am new to USR TC). I am running radius v 1.16 with esv
a patch to control multiple logins, with 4 portmaster3 and 2 portmaster2 ( I am familiar with PM2 and PM3s) o
n a Linux redhat machine and everything is fine. Now I am trying to use the same radius server with USR TC, b
ut authentication fails with bad password..! I am using my laptop to login via dialup, If I dial into portmas
ter to login using same username and password, I get authenticated OK. If I change the phone number and dial
into USR TC (only phone number changed on laptop, nothing else), I do not get authenticated..radius with -x o
utput shows bad password..I change the phone number to login via PM and I am OK..
I tried seraching 3Com website and Livingston Mailing archive and dont see anything that can help..I have pos
ted the radius -x output from both PM and TC, while trying to log on..
Any one has seen these problem..or what am I doing wrong with TC..?
thanks in advance.
harish
radius entry for user test
#
test Password="UNIX"
#
OUTPUT (radius) while trying to logon via PM2:
radrecv: Request from host d0dcc009 code=1, id=11, length=74
User-Name = "test"
Password = "@\334\036\372d\375\374\346vH}\370\204\354\035\020"
NAS-IP-Address = xxx.xxx.192.9
NAS-Port = 1
NAS-Port-Type = Async
Service-Type = Framed-User
Framed-Protocol = PPP
Sending Ack of id 11 to d0dcc009 (516comm)
radrecv: Request from host d0dcc009 code=4, id=12, length=90
Acct-Session-Id = "65000004"
User-Name = "test"
NAS-IP-Address = xxx.xxx.192.9
NAS-Port = 1
NAS-Port-Type = Async
Acct-Status-Type = Start
Acct-Authentic = RADIUS
Service-Type = Framed-User
Framed-Protocol = PPP
Framed-IP-Address = xxx.xxx.192.248
Acct-Delay-Time = 0
User test logged on 516comm S1
Sending Accounting Ack of id 12 to d0dcc009 (516comm)
OUTPUT (radius) while trying to logon via TC:
radrecv: Request from host cf618602 code=1, id=52, length=188
User-Name = "test"
Password = ".\321\202\325\374\202\304\356\202\241\255\335\024\002?g"
NAS-IP-Address = xxx.xxx.134.2
NAS-Port = 31
Service-Type = Framed-User
Framed-Protocol = PPP
Vendor-Specific = ""
Vendor-Specific = ""
Vendor-Specific = ""
Vendor-Specific = ""
Vendor-Specific = ""
Vendor-Specific = ""
Vendor-Specific = ""
NAS-Identifier = "usr1"
Acct-Session-Id = "0802016c"
NAS-Port-Type = Async
Authenticate: from xxx.xxx.134.2 - Bad password: test
Sending Reject of id 52 to cf618602 (xxx.xxx.134.2)
radrecv: Request from host cf618602 code=1, id=52, length=188
User-Name = "test"
Password = ".\321\202\325\374\202\304\356\202\241\255\335\024\002?g"
NAS-IP-Address = xxx.xxx.134.2
NAS-Port = 31
Service-Type = Framed-User
Framed-Protocol = PPP
Vendor-Specific = ""
Vendor-Specific = ""
Vendor-Specific = ""
Vendor-Specific = ""
Vendor-Specific = ""
Vendor-Specific = ""
Vendor-Specific = ""
NAS-Identifier = "usr1"
Acct-Session-Id = "0802016c"
NAS-Port-Type = Async
Dropping duplicate: from xxx.xxx.134.2 - ID: 52
USR TC Version:
Command> ver
U.S. Robotics
Total Control (tm) NETServer Card V.34/ISDN with Frame Relay V3.8.1
Build date: Aug 11 1998
Build time: 13:49:21
Network Interface Card: Ethernet & Frame Relay Combination (26)
ISDN Interface Card : MUNICH32 (4)
Packet Bus Circuit : Enhanced
Check the secret keys. When you dial into the USR-Tc NETServer your
authentication is denied because of wrong password. Basically it looks
like a password mismatch. Check the secret. Just re-entry the radius
secret on the NETServer and in the radius clients file.
krish
\ T.S.V. Krishnan \
\ Network System Engineer \ ( : - : )
\ 3Com ............ \
----------------------------------------------/
tkrishna@bubba.ae.usr.com
----------------------------/ http://interproc.ae.usr.com ----/
The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
Any Sufficiently advanced bug is indistinguishable for a feature.
- Rick Kulawiec
On Sat, 28 Nov 1998 newbie@www.datatone.com wrote:
> Hi,
> I posted this to portmaster-radius and didnt get any reply..so I am trying this list now..any help is greatly appreicated..
>
> TIA
> harish
>
> Greetings:
>
> I have a problem with radius and USR Total Control( I am new to USR TC). I am running radius v 1.16 with esv
> a patch to control multiple logins, with 4 portmaster3 and 2 portmaster2 ( I am familiar with PM2 and PM3s) o
> n a Linux redhat machine and everything is fine. Now I am trying to use the same radius server with USR TC, b
> ut authentication fails with bad password..! I am using my laptop to login via dialup, If I dial into portmas
> ter to login using same username and password, I get authenticated OK. If I change the phone number and dial
> into USR TC (only phone number changed on laptop, nothing else), I do not get authenticated..radius with -x o
> utput shows bad password..I change the phone number to login via PM and I am OK..
>
> I tried seraching 3Com website and Livingston Mailing archive and dont see anything that can help..I have pos
> ted the radius -x output from both PM and TC, while trying to log on..
> Any one has seen these problem..or what am I doing wrong with TC..?
>
> thanks in advance.
>
> harish
>
> radius entry for user test
>
> #
> test Password="UNIX"
> #
>
> OUTPUT (radius) while trying to logon via PM2:
>
> radrecv: Request from host d0dcc009 code=1, id=11, length=74
> User-Name = "test"
> Password = "@\334\036\372d\375\374\346vH}\370\204\354\035\020"
> NAS-IP-Address = xxx.xxx.192.9
> NAS-Port = 1
> NAS-Port-Type = Async
> Service-Type = Framed-User
> Framed-Protocol = PPP
> Sending Ack of id 11 to d0dcc009 (516comm)
> radrecv: Request from host d0dcc009 code=4, id=12, length=90
> Acct-Session-Id = "65000004"
> User-Name = "test"
> NAS-IP-Address = xxx.xxx.192.9
> NAS-Port = 1
> NAS-Port-Type = Async
> Acct-Status-Type = Start
> Acct-Authentic = RADIUS
> Service-Type = Framed-User
> Framed-Protocol = PPP
> Framed-IP-Address = xxx.xxx.192.248
> Acct-Delay-Time = 0
> User test logged on 516comm S1
> Sending Accounting Ack of id 12 to d0dcc009 (516comm)
>
>
> OUTPUT (radius) while trying to logon via TC:
>
> radrecv: Request from host cf618602 code=1, id=52, length=188
> User-Name = "test"
> Password = ".\321\202\325\374\202\304\356\202\241\255\335\024\002?g"
> NAS-IP-Address = xxx.xxx.134.2
> NAS-Port = 31
> Service-Type = Framed-User
> Framed-Protocol = PPP
> Vendor-Specific = ""
> Vendor-Specific = ""
> Vendor-Specific = ""
> Vendor-Specific = ""
> Vendor-Specific = ""
> Vendor-Specific = ""
> Vendor-Specific = ""
> NAS-Identifier = "usr1"
> Acct-Session-Id = "0802016c"
> NAS-Port-Type = Async
> Authenticate: from xxx.xxx.134.2 - Bad password: test
> Sending Reject of id 52 to cf618602 (xxx.xxx.134.2)
> radrecv: Request from host cf618602 code=1, id=52, length=188
> User-Name = "test"
> Password = ".\321\202\325\374\202\304\356\202\241\255\335\024\002?g"
> NAS-IP-Address = xxx.xxx.134.2
> NAS-Port = 31
> Service-Type = Framed-User
> Framed-Protocol = PPP
> Vendor-Specific = ""
> Vendor-Specific = ""
> Vendor-Specific = ""
> Vendor-Specific = ""
> Vendor-Specific = ""
> Vendor-Specific = ""
> Vendor-Specific = ""
> NAS-Identifier = "usr1"
> Acct-Session-Id = "0802016c"
> NAS-Port-Type = Async
> Dropping duplicate: from xxx.xxx.134.2 - ID: 52
>
> USR TC Version:
>
> Command> ver
> U.S. Robotics
> Total Control (tm) NETServer Card V.34/ISDN with Frame Relay V3.8.1
> Build date: Aug 11 1998
> Build time: 13:49:21
>
> Network Interface Card: Ethernet & Frame Relay Combination (26)
> ISDN Interface Card : MUNICH32 (4)
> Packet Bus Circuit : Enhanced
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Thus spake Mark R. Lindsey
>I like RADIUS the way it is. (Maybe I'm simplistic.)
Of course, if RADIUS worked as it was *supposed* to with all that, most
of this would be unnecessary...*some* of it would still be needed, but
not nearly as much of it.
>Each NAS has a table available via SNMP that has a list of all current
>sessions, and the indeces of that table are the Port numbers, as sent to
>the RADIUS accounting server.
Someone (David?) correct me here if I'm wrong, but I don't know of any
pre-defined SNMP trees that have this information in it, meaning we're
back to each vendor defines thier own tree instance and table structure
that the system then has to be able to handle. Now, standardizing it
on SNMP is an improvement...but that does leave some of the boxes out
in the cold...ala NETServer.
--
Jeff McAdams Email: jeffm@iglou.com
Head Network Administrator Voice: (502) 966-3848
IgLou Internet Services (800) 436-4456
On Sat, 28 Nov 1998 newbie@www.datatone.com wrote:
> Hi,
> I have checked and entered the secret twice while trying to trouble shoot the problem..so I dont think it is the secret...
>
If secret keys are fine then it is a problem in terms of using PAP /
CHAP. Also check your user make sure that the netserver itself on flash
does not have a user with the same name.
krish
> thanks for the reply...
>
> harish
>
> >From owner-usr-tc@lists.xmission.com Sat Nov 28 16:44:30 1998
> Date: Sat, 28 Nov 1998 15:50:48 -0600 (CST)
> From: Tatai SV Krishnan <tkrishna@bubba.ae.usr.com>
> To: newbie@www.datatone.com
> cc: -v@www.datatone.com, usr-tc@lists.xmission.com
> Subject: Re: (usr-tc) radius on usr-tc
>
> Check the secret keys. When you dial into the USR-Tc NETServer your
> authentication is denied because of wrong password. Basically it looks
> like a password mismatch. Check the secret. Just re-entry the radius
> secret on the NETServer and in the radius clients file.
>
> krish
>
> -----------------------------------------
> \ T.S.V. Krishnan \
> \ Network System Engineer \ ( : - : )
> \ 3Com ............ \
> ----------------------------------------------/
> tkrishna@bubba.ae.usr.com
> ----------------------------/ http://interproc.ae.usr.com ----/
> The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
> -------------------------------------------------------------------------\
> Any Sufficiently advanced bug is indistinguishable for a feature.
> - Rick Kulawiec
> -------------------------------------------------------------------------/
>
> On Sat, 28 Nov 1998 newbie@www.datatone.com wrote:
>
> > Hi,
> > I posted this to portmaster-radius and didnt get any reply..so I am trying this list now..any help is greatly appreicated..
> >
> > TIA
> > harish
> >
> > Greetings:
> >
> > I have a problem with radius and USR Total Control( I am new to USR TC). I am running radius v 1.16 with esv
> > a patch to control multiple logins, with 4 portmaster3 and 2 portmaster2 ( I am familiar with PM2 and PM3s) o
> > n a Linux redhat machine and everything is fine. Now I am trying to use the same radius server with USR TC, b
> > ut authentication fails with bad password..! I am using my laptop to login via dialup, If I dial into portmas
> > ter to login using same username and password, I get authenticated OK. If I change the phone number and dial
> > into USR TC (only phone number changed on laptop, nothing else), I do not get authenticated..radius with -x o
> > utput shows bad password..I change the phone number to login via PM and I am OK..
> >
> > I tried seraching 3Com website and Livingston Mailing archive and dont see anything that can help..I have pos
> > ted the radius -x output from both PM and TC, while trying to log on..
> > Any one has seen these problem..or what am I doing wrong with TC..?
> >
> > thanks in advance.
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:Re: (usr-tc) radius on usr-tc From: newbie@www.datatone.com Date: 1998-11-28 17:47:51
Hi,
I have checked and entered the secret twice while trying to trouble shoot the problem..so I dont think it is the secret...
thanks for the reply...
harish
From owner-usr-tc@lists.xmission.com Sat Nov 28 16:44:30 1998
cc: -v@www.datatone.com, usr-tc@lists.xmission.com
Check the secret keys. When you dial into the USR-Tc NETServer your
authentication is denied because of wrong password. Basically it looks
like a password mismatch. Check the secret. Just re-entry the radius
secret on the NETServer and in the radius clients file.
krish
\ T.S.V. Krishnan \
\ Network System Engineer \ ( : - : )
\ 3Com ............ \
----------------------------------------------/
tkrishna@bubba.ae.usr.com
----------------------------/ http://interproc.ae.usr.com ----/
The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
Any Sufficiently advanced bug is indistinguishable for a feature.
- Rick Kulawiec
On Sat, 28 Nov 1998 newbie@www.datatone.com wrote:
> Hi,
> I posted this to portmaster-radius and didnt get any reply..so I am trying this list now..any help is greatly appreicated..
>
> TIA
> harish
>
> Greetings:
>
> I have a problem with radius and USR Total Control( I am new to USR TC). I am running radius v 1.16 with esv
> a patch to control multiple logins, with 4 portmaster3 and 2 portmaster2 ( I am familiar with PM2 and PM3s) o
> n a Linux redhat machine and everything is fine. Now I am trying to use the same radius server with USR TC, b
> ut authentication fails with bad password..! I am using my laptop to login via dialup, If I dial into portmas
> ter to login using same username and password, I get authenticated OK. If I change the phone number and dial
> into USR TC (only phone number changed on laptop, nothing else), I do not get authenticated..radius with -x o
> utput shows bad password..I change the phone number to login via PM and I am OK..
>
> I tried seraching 3Com website and Livingston Mailing archive and dont see anything that can help..I have pos
> ted the radius -x output from both PM and TC, while trying to log on..
> Any one has seen these problem..or what am I doing wrong with TC..?
>
> thanks in advance.
Subject:Re: (usr-tc) Seeking a way to 'hide' users from "list connections" From: Ricky Beam <jfbeam@interpath.net> Date: 1998-11-29 18:41:58
David Bolen was heard to say:
>Ricky Beam <jfbeam@Interpath.net> writes:
>
>> Hmm, this makes me wonder what happens if the RADIUS server rewrites the
>> username in the access-accept.
>
>Hmm, since a Username attribute isn't even permitted in an
>Access-Accept (see RFC2138, although I think I saw some e-mail
>comments about changing that to permit it fly by recently), other than
>ignoring it, I'm guessing that nothing would happen :-)
Like 3Com is known for following standards?
--Ricky
Hi
Thank you for the response....
I have these too...
My box containing one NMC , one ARC, and two HiPer DSP
which connected to T1 ckt.
Pls help me out...
TIA
Regards
Shibu
----------
From: Tatai SV Krishnan[SMTP:tkrishna@bubba.ae.usr.com]
Reply To: usr-tc@lists.xmission.com
Sent: Wednesday, November 25, 1998 4:04 PM
To: Nair, Shibu (MED, Wipro Systems, Inc)
Cc: 'usr-tc@xmission.com'
Subject: Re: (usr-tc) Dial out from HiPer DSP card
On Wed, 25 Nov 1998, Nair, Shibu (MED, Wipro Systems, Inc)
wrote:
> Hi
>
> Iam new to Total Control Enterprise Network Hub.
>
> I have configured Channelised T1 ckt for the HiPer DSP card.
>
> Since i do no have Total Control Manager software, iam testing
using the
> console connection.
>
> I did the following test to ensure the dialout from the
HiPerDSP ...
>
> 1) I have dialed to my local telephone and received the dial
tone..
> >mdm <telephone number>
>
> 2) I have dialed to a machine which has dialup setup
>
> At this time it is trying for dialing out and comes back to
the mdm
> prompt.
>
> Iam not getting machine access when i dial....
The DSP cards also needs a packetbus card in the chassis like
the hiper
arc or the NETServer - Do you have those?
krish
>
> mdm2> atdt<machine dialup number>
> mdm2>
>
> HiPer DSP Rev. 1.0.7
>
> Pls help on this problem....
>
> Regards
> Shibu
>
> -
> To unsubscribe to usr-tc, send an email to
"majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old
messages send
> "help" to the same address. Do not use quotes in your
message.
>
-
To unsubscribe to usr-tc, send an email to
"majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages
send
"help" to the same address. Do not use quotes in your message.
Hi
Thank you for the response....
I have these too...
My box containing one NMC , one ARC, and two HiPer DSP
which connected to T1 ckt.
Pls help me out...
TIA
Regards
Shibu
----------
From: Tatai SV Krishnan[SMTP:tkrishna@bubba.ae.usr.com]
Reply To: usr-tc@lists.xmission.com
Sent: Wednesday, November 25, 1998 4:04 PM
To: Nair, Shibu (MED, Wipro Systems, Inc)
Cc: 'usr-tc@xmission.com'
Subject: Re: (usr-tc) Dial out from HiPer DSP card
On Wed, 25 Nov 1998, Nair, Shibu (MED, Wipro Systems, Inc)
wrote:
> Hi
>
> Iam new to Total Control Enterprise Network Hub.
>
> I have configured Channelised T1 ckt for the HiPer DSP card.
>
> Since i do no have Total Control Manager software, iam testing
using the
> console connection.
>
> I did the following test to ensure the dialout from the
HiPerDSP ...
>
> 1) I have dialed to my local telephone and received the dial
tone..
> >mdm <telephone number>
>
> 2) I have dialed to a machine which has dialup setup
>
> At this time it is trying for dialing out and comes back to
the mdm
> prompt.
>
> Iam not getting machine access when i dial....
The DSP cards also needs a packetbus card in the chassis like
the hiper
arc or the NETServer - Do you have those?
krish
>
> mdm2> atdt<machine dialup number>
> mdm2>
>
> HiPer DSP Rev. 1.0.7
>
> Pls help on this problem....
>
> Regards
> Shibu
>
> -
> To unsubscribe to usr-tc, send an email to
"majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old
messages send
> "help" to the same address. Do not use quotes in your
message.
>
-
To unsubscribe to usr-tc, send an email to
"majordomo@xmission.com"
with "unsubscribe usr-tc" in the body of the message.
For information on digests or retrieving files and old messages
send
"help" to the same address. Do not use quotes in your message.
Subject:RE: (usr-tc) default routes From: Brian <signal@shreve.net> Date: 1998-11-29 21:08:29
On Wed, 25 Nov 1998, Brian K McIntire wrote:
> >-----Original Message-----
> >From: owner-usr-tc@lists.xmission.com
> >[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Tatai SV Krishnan
> >Sent: Tuesday, November 24, 1998 5:32 PM
> >To: Brian
> >Cc: USRobotics TC Mailing List
> >Subject: Re: (usr-tc) default routes
> >
> >
> >list rtab pre
> >
> >will show you the route table and the routes that is stored in flash
>
> I tried this. list rtab preferred did not show me the second default route
> i added. However, if i delete the primary default route the one I had
> enetered before with a metric of 2 showed up under list rtab preferred and
> list ip routes.
exactly, so how can you see both the defaults that you have added?
> >
> >krish
> >
> >-----------------------------------------
> > \ T.S.V. Krishnan \
> > \ Network System Engineer \ ( : - : )
> > \ 3Com ............ \
> > ----------------------------------------------/
> >tkrishna@bubba.ae.usr.com
> >----------------------------/ http://interproc.ae.usr.com ----/
> >The Yadda Yadda Search - for simple anwers -
> http://interproc.ae.usr.com/tkb.html
> -------------------------------------------------------------------------\
> Any Sufficiently advanced bug is indistinguishable for a feature.
> - Rick Kulawiec
> -------------------------------------------------------------------------/
>
> On Tue, 24 Nov 1998, Brian wrote:
>
> >
> > How do you view the default routes on an arc that are set?
> >
> > Brian
> >
> >
> > --------------------------------------------------------------------------
> > Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> > Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> > signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> > (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:RE: (usr-tc) default routes From: Brian <signal@shreve.net> Date: 1998-11-29 21:09:51
On Wed, 25 Nov 1998, Brian K McIntire wrote:
> >-----Original Message-----
> >From: owner-usr-tc@lists.xmission.com
> >[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Brian
> >Sent: Tuesday, November 24, 1998 7:00 PM
> >To: Tatai SV Krishnan
> >Cc: USRobotics TC Mailing List
> >Subject: Re: (usr-tc) default routes
> >
> >
> >
> >It doesnt show you all of them, only the one it is using. Watch, I start
> >off with 208.206.76.1 as my default route. Then add 208.206.76.44, which
> >then becomes the preffered route, then I drop 208.206.76.44 and it goes
> >back to 208.206.76.1.
>
> Did you specify a metric of 2 for the second entry? When i did it did not
> over ride the primary until the primary went down
no i used a metric of 1 for both.......which may be wrong, but you would
think it would still be able to display it somewhere.
when you say the primary went down, how does it know the state of the
primary?
>
> Where in the arc can I get a list of all the
> >currently added defaultroutes?
> >
> >Brian
> >
> >
> >HiPer>> list rtab pre
> >ROUTING TABLE PREFERRED ROUTES
> >Destination Prot Age NextHop Metric Interface
> > 0.0.0.0/0 REMOTE 25 208.206.76.1 1 eth:1
> >.
> >.
> >.
> >
> >HiPer>> add ip deFAULTROUTE gatEWAY 208.206.76.44 metric 1
> >HiPer>> list rtab pre
> >ROUTING TABLE PREFERRED ROUTES
> >Destination Prot Age NextHop Metric Interface
> > 0.0.0.0/0 REMOTE 25 208.206.76.44 1 eth:1
> >.
> >.
> >.
> >HiPer>> del ip deFAULTROUTE gatEWAY 208.206.76.44
> >HiPer>> list rtab pre
> >ROUTING TABLE PREFERRED ROUTES
> >Destination Prot Age NextHop Metric Interface
> > 0.0.0.0/0 REMOTE 25 208.206.76.1 1 eth:1
> >.
> >.
> >.
> >
> >
> >
> >
> >
> >
> >On Tue, 24 Nov 1998, Tatai SV Krishnan wrote:
> >
> >> list rtab pre
> >>
> >> will show you the route table and the routes that is stored in flash
> >>
> >> krish
> >>
> >> -----------------------------------------
> >> \ T.S.V. Krishnan \
> >> \ Network System Engineer \ ( : - : )
> >> \ 3Com ............ \
> >> ----------------------------------------------/
> >> tkrishna@bubba.ae.usr.com
> >> ----------------------------/ http://interproc.ae.usr.com ----/
> >> The Yadda Yadda Search - for simple anwers -
> http://interproc.ae.usr.com/tkb.html
> > -------------------------------------------------------------------------\
> > Any Sufficiently advanced bug is indistinguishable for a feature.
> > - Rick Kulawiec
> > -------------------------------------------------------------------------/
> >
> > On Tue, 24 Nov 1998, Brian wrote:
> >
> > >
> > > How do you view the default routes on an arc that are set?
> > >
> > > Brian
> > >
> > >
> >
> > --------------------------------------------------------------------------
> > > Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service
> Provider
> > > Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> > > signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> > > (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
> > >
> > >
> > > -
> > > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > > with "unsubscribe usr-tc" in the body of the message.
> > > For information on digests or retrieving files and old messages send
> > > "help" to the same address. Do not use quotes in your message.
> > >
> >
>
> --------------------------------------------------------------------------
> Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:(usr-tc) systest007 From: Brian Jacklin <csabmj@mail.tds.net> Date: 1998-11-29 21:10:16
I've got a netserver that is telling me that the ports are in systest007
state.
They work fine, user's are able to log and function without a problem.
How can I get rid of this message?
Thanks
Subject:RE: (usr-tc) default routes From: Brian <signal@shreve.net> Date: 1998-11-29 21:11:11
On Wed, 25 Nov 1998, Tatai SV Krishnan wrote:
> _list rtab en
HiPer>> _list rtab en
ROUTING TABLE ENTRIES
Destination Prot Fwd Pref Age NextHop Metric
Interface Flags
0.0.0.0/0 REMOTE 1 30 0 208.206.76.44 1
eth:1
yet if I delete that above default route, it will start using the first I
added which is 208.206.76.1............no where does it show the
208.206.76.1 in this command.
>
> krish
>
> -----------------------------------------
> \ T.S.V. Krishnan \
> \ Network System Engineer \ ( : - : )
> \ 3Com ............ \
> ----------------------------------------------/
> tkrishna@bubba.ae.usr.com
> ----------------------------/ http://interproc.ae.usr.com ----/
> The Yadda Yadda Search - for simple anwers - http://interproc.ae.usr.com/tkb.html
> -------------------------------------------------------------------------\
> Any Sufficiently advanced bug is indistinguishable for a feature.
> - Rick Kulawiec
> -------------------------------------------------------------------------/
>
> On Wed, 25 Nov 1998, Brian K McIntire wrote:
>
> > >-----Original Message-----
> > >From: owner-usr-tc@lists.xmission.com
> > >[mailto:owner-usr-tc@lists.xmission.com]On Behalf Of Tatai SV Krishnan
> > >Sent: Tuesday, November 24, 1998 5:32 PM
> > >To: Brian
> > >Cc: USRobotics TC Mailing List
> > >Subject: Re: (usr-tc) default routes
> > >
> > >
> > >list rtab pre
> > >
> > >will show you the route table and the routes that is stored in flash
> >
> > I tried this. list rtab preferred did not show me the second default route
> > i added. However, if i delete the primary default route the one I had
> > enetered before with a metric of 2 showed up under list rtab preferred and
> > list ip routes.
> > >
> > >krish
> > >
> > >-----------------------------------------
> > > \ T.S.V. Krishnan \
> > > \ Network System Engineer \ ( : - : )
> > > \ 3Com ............ \
> > > ----------------------------------------------/
> > >tkrishna@bubba.ae.usr.com
> > >----------------------------/ http://interproc.ae.usr.com ----/
> > >The Yadda Yadda Search - for simple anwers -
> > http://interproc.ae.usr.com/tkb.html
> > -------------------------------------------------------------------------\
> > Any Sufficiently advanced bug is indistinguishable for a feature.
> > - Rick Kulawiec
> > -------------------------------------------------------------------------/
> >
> > On Tue, 24 Nov 1998, Brian wrote:
> >
> > >
> > > How do you view the default routes on an arc that are set?
> > >
> > > Brian
> > >
> > >
> > > --------------------------------------------------------------------------
> > > Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> > > Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> > > signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> > > (318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
> > >
> > >
> > > -
> > > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > > with "unsubscribe usr-tc" in the body of the message.
> > > For information on digests or retrieving files and old messages send
> > > "help" to the same address. Do not use quotes in your message.
> > >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
> >
> > -
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:Re: (usr-tc) default routes From: Brian <signal@shreve.net> Date: 1998-11-29 21:13:46
On Tue, 24 Nov 1998, John Rockwell wrote:
> 'list ip defaultroute' should do the trick.
>
Thanks!
HiPer>> list ip defaULTROUTE
Configured default routes
Address Mask Gateway Metric State
0.0.0.0 0.0.0.0 208.206.76.44 1 ENABLED
0.0.0.0 0.0.0.0 208.206.76.1 1 ENABLED
btw, I know I should have made the second one a metric of 2
>
> At 05:16 PM 11/24/98 -0600, you wrote:
> >
> >How do you view the default routes on an arc that are set?
> >
> >Brian
> >
> >
> >--------------------------------------------------------------------------
> >Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
> >Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
> >signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
> >(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
> >
> >
> >-
> > To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> > with "unsubscribe usr-tc" in the body of the message.
> > For information on digests or retrieving files and old messages send
> > "help" to the same address. Do not use quotes in your message.
> >
> -------------------------------------
> John Rockwell
> e-mail: jrockwel@clarityconnect.com
> Network Engineer
> Clarityconnect, Inc.
> Ithaca Area: (607)257-8268
> Outside Ithaca Area: (888)322-4900
> Try us: http://www.clarityconnect.com
> -------------------------------------
>
> -
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
Subject:(usr-tc) duck sound on modems From: Brian <signal@shreve.net> Date: 1998-11-29 21:16:46
Anyone ever dial up their TC hub (hdm's) and hear what sounds like a duck
quack? I think its the sound of a modem half way thru negotiation to be
honest...........
Brian Feeny (BF304) | ShreveNet Inc. - Premium Internet Service Provider
Network Administrator | Shreveport, Louisiana - http://www.shreve.net/
signal@shreve.net | Web Hosting, Virtual Domains, Storefronts,
(318)222-2NET x 109 | Database/Web Integration, 56k, ISDN, T1
I have a HyperDSP running a chanelized T1. We had a problem with only 9 of
the 24 channels working tonight. When I finally got someone to look at the
provisioning at the telco (Ameritech), they told me that 15 channels had
been removed from service by the switch. The reason the person gave me was
that the wink that the switch was sending out was set to happen every 250
milliseconds, and wink back from our DSP was happening every 150 ms. They
explained that after a while, the switch determines that it is getting no
response from my equipment, and it will automatically remove the channel
from service.
Anyone ever hear of this? Is there a setting for wink timing on the
HiperDSP?
Russ Miescke
Power Web Connect
russm@powerweb.net
http://www.powerweb.net
I experienced the same issue with Pacific Bell. Make sure that the Telco has
no custom channel blocking features enabled. You can see if this is so from
HiPer DSP NIC console port. I was showing several channels out of service
due to custom channel blocking.
-----Original Message-----
>I have a HyperDSP running a chanelized T1. We had a problem with only 9 of
>the 24 channels working tonight. When I finally got someone to look at the
>provisioning at the telco (Ameritech), they told me that 15 channels had
>been removed from service by the switch. The reason the person gave me was
>that the wink that the switch was sending out was set to happen every 250
>milliseconds, and wink back from our DSP was happening every 150 ms. They
>explained that after a while, the switch determines that it is getting no
>response from my equipment, and it will automatically remove the channel
>from service.
>
>Anyone ever hear of this? Is there a setting for wink timing on the
>HiperDSP?
>
>
>Russ Miescke
>Power Web Connect
>russm@powerweb.net
>http://www.powerweb.net
>
>
>
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
Subject:(usr-tc) PRI card, or Used Rack with PRI card and V.90 wanted From: Dwight G. Jones <djones@imagen.net> Date: 1998-11-30 10:52:21
Please contact djones@imagen.net
Subject:RE: (usr-tc) rcvdGatewayDiscCmd From: David Bolen <db3l@ans.net> Date: 1998-11-30 12:29:06
Tatai SV Krishnan <tkrishna@bubba.ae.usr.com> writes:
> A call can be disconnected by a Gateway card (NETServer, EdgeServer,
> Hiper arc, X.25 Card, App Card ... etc) for the simple reason that it did
> not like the call. Say for example during a ppp session you the hiper
> arc receives lot of errors and this continues to happen to a certain
> level - the the hiper arc can tell the modem to disconnect.
While the majority of services discussed here relate to PPP sessions,
just for another data point, the disconnect originating from the
NETServer/ARC is also fairly typical if you are using tunneled TCP
sessions (e.g., netdata) to service async users, over which a higher
order protocol is running.
In such a case, a proper shutdown by the end user creates a disconnect
sequence in the protocol being tunneled, at which point the TCP
session is going to tear down. But you end up in a race condition;
either the customer is going to disconnect the modem which will be
seen over the telco call path, or the server is going to disconnect
the TCP session which is seen by the NETServer. Depending on which
arrives at the chassis first, that call will either disconnect (as
perceived by the modem which is in the "middle") from the network
(NETServer/ARC) side or the telco (DS0 channel) side.
One could also imagine that if a customer's PC is slow to disconnect
(or has some fixed delay before hangup) the actual phone line after
shutting down the PPP stack, that a similar race condition might exist
even for PPP sessions.
On our network (which takes a lot of tunneled TCP calls) we see
somewhere around a 15% disconnect as a system wide average from the
network side with the rcvdGatewayDiscCmd, generally second only
(although by a large margin) to a normal V.42 disconnect from the
telco side.
-- David
/-----------------------------------------------------------------------\
\ David Bolen \ Internet: db3l@ans.net /
| ANS Communications, Inc. \ Phone: (914) 701-5327 |
/ 100 Manhattanville Rd, Purchase, NY 10577 \ Fax: (914) 701-5310 \
\-----------------------------------------------------------------------/
Subject:Re: (usr-tc) Seeking a way to 'hide' users from "list connections" From: David Bolen <db3l@ans.net> Date: 1998-11-30 12:31:53
Ricky Beam <jfbeam@Interpath.net> writes:
> Like 3Com is known for following standards?
Hmm - it's hard to answer a statement like that in blanket terms, but
it seems to me the TC chassis does pretty well in that regard.
Certainly with respect to this discussion and RADIUS it fits the RFC.
There are some nits but nothing overly significant to my mind -
probably one of the bigger ones is their construction of vendor
specific attributes, but that's not in violation of the RFC, just not
the suggested format - and I sort of like 3Com's format better
anyway.
-- David
Subject:Re: (usr-tc) Upgrading Network Manager Card 4Meg. .... From: Victor J. Velazquez <victorv@infi.net> Date: 1998-11-30 13:57:17
Copy the 5.3.2 file to the sdl subdirectory of the USRSuite directory. To
upgrade the NMC to 5.3.2 you need to use TCM. In TCM select Configure then
select download and check the box for the NMC and click OK.
At 09:39 PM 11/30/98 +0400, you wrote:
>Hi,
>I wish to upgrade NMC 4Meg to ver. 5.3.2 for Y2K compliance. Present
version on NMC is 4.3.9. I have already downloaded the file. What I don't
know is
>how to apply this patch. I can see a executable file pcsdl.exe upon
>unzipping the code. Can someone please help me in knowing the correct
>procedure for applying this patch.
>
>Regards
>mahinderj@ohitelecom.com
>
>-
> To unsubscribe to usr-tc, send an email to "majordomo@xmission.com"
> with "unsubscribe usr-tc" in the body of the message.
> For information on digests or retrieving files and old messages send
> "help" to the same address. Do not use quotes in your message.
>
>
Hi,
I wish to upgrade NMC 4Meg to ver. 5.3.2 for Y2K compliance. Present =
version on NMC is 4.3.9. I have already downloaded the file. What I =
don't know is
how to apply this patch. I can see a executable file pcsdl.exe upon
unzipping the code. Can someone please help me in knowing the correct
procedure for applying this patch.
Regards
mahinderj@ohitelecom.com