May 1997

With what format should I add the Vendor Specific attributes to my (Merit 2.4.23) RADIUS dictionary? I've tried this; VALUE Vendor-Specific IP-Input-Filter 0x9001 VALUE Vendor-Specific IPX-Input-Filter 0x9002 VALUE Vendor-Specific IP-Output-Filter 0x9003 ...etc... but I'm still getting entries in my detail file showing (amongst other correct data); Acct-Authentic = RADIUS Vendor-Specific = Vendor-Specific = Vendor-Specific = Vendor-Specific = NAS-Identifier = nsc-vt1 ...etc... TIA -- Phil Dye | Work: pmd@tcp.net.uk Network Manager | Play: phil@lart.ing.co.uk Total Connectivity Providers | Consider myself properly disclaimed "Make it idiot-proof and someone will make a better idiot" - Anon

On Fri, 2 May 1997, Phil Dye wrote: > With what format should I add the Vendor Specific attributes to > my (Merit 2.4.23) RADIUS dictionary? > > I've tried this; > VALUE Vendor-Specific IP-Input-Filter 0x9001 > VALUE Vendor-Specific IPX-Input-Filter 0x9002 > VALUE Vendor-Specific IP-Output-Filter 0x9003 > ...etc... > > but I'm still getting entries in my detail file showing (amongst > other correct data); > Acct-Authentic = RADIUS > Vendor-Specific = > Vendor-Specific = > Vendor-Specific = > Vendor-Specific = > NAS-Identifier = nsc-vt1 > ...etc... I would like to know the same. Apparently, from what I understand, you must use a hacked up version of radius, and a special dictionary file.........is ANYONE logging vendor specific USR info on there UNIX radius? such as nmc logging etc.? Brian > > > TIA > > -- > Phil Dye | Work: pmd@tcp.net.uk > Network Manager | Play: phil@lart.ing.co.uk > Total Connectivity Providers | Consider myself properly disclaimed > "Make it idiot-proof and someone will make a better idiot" - Anon > oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo Brian Feeny oo ShreveNet, Inc. oo Phone: (318) 222-2NET UNIX Administrator oo 333 Texas St #619 oo FAX: (318) 221-6612 signal@shreve.net oo Shreveport, LA 71101 oo http://www.shreve.net/ oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo

Brian <signal@shreve.net> writes: > I would like to know the same. Apparently, from what I understand, you > must use a hacked up version of radius, and a special dictionary > file......... (Note, this note gives a bit of background/summary of what needs to be done with the Vendor-Specific attribute but in the end, I'm not sure of publically available servers fully supporting it - and perhaps someone from USR can comment on any availability of their Merit mods) Yes, at a minimum you will need a version of a RADIUS server (or accounting server) that can parse vendor specific fields. That's because at the uppermost level, a "Vendor-Specific" attribute is just a string attribute, an opaque series of character. So it doesn't make sense to define a numeric "VALUE" for a Vendor-Specific attribute because the attribute is more than just that value. The first four bytes are a vendor id (the enterprise ID of the vendor - 429 for USR) and the rest are opaque bytes. RFC 2058 suggests a standard TLV (type-length-value) format, but even that can differ among vendors since some use differing sizes for the length, and the use of a separate length byte for the attribute can be considered redundant. For the NETServer, Vendor-Specific fields are a string with the first 4 (vendor id) bytes set to 429. The next four bytes define the USR-specific attribute number. And finally the remainder of the field is the attribute value (whose length is computed based on the overall length of the vendor-specific attribute). I believe USR had a modified version of Merit's server (based on 2.4.20 I think) that was changed to support this form of vendor specific attributes. Presumably the Windows version also does. I don't know if direct support made it into a main Merit release or not. How to define vendor attributes in the dictionary was a discussion a while back on the RADIUS list. Different dictionaries may do it differently (my own is a hierarchical, dotted notation format) but the USR mods to the Merit stuff followed the mailing list discussion, and basically had two parts: * Define vendors in a vendors table with a attribute and value keyword for use in the main dictionary along with the vendor identifier. (And in the USR version of the Merit server the length of the attribute type field). Something like: USR-ATTRIBUTE USR-VALUE 429 4 * Then, in the main dictionary you define vendor-specific attributes using the new keywords. By using the keywords you tell the server to parse out the vendor specific stuff out of the attribute string. USR-ATTRIBUTE IP-Input-Filter 0x9001 or for something with a value: USR-ATTRIBUTE RIPV2 36873 USR-VALUE RIPV2 Off 0 USR-VALUE RIPV2 On 1 > is ANYONE logging vendor specific USR info on there UNIX > radius? such as nmc logging etc.? The answer to this is most definitely yes, however at least in our case we are using a locally developed authentication system and not something basic on Merit or other publically available code. -- David /-----------------------------------------------------------------------\ \ David Bolen \ Internet: db3l@ans.net / | ANS Communications \ Phone: (914) 789-5327 | / 100 Clearbrook Road, Elmsford, NY 10523 \ Fax: (914) 789-5310 \ \-----------------------------------------------------------------------/

When running 3.5.29 on the netserver (20mb's ram) and merit radius 2.4.23, I'm finding that most logins are coming in as 'dumb' instead of 'PPP' -- radius log -- Sun May 4 05:36:27 1997: rad_authenticate: 17/103 'master' at Total-Control dumb -- Does this have anything to do with the Sxx login port settings, are there any defaults I need to change in the netserver? I've downgraded back to 3.3.28 in the meantime. - lv

On Sun, 4 May 1997, Laszlo Vecsey wrote: > When running 3.5.29 on the netserver (20mb's ram) and merit radius 2.4.23, > I'm finding that most logins are coming in as 'dumb' instead of 'PPP' > > -- radius log -- > Sun May 4 05:36:27 1997: rad_authenticate: 17/103 'master' at Total-Control dumb > -- > > Does this have anything to do with the Sxx login port settings, are there > any defaults I need to change in the netserver? I've downgraded back to > 3.3.28 in the meantime. yes, have the port type set as "Netwrk" or "Login/Netwrk" instead of just "Login" Login=dumb Netwrk=PPP,SLIP,etc Brian also, when doing a "show SXX" the ACTIVE and DEFAULT profiles are reversed, a simple bug USR still hasnt fixed. Brian > > - lv > > oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo Brian Feeny oo ShreveNet, Inc. oo Phone: (318) 222-2NET UNIX Administrator oo 333 Texas St #619 oo FAX: (318) 221-6612 signal@shreve.net oo Shreveport, LA 71101 oo http://www.shreve.net/ oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo

On Sun, 4 May 1997, Laszlo Vecsey wrote: > When running 3.5.29 on the netserver (20mb's ram) and merit radius 2.4.23, > I'm finding that most logins are coming in as 'dumb' instead of 'PPP' > > -- radius log -- > Sun May 4 05:36:27 1997: rad_authenticate: 17/103 'master' at Total-Control dumb > -- > > Does this have anything to do with the Sxx login port settings, are there > any defaults I need to change in the netserver? I've downgraded back to > 3.3.28 in the meantime. yes, have the port type set as "Netwrk" or "Login/Netwrk" instead of just "Login" Login=dumb Netwrk=PPP,SLIP,etc Brian also, when doing a "show SXX" the ACTIVE and DEFAULT profiles are reversed, a simple bug USR still hasnt fixed. Brian > > - lv > > oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo Brian Feeny oo ShreveNet, Inc. oo Phone: (318) 222-2NET UNIX Administrator oo 333 Texas St #619 oo FAX: (318) 221-6612 signal@shreve.net oo Shreveport, LA 71101 oo http://www.shreve.net/ oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo

I just started hitting a situation where my Radius accounting session IDs have rolled over and started being used again in 2 months! Everyone I've asked said this should happen at least a year later. This is VERY bad because my SQL server is not inserting those accounting IDs because they are duplicates and I'm losing hundreds of transactions per hour! Is this happening to everyone? How does the ENH generate the ID? Can the ID length be increased? Can I somehow "bump" the ID to a new range? HELP! Tom

> On Sun, 4 May 1997, Laszlo Vecsey wrote: > > > When running 3.5.29 on the netserver (20mb's ram) and merit radius 2.4.23, > > I'm finding that most logins are coming in as 'dumb' instead of 'PPP' > > > > yes, have the port type set as "Netwrk" or "Login/Netwrk" instead of just > "Login" > I just checked the ports. None of the ports were set to Login, the majority were set to Login/Netwrk (producing dumb), and some were set to just Netwrk (which explain the small percentage of PPP logins in the trials I made). So I guess I'll be changing them all to Netwrk when I upgrade next time. - lv

: I just started hitting a situation where my Radius accounting session IDs : have rolled over and started being used again in 2 months! Everyone : I've asked said this should happen at least a year later. I just checked RFC 2139 on RADIUS accounting, and, while it's intentionally vague on that, it doesn't sound unreasonable to assume that a period of two months is enough time to rotate the logs. : This is VERY bad because my SQL server is not inserting those accounting : IDs because they are duplicates and I'm losing hundreds of transactions : per hour! Perhaps you could prepend the year and julian day or some such to the ID before passing it to the SQL server. (Unless, of course, you're using NT and don't have access to any of the sources, in which case I'll say a prayer for you.) (As if this is an industry so well established and so well developed that its participants need to throw out the tool of their trade -- programming -- and rely on manufacturers for perfect solutions. But, alas, we have to work with what we can get. Please pardon this diatribe.) --- Mark R. Lindsey, mark@datasys.net DSS Online Network Engineer (912) 241-0607, Fax: 241-0190

On Sun, 4 May 1997, Laszlo Vecsey wrote: > > On Sun, 4 May 1997, Laszlo Vecsey wrote: > > > > > When running 3.5.29 on the netserver (20mb's ram) and merit radius 2.4.23, > > > I'm finding that most logins are coming in as 'dumb' instead of 'PPP' > > > > > > > yes, have the port type set as "Netwrk" or "Login/Netwrk" instead of just > > "Login" > > > > I just checked the ports. None of the ports were set to Login, the > majority were set to Login/Netwrk (producing dumb), and some were set to > just Netwrk (which explain the small percentage of PPP logins in the > trials I made). So I guess I'll be changing them all to Netwrk when I > upgrade next time. thats not the problem though. The ones that said just "Netwrk" said just "Netwrk" because they were in use, and being used as a Netwrk connection. while the ones that said Login/Netwrk were waiting for callers. Brian > > - lv > > oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo Brian Feeny oo ShreveNet, Inc. oo Phone: (318) 222-2NET UNIX Administrator oo 333 Texas St #619 oo FAX: (318) 221-6612 signal@shreve.net oo Shreveport, LA 71101 oo http://www.shreve.net/ oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo

I'm using RadiusNT and for an Insert to happen it needs to have a unique Accounting ID, NAS, Nas Port and Account Status Type. I added the User's ID to this constraint and that seems to have cleared up the problem. Unless the same user logins into the same port X number of months ago and gets the same Accounting ID. Unlikely (I hope). Tom At 01:54 PM 5/4/97 -0400, you wrote: >: I just started hitting a situation where my Radius accounting session IDs >: have rolled over and started being used again in 2 months! Everyone >: I've asked said this should happen at least a year later. > >I just checked RFC 2139 on RADIUS accounting, and, while it's intentionally >vague on that, it doesn't sound unreasonable to assume that a period >of two months is enough time to rotate the logs. > >: This is VERY bad because my SQL server is not inserting those accounting >: IDs because they are duplicates and I'm losing hundreds of transactions >: per hour! > >Perhaps you could prepend the year and julian day or some such to the >ID before passing it to the SQL server. (Unless, of course, you're using >NT and don't have access to any of the sources, in which case I'll say >a prayer for you.) > >(As if this is an industry so well established and so well developed that >its participants need to throw out the tool of their trade -- programming -- >and rely on manufacturers for perfect solutions. But, alas, we have to work >with what we can get. > >Please pardon this diatribe.) > >--- >Mark R. Lindsey, mark@datasys.net >DSS Online Network Engineer >(912) 241-0607, Fax: 241-0190 > >

This has been going on for months and noone seems to know why. I'm using RadiusNT but my Portmasters return all the appropriate reasons, just not the USR TCs. Talking to USR is like talking to a rock. I'm tired of "I dunno, just run our stuff." Is anyone running something other than USR accounting stuff? If so, then are you getting real disconnect reasons? The lack of good accounting makes it difficult to explain to a user what is going on. Thanks, Tom

On Wed, 7 May 1997, Tom Bilan wrote: > Is anyone running something other than USR accounting stuff? If > so, then are you getting real disconnect reasons? The lack of good > accounting makes it difficult to explain to a user what is going on. I have the same problem with Livingston Radius 2.0. I was grepping for disconnect reasons, and they all say lost-carrier. It doesn't appear to be a Merit Radius specific thing. I was thinking about calling USR about this. There is no reason that this should function right only with TC security server. Brian

I have a few users running netmedic on their Win95 systems and occasionally (I believe when the Total Control unit is mostly full) I get email reports stating that there is a high entry router delay.. -- ISP Entry Router Delay: 134 ISP Egress Router Delay: 0 -- The router is a Cisco 2511 running 10.3(10), which is connected to a 3com 10baseT switch and the Total Control. Any ideas of whether this might be a problem with the netserver, cisco software, or even netmedic? Also some people are complaining about Quake delays, and this appears to be related. I have PPP in modem on. - lv

Laszlo Vecsey said once upon a time: >I have a few users running netmedic on their Win95 systems and >occasionally (I believe when the Total Control unit is mostly full) I get >email reports stating that there is a high entry router delay.. > >-- >ISP Entry Router Delay: 134 >ISP Egress Router Delay: 0 >-- > >The router is a Cisco 2511 running 10.3(10), which is connected to a 3com >10baseT switch and the Total Control. > >Any ideas of whether this might be a problem with the netserver, cisco >software, or even netmedic? Also some people are complaining about Quake >delays, and this appears to be related. I have PPP in modem on. Routing table entries are being held-down in your Cisco. You need to reduce your hold-down time to the minimum, which is 1 second. Otherwise, new entries can not make it into your table. It is also a good idea to make a static route for your address pool in your Cisco, so the delays only affect the on-demand networks. I've bitched at Cisco about the inability to remove hold-down completely from RIP, and they've pretty much ignored it. It causes a lot of headaches when you do extensive BGP. Let me know if you need any help on any of the above. P.S. This isn't related to the Quake problem, which is UDP throughput, not routing.

On Wed, 7 May 1997, Laszlo Vecsey wrote: > I have a few users running netmedic on their Win95 systems and > occasionally (I believe when the Total Control unit is mostly full) I get > email reports stating that there is a high entry router delay.. > > -- > ISP Entry Router Delay: 134 > ISP Egress Router Delay: 0 > -- > > The router is a Cisco 2511 running 10.3(10), which is connected to a 3com > 10baseT switch and the Total Control. > > Any ideas of whether this might be a problem with the netserver, cisco > software, or even netmedic? I've played with Netmedic and have noticed that it guesses on alot of things and also is set to be terribly sensitive--pointing out quite normal activity with yellow and even red lights making it seem like a network is more congested than it really is. Hell, it even gives me a warning that I need to upgrade my machine after playing Quake saying that the cpu has been at 100%. Although a handy and amusing tool, I'm worried about all of the emails we're all going to receive from well-intentioned customers.

1> Can anyone tell me what this trap message means: No loop current. Apparently 10-15% of our connections are failing because of this problem. I have a Total Control chassie with a netserver(3.3.28) and quad modems with analog lines. 2> On this same chassie from time to time dial in users once authenticated recieve 0.0.0.0 as their ip address. Then they are imediately disconnected. I am running Livingston's 2.0 radius server, although I don't think radius is the problem. Has anyone seen this or have an idea of what might be happening here? - Corey

I brought this up to the mailing list before... only change I was able to make was to put in a static IP entry in my router for the netserver. It still isn't clear to me how that would create any noticable changes, because my 3com hub is switched and (I would assume) would direct packets directly to and from the Total Control unit. -- Number of Occurrences: 15 ISP Entry Router Delay: 1713 ISP Egress Router Delay: 0 -- Again, the hardware configuration includes a 10baseT 3com switched superstack, Cisco 2511, Kentrox csu/dsu, and USR Total Control PRI (netserver 3.3.28). And of course, netmedic. Any suggestions are appreciated... - lv

Once upon a time Corey Piggott shaped the electrons to say... >2> On this same chassie from time to time dial in users once authenticated > recieve 0.0.0.0 as their ip address. Then they are imediately That's a bug from our 3.1.4 ComOS - which is what the USR version is based on. I suspect you need to have a later version of their OS installed. -MZ -- Livingston Enterprises - Chair, Department of Interstitial Affairs Phone: 800-458-9966 510-737-2100 FAX: 510-737-2110 megazone@livingston.com For support requests: support@livingston.com <http://www.livingston.com/> Snail mail: 4464 Willow Road, Pleasanton, CA 94588

> 1> Can anyone tell me what this trap message means: No loop current. > Apparently 10-15% of our connections are failing because of this > problem. I have a Total Control chassie with a netserver(3.3.28) and > quad modems with analog lines. We had this as well and found it to be faulty PSTN wiring. In our case the Telco had crossed up our lines in the street. When one of the lines was busy and a call came in on the other crossed up line, both would drop with this error. The Telco fixed the wiring and we've not seen the 'No Loop Current' again. > 2> On this same chassie from time to time dial in users once authenticated > recieve 0.0.0.0 as their ip address. Then they are imediately > disconnected. Not seen this one... We run Merit Radius here though. This *was* a very old Livingston ComOS bug, and of course the NETServer code is based on ComOS. I'd imagine that USR would have fixed this bug by now though. Regards, Bob Purdon, Technical Manager, Southern Internet Services.

Greetings, Can anyone point to success at setting up a Total Control chassis for DOVBS (56kb or 64Kb). My current understanding is that a digital tone needs to be sent that whether an voice level call if itentified a ISDN digital response is given. A separate number has been assigned in a channelized T-1, but the USR TC won't send the right tone. Thanks, Sandy =============================================================== Mr.Sandy Culver BBN Planet HR Consultant 150 CambridgePark Drive sculver@std.com Cambridge, MA 02140 voice: (508) 623-0942 sculver@bbn.com fax: (508) 623-0082 http://www.bbn.com ===============================================================

Once upon a time Sandy Culver shaped the electrons to say... >Can anyone point to success at setting up a Total Control chassis for DOVBS >(56kb or 64Kb). My current understanding is that a digital tone needs to be AFAIK you can't do ISDN DOSBS over a channelized-t1 line to a TC, only PRI. Least that's what users have told me. The only box I know of that does that automatically is the PM-3. I also believe the Ascend MAX can do it, but it can't auto detect it - it needs to be configured to always do it. Im fuzzy on that - I know it cannot auto-detect like the PM-3 does, but users have told me running it is also possible. -MZ -- Livingston Enterprises - Chair, Department of Interstitial Affairs Phone: 800-458-9966 510-737-2100 FAX: 510-737-2110 megazone@livingston.com For support requests: support@livingston.com <http://www.livingston.com/> Snail mail: 4464 Willow Road, Pleasanton, CA 94588

Hello, I want to know whether TCENH supports fully Livingston's Radius 2.0 spec. Especially, prefix option. i.e. to provide ppp, login service to each user. I heard 3com's AB8000 did that by giving options. For exmaple, userid ...p ..? Is it correct? Does anoyone know about 3com box please give me that privately. If possible, I want to know how many vendors offer ppp service without authentication. From Seoul, Seob

Sorry about that. It's late :-) -Dave

At 06:48 PM 5/10/97 -0700, MegaZone wrote: >Once upon a time Sandy Culver shaped the electrons to say... >>Can anyone point to success at setting up a Total Control chassis for DOVBS >>(56kb or 64Kb). My current understanding is that a digital tone needs to be > >AFAIK you can't do ISDN DOSBS over a channelized-t1 line to a TC, only PRI. >Least that's what users have told me. > >The only box I know of that does that automatically is the PM-3. I also >believe the Ascend MAX can do it, but it can't auto detect it - it needs >to be configured to always do it. Im fuzzy on that - I know it cannot >auto-detect like the PM-3 does, but users have told me running it is also >possible. This is correct. Kevin Smith Updated Service and Support Ascend Communications Resources are now at: http://www.ascend.com/service

Greetings, Can anyone point to success at setting up a Total Control chassis for DOVBS (56kb or 64Kb). My current understanding is that a digital tone needs to be sent whenever a voice level call is made, and whatever is indicated, ISDN digital response is given. A separate number has been assigned in a channelized T-1, but the USR TC won't send the right tone. Any suggestions most welcome. Regards, Sandy =============================================================== Mr.Sandy Culver BBN Planet HR Consultant 150 CambridgePark Drive sculver@std.com Cambridge, MA 02140 voice: (508) 623-0942 sculver@bbn.com fax: (508) 623-0082 http://www.bbn.com ===============================================================

We're getting one TC rack after another, and I feel a little left behind having read only some of what USR ships with the units. The technical descriptive documentation seems to be pretty good, but it's spread out throughout several sources; for example, the Netserver command reference includes an overview of the TC architecture, and the T1 NAC's papers discuss signalling. Is there any more collected, comprehensive documentation on these machines? Many thanks. --- Mark R. Lindsey, mark@datasys.net DSS Online Network Engineer (912) 241-0607, Fax: 241-0190

I'm setting up a chassis in a new location with the dual pri card. Southwestern Bell is asking me if the PRI's I'm ordering need to be set up for "dial pulse, MF, or DTMF". My best guess is that it doesn't matter since we won't be dialing out. -CH

Nevermind, I found that you can use MF or DTMF in call control options in the modem cards. MF is the default, so I'm going with that. These trunk provisioning questionnaires are giving me flashbacks to final exams. How much for Bellcore training? -CH On Tue, 13 May 1997, Charles Hill wrote: > I'm setting up a chassis in a new location with the dual pri card. > Southwestern Bell is asking me if the PRI's I'm ordering need to be set up > for "dial pulse, MF, or DTMF". My best guess is that it doesn't matter > since we won't be dialing out. > > -CH

> >Speaking of PRI cards.....just got one of the ISP bundles that USR is >pushing (in order to get rid of all their old fan trays, ie, the ones that >don't have the integrated fantray)...supposed to have a dual PRI NAC in >it...and sure enough that's how its labeled, but on connecting to >it...guess what? It reports that its a dual T1 NAC....management software >confirms it. > >Is this common for these things to get mixed up and mis-labeled like that? ></delurk> >-- The cards are software upgradable between PRI / T1. I couldn't tell you about the mislabeling though :-(

I'm curious about the serial number. If it says 000944, then it has a dual PRI serial number too. -CH On Wed, 14 May 1997, Francis Fong wrote: > Take a look at the T1 card. If this is a 386 cpu inside, you can upgrade > it to T1 PRI. If this is only 186 CPU, then this T1 card is only capable > for T1 only, as far as I know. > > Rgds, > > Francis > > > On Tue, 13 May 1997, Jeff Mcadams wrote: > > > Thus spake Charles Hill > > >I'm setting up a chassis in a new location with the dual pri card. > > >Southwestern Bell is asking me if the PRI's I'm ordering need to be set up > > >for "dial pulse, MF, or DTMF". My best guess is that it doesn't matter > > >since we won't be dialing out. > > > > <delurk> > > > > Speaking of PRI cards.....just got one of the ISP bundles that USR is > > pushing (in order to get rid of all their old fan trays, ie, the ones that > > don't have the integrated fantray)...supposed to have a dual PRI NAC in > > it...and sure enough that's how its labeled, but on connecting to > > it...guess what? It reports that its a dual T1 NAC....management software > > confirms it. > > > > Is this common for these things to get mixed up and mis-labeled like that? > > </delurk> > > -- > > Jeff McAdams | "OK, we have enough youth... > > IgLou Internet Services | How about a fountain of smart?" > > e-mail: jeffm@iglou.com | - unknown > > > > > >

Thus spake Charles Hill >I'm setting up a chassis in a new location with the dual pri card. >Southwestern Bell is asking me if the PRI's I'm ordering need to be set up >for "dial pulse, MF, or DTMF". My best guess is that it doesn't matter >since we won't be dialing out. <delurk> Speaking of PRI cards.....just got one of the ISP bundles that USR is pushing (in order to get rid of all their old fan trays, ie, the ones that don't have the integrated fantray)...supposed to have a dual PRI NAC in it...and sure enough that's how its labeled, but on connecting to it...guess what? It reports that its a dual T1 NAC....management software confirms it. Is this common for these things to get mixed up and mis-labeled like that? </delurk> -- Jeff McAdams | "OK, we have enough youth... IgLou Internet Services | How about a fountain of smart?" e-mail: jeffm@iglou.com | - unknown

Jeff Mcadams <jeffm@iglou.com> writes: > Speaking of PRI cards.....just got one of the ISP bundles that USR is > pushing (in order to get rid of all their old fan trays, ie, the ones that > don't have the integrated fantray)...supposed to have a dual PRI NAC in > it...and sure enough that's how its labeled, but on connecting to > it...guess what? It reports that its a dual T1 NAC....management software > confirms it. It might just be running the wrong code. You can download either channelized T1 or PRI code to the newer 386 based cards. Depending on what the management tool in question is checking, while the physical card type will always show up as a PRI, the management type may be channelized T1 or PRI depending on the code running on the card. If the card is shown to be running code 4.x.x then this is likely the case, but if it's running a 3.5.x code (or earlier) then it is likely an actual older Dual T1 card that only supports T1 lines. Presuming the former, you can just download the latest PRI code (dp*.nac) and the card will switch to supporting PRI lines. -- David /-----------------------------------------------------------------------\ \ David Bolen \ Internet: db3l@ans.net / | ANS Communications \ Phone: (914) 789-5327 | / 100 Clearbrook Road, Elmsford, NY 10523 \ Fax: (914) 789-5310 \ \-----------------------------------------------------------------------/

We have SA server 4.3 which claims to support proxy TACACS+. Has anyone got this feature to work? We can get the SA server to pass a username to our T+ server (Cisco freeware v2.1) but when the T+ server attempts to read the password from the SA server (acting as a T+ client) it fails with Mon May 12 15:47:33 1997 [16025]: IP: fd 1 eof (connection closed) Mon May 12 15:47:33 1997 [16025]: Error Read -1 bytes from IP, expecting 12 Mon May 12 15:47:33 1997 [16025]: Error IP: Null reply packet when expecting CONTINUE (12 is the number of bytes in the paasword for this user) We are trying to get USR to help but our friendly engineer is off ill - so maybe someone else can help... TIA Malcolm Dobson, Scotland On Line www.scotland.net

On Tue, 13 May 1997 13:09:53 -0400 (EDT), you wrote: > <delurk> >=20 > Speaking of PRI cards.....just got one of the ISP bundles that USR is > pushing (in order to get rid of all their old fan trays, ie, the ones = that > don't have the integrated fantray)...supposed to have a dual PRI NAC in > it...and sure enough that's how its labeled, but on connecting to > it...guess what? It reports that its a dual T1 NAC....management = software > confirms it. >=20 > Is this common for these things to get mixed up and mis-labeled like = that? > </delurk> =46ortunally not :-) But do you remember: Shit happens.... Have a nice day, Holger Koepke

Take a look at the T1 card. If this is a 386 cpu inside, you can upgrade it to T1 PRI. If this is only 186 CPU, then this T1 card is only capable for T1 only, as far as I know. Rgds, Francis On Tue, 13 May 1997, Jeff Mcadams wrote: > Thus spake Charles Hill > >I'm setting up a chassis in a new location with the dual pri card. > >Southwestern Bell is asking me if the PRI's I'm ordering need to be set up > >for "dial pulse, MF, or DTMF". My best guess is that it doesn't matter > >since we won't be dialing out. > > <delurk> > > Speaking of PRI cards.....just got one of the ISP bundles that USR is > pushing (in order to get rid of all their old fan trays, ie, the ones that > don't have the integrated fantray)...supposed to have a dual PRI NAC in > it...and sure enough that's how its labeled, but on connecting to > it...guess what? It reports that its a dual T1 NAC....management software > confirms it. > > Is this common for these things to get mixed up and mis-labeled like that? > </delurk> > -- > Jeff McAdams | "OK, we have enough youth... > IgLou Internet Services | How about a fountain of smart?" > e-mail: jeffm@iglou.com | - unknown > >

Is anyone else having a problem with modem timeout values? I have them set to 60 minutes, but when someone logs in through the modem the value jumps to a very high random number. When the user gets off, the number goes back to 60. I'm using the standard Netserver with 4mb and version 3.3.3 of the software. Thanks Wayne Barber - barberw@tidewater.net Internet System Administrator Coastal Telco Services

Wayne Barber said once upon a time: > >Is anyone else having a problem with modem timeout values? I have >them set to 60 minutes, but when someone logs in through the modem >the value jumps to a very high random number. When the user gets off, >the number goes back to 60. You have to specifically set the Idle-Timeout via RADIUS. The Netserver will not pick it up from the configured settings for each port. I had this problem as well before I specifically set the Idle-Timeout in my RADIUS users file. Be sure to note that this value is in SECONDS!

Does anyone give me a precompiled Merit Radius server for Linux? I use Redhat 4.1 and compiled that without error. But it immediately exited with signal 225. Thanks.

Has anyone been able to get 4.3 to authenticate through the NT user database? My 4.2 server does it just fine, but no luck with 4.3. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Richard Yoakley BlitzNet Online Communications, Inc. (423) 283-7504 =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

Once upon a time Stephen Corbesero shaped the electrons to say... >authentication from the server. I have already checked the obvious >things (routing is working, there is a radius clients entry, ...). > >Do radius clients need a radius server on their local subnet? They shouldn't - run a radius debug on the server to see if you are getting Access-Request packets. And if you are, is a reply being sent. If that checks out you probably want to do a packet trace on the NAS to see if the responses are making it all the way back. -MZ -- Livingston Enterprises - Chair, Department of Interstitial Affairs Phone: 800-458-9966 510-737-2100 FAX: 510-737-2110 megazone@livingston.com For support requests: support@livingston.com <http://www.livingston.com/> Snail mail: 4464 Willow Road, Pleasanton, CA 94588

We have just set up a remote location that is connected back to our main site via frame relay. We already have two usr's back at our main site cheerfully talking with our radius server (running on a solaris box). Unfortunately, the remote usr's cannot seem to get any valid authentication from the server. I have already checked the obvious things (routing is working, there is a radius clients entry, ...). Do radius clients need a radius server on their local subnet? -- Stephen Corbesero This message has been brought to you by electrons. flash@early.com Electrons -- The other charged particle.

Hi, I have a mix of Livingston PM's and TC racks. My radius server is the stock 1.16 from Livingston. It works OK with the PM's, but I'm getting a lot of "Unknown attribute" when the TC rack tries to authenticate. Anything obvious I should be looking for? The radius server is running on Solaris X86 and FreeBSD. Any tip appreciated.

--IMA.Boundary.838128368 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Description: cc:Mail note part >> Is this common for these things to get mixed up and mis-labeled like that? >> </delurk> > >Fortunally not :-) But do you remember: Shit happens.... > >Have a nice day, >Holger Koepke That's true, Holger, but in this case it's planned. Kurtiss ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~ __ __ ~ ~ Kurtiss Johnson | | | | RRRRR ~ ~ Product Manager | | | | ***RR RR ~ ~ US Robotics | \_/ |*** RRRRR ~ ~ kjohnson@usr.com \___/ RR RR ~ ~ See us at www.usr.com! ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ --IMA.Boundary.838128368 Content-Type: text/plain; charset=US-ASCII; name="RFC822 message headers" Content-Transfer-Encoding: 7bit Content-Description: cc:Mail note part Content-Disposition: attachment; filename="RFC822 message headers" Received: from usr.com (mailgate.usr.com) by robogate2.usr.com with SMTP (IMA Internet Exchange 2.02 Enterprise) id 378B0BB0; Tue, 13 May 97 13:19:39 -0500 Received: from mail.xmission.com by usr.com (8.7.5/3.1.090690-US Robotics) id MAA24232; Tue, 13 May 1997 12:58:38 -0500 (CDT) Received: from domo by mail.xmission.com with local (Exim 1.62 #1) id 0wRLHf-0001rX-00; Tue, 13 May 1997 11:22:43 -0600 Received: from mms.de [195.21.213.42] (root) by mail.xmission.com with smtp (Exim 1.62 #1) id 0wRLHc-0001qg-00; Tue, 13 May 1997 11:22:41 -0600 Organization: MMS GmbH Message-ID: <3383a2f4.32163048@mms> References: <E0wRL5G-0003sC-00@iglou1> In-Reply-To: <E0wRL5G-0003sC-00@iglou1> X-Mailer: Forte Agent 1.0/32.390 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Sender: owner-usr-tc@xmission.com Precedence: bulk Reply-To: usr-tc@mail.xmission.com --IMA.Boundary.838128368--

Kurtiss, Or anyone else that knows, Do you know if the Total Control, using ISDN PRI, can be set up to accept data over voice level (DOV or DOSBS) calls? I'd appreciate some definitive help. Regards, Sandy =============================================================== Mr.Sandy Culver BBN Planet HR Consultant 150 CambridgePark Drive sculver@std.com Cambridge, MA 02140 voice: (508) 623-0942 sculver@bbn.com fax: (508) 623-0082 http://www.bbn.com ===============================================================

Is VLSM supported in the following code levels on the Netserver cards? - Netserver 4MB card running version 3.3.3 - Netserver PRI card running version 3.4.23 Does the subnetting for the ethernet (or net0) interface get applied to the assigned address to determine the range of addresses assigned per card, or does the option to set the number of assignable addresses in the Netserver Management software override this? Thanks, Timothy

Laszlo Vecsey said once upon a time: > >To those who experienced the local quake lag phenomena, do the new TCS >2.5.1 quad/netserver updates appear to resolve it? I can't seem to find this release anywhere on the totalservice site.

Is anybody using the Viper xDSL cards yet? Do they work? Are you offering DSL service to your customers?

To those who experienced the local quake lag phenomena, do the new TCS 2.5.1 quad/netserver updates appear to resolve it? - lv

USR have resolved this one - their client does not use encryption - quick fix remove secret from T+ server. However the accounting does not end up in the tacacs accounting file, but in the radius file! On Tue, 13 May 1997, Malcolm Dobson wrote: > We have SA server 4.3 which claims to support proxy TACACS+. > Has anyone got this feature to work? We can get the SA server to > pass a username to our T+ server (Cisco freeware v2.1) but > when the T+ server attempts to read the password from the SA > server (acting as a T+ client) it fails with > > Mon May 12 15:47:33 1997 [16025]: IP: fd 1 eof (connection closed) > Mon May 12 15:47:33 1997 [16025]: Error Read -1 bytes from IP, expecting 12 > Mon May 12 15:47:33 1997 [16025]: Error IP: Null reply packet when expecting CONTINUE > > (12 is the number of bytes in the paasword for this user) > > We are trying to get USR to help but our friendly engineer is off ill - > so maybe someone else can help... > > > TIA > > > Malcolm Dobson, Scotland On Line > www.scotland.net > > >

Last Thursday, US Worst did something to their Salt Lake network to make X2 completely unusable. To make matters worse, our clients need to turn X2 off on their modems in order to make a connection to us at all. We have 14 PRIs, all exhibiting the same behavior, so I'm sure it isn't our problem. First off, does anyone have any theories as to what USW has done? They of course are denying everything, even though we can point to an exact time the change was made. I'm theorizing that we're somehow being multiplexed or compressed where we shouldn't be. Second, is there a quick and dirty way to shut off X2 at the TC chassis? I'd like to disable this so people don't have to adjust their modems to connect to us. -- Pete XMission

If the NIC on the NMC is 10baseT, why does it take forever to do software downloads via TCM? It feels like 57,600 console speed. - lv

On Mon, 19 May 1997, Laszlo Vecsey wrote: > If the NIC on the NMC is 10baseT, why does it take forever to do software > downloads via TCM? It feels like 57,600 console speed. > > - lv Have you tried pinging the NMC? I get 30-50ms using a machine on the same segment!

What a miserable excuse for a company, that UseLessWest. We ordered a channelized T1 here in Casper in January after finding out what kind of line that we would need for x2. Talked extensively to USW and USR. Faxed the specs from USR to USW and finally bought the equipment and ordered a T1. It has been a nightmare. Three weeks after we ordered the line, the lady coding the line called me and asked for details that we worked on for weeks. I gave her info on the spot, had to make decisions about the coding then and there. We made some changes and got it working. Then fought with USR to get the software upgrade. We got it about 5 weeks ago but alas no x2 connections. USR told us there was a conversion between us and the central office which is 2 blocks away from us. Our rep insisted that there was no conversion from analog to digital on their end. We talked to the local engineer 10 days ago and she insisted that there was no way that there was a conversion. We called repair and within 5 mins the guy informs us that there is indeed a conversion between the switch and digital equipment. Our rep now tells us that we need to go to $70 something per line per month from $26 per month AND reinstall the lines. They now quote chapter and verse that we have line side trunks, voice quality instead of what we really need. Ten days ago, they didn't know the difference. I didn't know beans about this stuff when I started this ordeal but learned a lot coming through it. We told them early and often that we needed trunk side lines, USR was pretty clear about that. I'm more angry and frustrated with that bunch of losers at USWorst than I've ever been in the past couple of years of them screwing with us. In the middle of all of this, they moved us into the ISP (Special Group) to handle us businesses with special needs. What a joke, I got better service from the other rep or simply calling the business office number in the front of the phone book. I envy anyone with ANY alternative to USW, you don't know just how lucky you are. Apparently there are two wires from the switch to the digital equipment and that needs to be upgraded to 4. They can do that but only at the higher rates. I filed a complaint with the PUC but I'm not holding my breath for any action there. USW won't even address it with them when I last talked to the state person. USW can hardly talk to me since I filed the complaint. At 07:03 PM 5/19/97 -0600, you wrote: >Last Thursday, US Worst did something to their Salt Lake network to make X2 >completely unusable. To make matters worse, our clients need to turn X2 >off on their modems in order to make a connection to us at all. > >We have 14 PRIs, all exhibiting the same behavior, so I'm sure it isn't our >problem. First off, does anyone have any theories as to what USW has done? >They of course are denying everything, even though we can point to an exact >time the change was made. I'm theorizing that we're somehow being >multiplexed or compressed where we shouldn't be. > >Second, is there a quick and dirty way to shut off X2 at the TC chassis? >I'd like to disable this so people don't have to adjust their modems to >connect to us. > >-- >Pete >XMission > > Thanks, Greg Coffey, CoffeyNet ~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~ 142 S. Center St. Local Internet for Casper, Rawlins, Douglas, Casper, WY 82601 Wheatland, Pinedale, Lander & Lusk, WY http://www.coffey.com (307) 234-5443 ~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~

I'd PAY them to reinstall the lines if they could provide the service at the rates quoted in January. The install side of this is not the issue here, its the monthly rate. I can't believe that adding two more wires to the mix justifies that kind of fare increase (from $26 to $71 per month). I would not have invested in this technology if I had know what the REAL rate would be. I'm out the money and stuck with the equipment. Their ISDN rates are outrageous too. I'll bet they haven't sold more than a handful of those lines in the state. People tell me that it doesn't cost anymore to install or operate than a regular line but USW charges a ton more for both. The Public Utilities Commission is opening a hearing on these issues in the near future. Perhaps we'll see some adjustment in the rates but what we really need is an alternative phone company. >So don't pay them....you told them trunk side, they gave you line >side...they didn't fulfill their end of the contract. :) Tell 'em they >won't get any money until the lines are installed as you want them...that >should get them moving. :) >-- >Jeff McAdams | "OK, we have enough youth... >IgLou Internet Services | How about a fountain of smart?" >e-mail: jeffm@iglou.com | - unknown > > Thanks, Greg Coffey, CoffeyNet ~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~ 142 S. Center St. Local Internet for Casper, Rawlins, Douglas, Casper, WY 82601 Wheatland, Pinedale, Lander & Lusk, WY http://www.coffey.com (307) 234-5443 ~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~

On Tue, 20 May 1997, Greg Coffey wrote: > I'd PAY them to reinstall the lines if they could provide the service at the > rates quoted in January. The install side of this is not the issue here, > its the monthly rate. I can't believe that adding two more wires to the mix > justifies that kind of fare increase (from $26 to $71 per month). I would > not have invested in this technology if I had know what the REAL rate would > be. I'm out the money and stuck with the equipment. Their ISDN rates are > outrageous too. I'll bet they haven't sold more than a handful of those > Make sure the PUC knows how you feel, around here a huge campaign was mounted against them, and the PUC finally ordered USW to drop their flat rate ISDN prices from $185 to $68 per month... Jason Kohles -- System Administrator -- XMission Internet Access robobob@xmission.com (at work) robobob@mindwell.com (at play) "If you can smile when things go wrong, you have someone in mind to blame."

Thus spake Greg Coffey >They now quote chapter and verse >that we have line side trunks, voice quality instead of what we really need. [snip] >We told them early and often that we needed trunk side lines, USR was pretty >clear about that. So don't pay them....you told them trunk side, they gave you line side...they didn't fulfill their end of the contract. :) Tell 'em they won't get any money until the lines are installed as you want them...that should get them moving. :) -- Jeff McAdams | "OK, we have enough youth... IgLou Internet Services | How about a fountain of smart?" e-mail: jeffm@iglou.com | - unknown

Hello, Because of a contract I can't use the following system (so it is is for sale): US Robotics Total Control System: - TC Enterprise Network HUB (220VAC) - 2x Power supply, 1x reserve Power supply - Fan Tray - Ethernet Network Management Card (SNMP) - TC Dual PRI Card - TC Netserver Card (60 channel with NIC) - 2x Quad Modem Card Total Package: New Price fl 60.000,00 Sale Price fl 40.000,00 The system is never used and inclusive the full installation and upgrade to X2 code by US Robotics. Joost procompNET (http://www.procomp.NET) Toernooiveld 126 - 6525 EC Nijmegen tel. 024-3500437 - fax. 024-3500613 Netherlands

Hello, Because of a contract I can't use the following system (so it is is for sale): US Robotics Total Control System: - TC Enterprise Network HUB (220VAC) - 2x Power supply, 1x reserve Power supply - Fan Tray - Ethernet Network Management Card (SNMP) - TC Dual PRI Card - TC Netserver Card (60 channel with NIC) - 2x Quad Modem Card Total Package: New Price fl 60.000,00 Sale Price fl 40.000,00 The system is never used and inclusive the full installation and upgrade to X2 code by US Robotics. Joost procompNET (http://www.procomp.NET) Toernooiveld 126 - 6525 EC Nijmegen tel. 024-3500437 - fax. 024-3500613 Netherlands ********************************************************************** Please visit list home page at http://www.neystadt.org/winnt/winnt.htm You will find subscription / unsubscription instructions there.

Hi, I also have a mix of these two product. We use Radius 2.0 and discover that there are many '0' in session-time in Total control record. If we try to upgrade Netserver Card 3.3.3, the radius deamon will automatically stop by itself. Can anyone have some suggestion? Kenneth Tse Jaye Mathisen wrote: > > Hi, I have a mix of Livingston PM's and TC racks. My radius server is > the stock 1.16 from Livingston. > > It works OK with the PM's, but I'm getting a lot of "Unknown attribute" > when the TC rack tries to authenticate. > > Anything obvious I should be looking for? The radius server is running on > Solaris X86 and FreeBSD. > > Any tip appreciated.

Hi, I also have a mix of these two product. We use Radius 2.0 and discover that there are many '0' in session-time in Total control record. If we try to upgrade Netserver Card 3.3.3, the radius deamon will automatically stop by itself. Can anyone have some suggestion? Kenneth Tse Jaye Mathisen wrote: > > Hi, I have a mix of Livingston PM's and TC racks. My radius server is > the stock 1.16 from Livingston. > > It works OK with the PM's, but I'm getting a lot of "Unknown attribute" > when the TC rack tries to authenticate. > > Anything obvious I should be looking for? The radius server is running on > Solaris X86 and FreeBSD. > > Any tip appreciated.

Thus spake Greg Coffey >I'd PAY them to reinstall the lines if they could provide the service at the >rates quoted in January. The install side of this is not the issue here, >its the monthly rate. Yeah, good point....it will cost you more, but since you specifically told them trunk side, I wouldn't pay them for the current service that you're (at least a little bit) using. >I can't believe that adding two more wires to the mix >justifies that kind of fare increase (from $26 to $71 per month). I would >not have invested in this technology if I had know what the REAL rate would >be. I'm out the money and stuck with the equipment. Their ISDN rates are >outrageous too. I'll bet they haven't sold more than a handful of those >lines in the state. People tell me that it doesn't cost anymore to install >or operate than a regular line but USW charges a ton more for both. The >Public Utilities Commission is opening a hearing on these issues in the near >future. Perhaps we'll see some adjustment in the rates but what we really >need is an alternative phone company. Indeed...here in BellSouth land, PRI is actually a bit cheaper than T1's....though not significantly. We're getting alternate providers around here....just started happening in the past few months...I can't wait personally....our phone bill will drop from about $16,000/month to about half that, maybe less. -- Jeff McAdams | "OK, we have enough youth... IgLou Internet Services | How about a fountain of smart?" e-mail: jeffm@iglou.com | - unknown

Laszlo Vecsey said once upon a time: > >If the NIC on the NMC is 10baseT, why does it take forever to do software >downloads via TCM? It feels like 57,600 console speed. I think the actual download is fairly quick. Its writing it into the NVRAM which takes all the time.

Thus spake Jason S Kohles >Make sure the PUC knows how you feel, around here a huge campaign was >mounted against them, and the PUC finally ordered USW to drop their flat >rate ISDN prices from $185 to $68 per month... BellSouth just within the past few months did away with flat-rate ISDN completely...at least here in Kentucky they did. :/ -- Jeff McAdams | "OK, we have enough youth... IgLou Internet Services | How about a fountain of smart?" e-mail: jeffm@iglou.com | - unknown

On Tue, 20 May 1997, Jason S Kohles wrote: > Make sure the PUC knows how you feel, around here a huge campaign was > mounted against them, and the PUC finally ordered USW to drop their flat > rate ISDN prices from $185 to $68 per month... Can the Utah PUC send the Minnesota PUC a few new employees? We're still paying US West $185 a month for flat rate ISDN. There was already a huge battle waged to get US West to offer ISDN to any Minneapolis/St. Paul location. Previously, they would charge for back haul if your CO did not have an ISDN capable switch. Brian

On Tuesday, May 20, 1997 12:52 PM, Kenneth Tse [SMTP:kentse@aicom.com] wrote: > Hi, I also have a mix of these two product. We use Radius 2.0 and > discover that there are many '0' in session-time in Total control > record. If we try to upgrade Netserver Card 3.3.3, the radius > deamon will automatically stop by itself. Can anyone have some > suggestion? > > Kenneth Tse > > Jaye Mathisen wrote: > > > > Hi, I have a mix of Livingston PM's and TC racks. My radius server is > > the stock 1.16 from Livingston. > > > > It works OK with the PM's, but I'm getting a lot of "Unknown attribute" > > when the TC rack tries to authenticate. > > > > Anything obvious I should be looking for? The radius server is running on > > Solaris X86 and FreeBSD. > > > > Any tip appreciated. We just upgraded to 2.0 of the Livingston Radius for both x86 and Sparc Solaris. We made several changes to the dictionary files in order to gain to most information from USR Netservers, Ascend Max and Livingston NAS. I am not aware of any problems like 0 for the session times however. We are running 3.4.x on the Netservers, 5.0Ap1 on the MAX and 3.3.x on the Livingston boxes. For Solaris, please do not forget to kill the nscd daemon or your log files will overflow with messages. We have seem many "killing child processes" messages on a very low utilized server and are tracking that down now. The dictionary file enhancements were: ATTRIBUTE Acct-Input-Packets 47 integer ATTRIBUTE Acct-Output-Packets 48 integer in the users file we made the following additions: # Username alex runs PPP if alex uses PAP to authenticate and port autodetects DEFAULT Auth-Type = System, Framed-Protocol = PPP Service-Type = Framed-User, Framed-IP-Address = 255.255.255.254, Framed-IP-Netmask = 255.255.255.255, Framed-Routing = None, Framed-Compression = Van-Jacobson-TCP-IP, Idle-Timeout = 1200, Session-Timeout = 28800, Port-Limit = 1, Framed-MTU = 1500 # for all other users, PPP is assumed. ISDN 128k Dialers will need the Port-Limit = 2 parameter. DEFAULT Auth-Type = System Service-Type = Framed-User, Framed-Protocol = PPP, Framed-IP-Address = 255.255.255.254, Framed-IP-Netmask = 255.255.255.255, Framed-Routing = None, Framed-Compression = Van-Jacobson-TCP-IP, Idle-Timeout = 1200, Session-Timeout = 28800, Port-Limit = 1, Framed-MTU = 1500 We also changed the username to be can insensitive due to the problems we have seen in the past by editing the radius.c file a bit. If anyone wants it, just ask the list and we will publicize it. Marshall Morgan President Internet Doorway, Inc. http://www.netdoor.com 601.969.1434 | 800.952.1570 | 601.969.3838 Fax

Pete Ashdown <pashdown@xmission.com> writes: > Second, is there a quick and dirty way to shut off X2 at the TC chassis? > I'd like to disable this so people don't have to adjust their modems to > connect to us. Set the "mdmScX2DisableServer" MIB object to "disable" for each modem (default="enable"). This is part of the "signal control" group for each modem and should be selectable in TCM with a current MIB file. This is for NMC 4.3.4 - if you're part of any of the later betas the variable changed name to "mdmScX2Server" (which is more reasonable given how "enable"/"disable" work for the variable). -- David /-----------------------------------------------------------------------\ \ David Bolen \ Internet: db3l@ans.net / | ANS Communications \ Phone: (914) 789-5327 | / 100 Clearbrook Road, Elmsford, NY 10523 \ Fax: (914) 789-5310 \ \-----------------------------------------------------------------------/

--IMA.Boundary.326152468 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Description: cc:Mail note part On 5-16-97 @ 12:13pm, mrcpe@cdsnet.net wrote: > >Hi, I have a mix of Livingston PM's and TC racks. My radius server is >the stock 1.16 from Livingston. > >It works OK with the PM's, but I'm getting a lot of "Unknown attribute" >when the TC rack tries to authenticate. > >Anything obvious I should be looking for? The radius server is running on >Solaris X86 and FreeBSD. > >Any tip appreciated. > Jaye, USR uses a number of vendor-specific attributes (attribute #26, with offsets) that deliver a number of proprietary extensions. In addition, if you have turned ON the RADIUS accounting of the NMC, then you are getting a number of additional vendor-specific attributes that reflects modem information not normally logged anywhere. This is what is causing the "unknown attribute" message, and unless you're looking for something special, this isn't a problem. Kurtiss ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~ __ __ ~ ~ Kurtiss Johnson | | | | RRRRR ~ ~ Product Manager | | | | ***RR RR ~ ~ US Robotics | \_/ |*** RRRRR ~ ~ kjohnson@usr.com \___/ RR RR ~ ~ See us at www.usr.com! ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ --IMA.Boundary.326152468 Content-Type: text/plain; charset=US-ASCII; name="RFC822 message headers" Content-Transfer-Encoding: 7bit Content-Description: cc:Mail note part Content-Disposition: attachment; filename="RFC822 message headers" Received: from usr.com (mailgate.usr.com) by robogate2.usr.com with SMTP (IMA Internet Exchange 2.02 Enterprise) id 37CE1B20; Fri, 16 May 97 17:37:38 -0500 Received: from mail.xmission.com by usr.com (8.7.5/3.1.090690-US Robotics) id OAA15972; Fri, 16 May 1997 14:52:20 -0500 (CDT) Received: from domo by mail.xmission.com with local (Exim 1.62 #1) id 0wSSS1-0006Td-00; Fri, 16 May 1997 13:14:01 -0600 Received: from mail.cdsnet.net [204.118.244.5] by mail.xmission.com with smtp (Exim 1.62 #1) id 0wSSRs-0006TI-00; Fri, 16 May 1997 13:13:52 -0600 Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by mail.cdsnet.net (8.8.5/8.7.3) with SMTP id MAA27124 for <usr-tc@xmission.com>; Fri, 16 May 1997 12:13:40 -0700 (PDT) Message-ID: <Pine.NEB.3.95.970516121200.1055j-100000@mail.cdsnet.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-usr-tc@xmission.com Precedence: bulk Reply-To: usr-tc@mail.xmission.com --IMA.Boundary.326152468--

Hello, Because of a contract I can't use the following system (so it is is for sale): US Robotics Total Control System: - TC Enterprise Network HUB (220VAC) - 2x Power supply, 1x reserve Power supply - Fan Tray - Ethernet Network Management Card (SNMP) - TC Dual PRI Card - TC Netserver Card (60 channel with NIC) - 2x Quad Modem Card Total Package: New Price fl 60.000,00 Sale Price fl 40.000,00 = 21.000 US$ (exclusive shipment) The system is never used and inclusive the full installation and upgrade to X2 code by US Robotics. Joost procompNET (http://www.procomp.NET) Toernooiveld 126 - 6525 EC Nijmegen tel. 024-3500437 - fax. 024-3500613 Netherlands

On Wed, 21 May 1997 kjohnson@usr.com wrote: > On 5-16-97 @ 12:13pm, mrcpe@cdsnet.net wrote: > > > >Hi, I have a mix of Livingston PM's and TC racks. My radius server is > >the stock 1.16 from Livingston. > > > >It works OK with the PM's, but I'm getting a lot of "Unknown attribute" > >when the TC rack tries to authenticate. > > > >Anything obvious I should be looking for? The radius server is running on > >Solaris X86 and FreeBSD. > > > >Any tip appreciated. > > > Jaye, > > USR uses a number of vendor-specific attributes (attribute #26, with offsets) > that deliver a number of proprietary extensions. In addition, if you have > turned ON the RADIUS accounting of the NMC, then you are getting a number of > additional vendor-specific attributes that reflects modem information not > normally logged anywhere. > > This is what is causing the "unknown attribute" message, and unless you're > looking for something special, this isn't a problem. So how can I see all these Attributes logged in the detail file, if I am using Merit radius? Does USR have a dictionary file made that I can just plug in? Brian > > Kurtiss > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > ~ __ __ ~ > ~ Kurtiss Johnson | | | | RRRRR ~ > ~ Product Manager | | | | ***RR RR ~ > ~ US Robotics | \_/ |*** RRRRR ~ > ~ kjohnson@usr.com \___/ RR RR ~ > ~ See us at www.usr.com! ~ > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > > oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo Brian Feeny oo ShreveNet, Inc. oo Phone: (318) 222-2NET UNIX Administrator oo 333 Texas St #619 oo FAX: (318) 221-6612 signal@shreve.net oo Shreveport, LA 71101 oo http://www.shreve.net/ oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo

On Wed, 21 May 1997 kjohnson@usr.com wrote: > On 5-16-97 @ 12:13pm, mrcpe@cdsnet.net wrote: > > > >Hi, I have a mix of Livingston PM's and TC racks. My radius server is > >the stock 1.16 from Livingston. > > > >It works OK with the PM's, but I'm getting a lot of "Unknown attribute" > >when the TC rack tries to authenticate. > > > >Anything obvious I should be looking for? The radius server is running on > >Solaris X86 and FreeBSD. > > > >Any tip appreciated. > > > Jaye, > > USR uses a number of vendor-specific attributes (attribute #26, with offsets) > that deliver a number of proprietary extensions. In addition, if you have > turned ON the RADIUS accounting of the NMC, then you are getting a number of > additional vendor-specific attributes that reflects modem information not > normally logged anywhere. > > This is what is causing the "unknown attribute" message, and unless you're > looking for something special, this isn't a problem. So how can I see all these Attributes logged in the detail file, if I am using Merit radius? Does USR have a dictionary file made that I can just plug in? Brian > > Kurtiss > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > ~ __ __ ~ > ~ Kurtiss Johnson | | | | RRRRR ~ > ~ Product Manager | | | | ***RR RR ~ > ~ US Robotics | \_/ |*** RRRRR ~ > ~ kjohnson@usr.com \___/ RR RR ~ > ~ See us at www.usr.com! ~ > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > > oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo Brian Feeny oo ShreveNet, Inc. oo Phone: (318) 222-2NET UNIX Administrator oo 333 Texas St #619 oo FAX: (318) 221-6612 signal@shreve.net oo Shreveport, LA 71101 oo http://www.shreve.net/ oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo

We have Computone terminal servers here which we are able to finger to find out who is logged on. We now have a Total Control box, and I am looking for a way to do the same thing. Is there any way to do this? I've looked a bit and haven't found anything yet. Thanks, Jim MacKenzie System Administrator ErieNet, Inc.

> > We have 14 PRIs, all exhibiting the same behavior, so I'm sure it isn't our > problem. First off, does anyone have any theories as to what USW has done? > They of course are denying everything, even though we can point to an exact > time the change was made. I'm theorizing that we're somehow being > multiplexed or compressed where we shouldn't be. > Why not hang a sportster or courier external modem off the back iof one of your quad cards (assuming you have the analog/digital quad cards) and manually dial out to the 800 line test # at usr and see for yourself if they are still X2 capible. Of course your PRI circuits would have to be configured to allow dialing out... Adam Wills Global 2000 Communications Director of Networking Systems 1840 Western Ave. sysadmin@global2000.net Albany, NY, 12203 http://www.global2000.net (518) 452-1465

On Thu, 22 May 1997, James MacKenzie wrote: > We have Computone terminal servers here which we are able to finger to > find out who is logged on. We now have a Total Control box, and I am > looking for a way to do the same thing. Is there any way to do this? > I've looked a bit and haven't found anything yet. > Thanks, > Jim MacKenzie I use pmwho. I had to modify the source just slightly. . . take an extra space out at the end of the expected prompt. It still works with both the portmasters and the netservers without the space. -CH

I've heard through the grapevine that USR has developed this software package for use with FreeBSD. Is this true? There are a load of BSDI users out here that could make use of such an animal. Since BSDI & USR are working together on the EdgeServer, it makes sense that USR would do this if they haven't. Comments? Webmaster http://www.wingnet.net

On 5/19/97 at 11:42am, Pete Ashdown wrote: >Laszlo Vecsey said once upon a time: >> >>To those who experienced the local quake lag phenomena, do the new TCS >>2.5.1 quad/netserver updates appear to resolve it? > >I can't seem to find this release anywhere on the totalservice site. > Laszlo and Pete (and others), This system release won't be found on the TotalService site until it's released. You should see it on the TotalService site in two to three weeks. There are no specific fixes for the "quakelag" issue, although we are working on it. So far, we've not had any luck reproducing the problem in-house (although I'm getting pretty good at Quake now :) You may see this problem diminish in the new release, as we have increased the NETServer's total system throughput by another 30% over TCSv2.5 and the per-port throughput capacity by about 10-15%. Kurtiss ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~ __ __ ~ ~ Kurtiss Johnson | | | | RRRRR ~ ~ Product Manager | | | | ***RR RR ~ ~ US Robotics | \_/ |*** RRRRR ~ ~ kjohnson@usr.com \___/ RR RR ~ ~ See us at www.usr.com! ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

I have two of the Total Control units but both have the 186 processor on the T1 cards so they can't be upgraded to PRI. Any of you have any extra dual cards that can do PRI that you want to sell or swap for my old ones and cash? This gives me some alternatives with USeLessWest and can drive down my monthly costs. Thanks, Greg Coffey, CoffeyNet ~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~ 142 S. Center St. 307-234-5443 307-234-5446 Fax Casper, WY 82601 Local Internet for Casper, Rawlins, Douglas, http://www.coffey.com Wheatland, Pinedale, Lander & Lusk, WY ~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~

On Thu, 22 May 1997, James MacKenzie wrote: > We have Computone terminal servers here which we are able to finger to > find out who is logged on. We now have a Total Control box, and I am > looking for a way to do the same thing. Is there any way to do this? > I've looked a bit and haven't found anything yet. > tcview will do what you are looking for, but from a unix prompt. If you really need finger service (to reach other platforms, like Win95) what I do is simply have it invoked as a wrapper from cfingerd running on a unix server. http://master.internexus.net/Projects/tcview.tar.gz Comments/suggestions are always appreciated. - lv

I set up four tc with four netserver pri (3.3.28). Two of each share same subnet. Say Tc1,Tc2's network is 1.1.1.0 and Tc3,Tc4 is 1.1.2.0. (netmask is 255.255.255.0) i.e. Tc 1 Netserver 1.1.1.2 Assigned ip adress starts 1.1.1.10. limit is 40 Tc 2 Netserver 1.1.1.4 " " 1.1.1.60 " Tc 3 Netserver 1.1.2.2 " " 1.1.2.10 " Tc 4 Netserver 1.1.2.4 " " 1.1.2.60 " Then these boxes connect router with ip. i.e. router just has each ip and there is no routing protocol between Netserver and router.(net0 routing off) Anyway I can ping each Netserver from router vice versa. So I wanted to ping via dial-up user. I could ping anywhere when connected via Tc1, Tc3 dialup user but couldn't ping router ,of course, other subnet when connected via Tc2, Tc4 but I could ping Tc2 and Tc4. Does anyone know why? Do I have to use RIP or something in router and Netserver? Thanks you. From Seoul, Seob

I think it is useful to provide this info. I set up network dailup user like this. set user ppp address assigned. netmask 255.255.255.0 And each Netserver has gateway. default gateway 1.1.1.1 for 1.1.1.2, 1.1.1.3 (Netservers) " " 1.1.2.1 for 1.1.2.2, 1.1.2,3 " More specific, say when dialup user's ip address was 1.1.1.20, he could ping 1.1.1.0, 1.1.2.0 but ip address was 1.1.1.70 or 1.1.2.70 he couldn't ping router(1.1.1.1 or 1.1.2.1) but could ping his netserver. i.e. 1.1.1.2 and 1.1.1.3. From Seoul, Seob

We have an old Netserver with 1 meg of RAM soldered on the board; this sounds antique compared to the 4 meg that seem to be the standard for quite some time now, and I wanted to be sure that I could just add what the rest of you are adding: I'd like to plug in a 4 or 8-meg non-parity 72-pin SIMM. Will that work? (Kurtiss?) Thanks. --- Mark R. Lindsey, mark@datasys.net DSS Online Network Engineer (912) 241-0607, Fax: 241-0190

I wrote in a while ago about the ability to check to see if someone is logged in using X2 or not from the unix command line. Someone replied with a scotty script that they were using, unfortunately, I seem to have lost that mail file. Is there an archive of the list, and if not could someone please repost the script that they use to check the users on a TCHub and whether or not they're using X2? Thanks in advance! Joe

Once upon a time Laszlo Vecsey shaped the electrons to say... >What needs to be configured in radius for a dialup user to establish a >Multilink PPP session? I'm using unix merit radius, and the only related Nothing. There is nothing at all special in RADIUS for this. It should be on by default. > ATTRIBUTE Ascend-Multilink-ID 187 integer > ATTRIBUTE Ascend-Num-In-Multilink 188 integer These are Ascend specific attributes for some extensions they alone created. -MZ -- Livingston Enterprises - Chair, Department of Interstitial Affairs Phone: 800-458-9966 510-737-2100 FAX: 510-737-2110 megazone@livingston.com For support requests: support@livingston.com <http://www.livingston.com/> Snail mail: 4464 Willow Road, Pleasanton, CA 94588

Once upon a time Laszlo Vecsey shaped the electrons to say... >What about restricting it by default for most users? Port-Limit = 1 The RFC provides for setting the max ports on an MP session with that attribute. Setting it to one effectively disables MP. -MZ -- Livingston Enterprises - Chair, Department of Interstitial Affairs Phone: 800-458-9966 510-737-2100 FAX: 510-737-2110 megazone@livingston.com For support requests: support@livingston.com <http://www.livingston.com/> Snail mail: 4464 Willow Road, Pleasanton, CA 94588

What needs to be configured in radius for a dialup user to establish a Multilink PPP session? I'm using unix merit radius, and the only related items I found in the dictionary are these: ATTRIBUTE Ascend-Multilink-ID 187 integer ATTRIBUTE Ascend-Num-In-Multilink 188 integer - lv

On Tuesday, May 27, 1997 5:42 PM, Laszlo Vecsey [SMTP:master@internexus.net] wrote: > What about restricting it by default for most users? > > On Tue, 27 May 1997, MegaZone wrote: > > > Once upon a time Laszlo Vecsey shaped the electrons to say... > > >What needs to be configured in radius for a dialup user to establish a > > >Multilink PPP session? I'm using unix merit radius, and the only related > > > > Nothing. There is nothing at all special in RADIUS for this. It should > > be on by default. > > > > > ATTRIBUTE Ascend-Multilink-ID 187 integer > > > ATTRIBUTE Ascend-Num-In-Multilink 188 integer > > > > These are Ascend specific attributes for some extensions they alone > > created. > > > > -MZ > > -- > > Livingston Enterprises - Chair, Department of Interstitial Affairs > > Phone: 800-458-9966 510-737-2100 FAX: 510-737-2110 megazone@livingston.com > > For support requests: support@livingston.com <http://www.livingston.com/> > > Snail mail: 4464 Willow Road, Pleasanton, CA 94588 > > > > > For RADIUS 2.0, somethink that includes the Port-Limit directive. DEFAULT Auth-Type = System Service-Type = Framed-User, Framed-Protocol = PPP, Framed-IP-Address = 255.255.255.254, Framed-IP-Netmask = 255.255.255.255, Framed-Routing = None, Framed-Compression = Van-Jacobson-TCP-IP, Idle-Timeout = 1200, Session-Timeout = 28800, Port-Limit = 1, Framed-MTU = 1500 --- Marshall http://www.netdoor.com

What about restricting it by default for most users? On Tue, 27 May 1997, MegaZone wrote: > Once upon a time Laszlo Vecsey shaped the electrons to say... > >What needs to be configured in radius for a dialup user to establish a > >Multilink PPP session? I'm using unix merit radius, and the only related > > Nothing. There is nothing at all special in RADIUS for this. It should > be on by default. > > > ATTRIBUTE Ascend-Multilink-ID 187 integer > > ATTRIBUTE Ascend-Num-In-Multilink 188 integer > > These are Ascend specific attributes for some extensions they alone > created. > > -MZ > -- > Livingston Enterprises - Chair, Department of Interstitial Affairs > Phone: 800-458-9966 510-737-2100 FAX: 510-737-2110 megazone@livingston.com > For support requests: support@livingston.com <http://www.livingston.com/> > Snail mail: 4464 Willow Road, Pleasanton, CA 94588 > >

Ok, that makes sense. I was using Port-Limit for Ascend Pipelines calling in with ISDN, and was under the impression that it would be different for regular Win95 people. Thanks, - lv On Tue, 27 May 1997, MegaZone wrote: > Once upon a time Laszlo Vecsey shaped the electrons to say... > >What about restricting it by default for most users? > > Port-Limit = 1 > > The RFC provides for setting the max ports on an MP session with that > attribute. Setting it to one effectively disables MP. > > -MZ > -- > Livingston Enterprises - Chair, Department of Interstitial Affairs > Phone: 800-458-9966 510-737-2100 FAX: 510-737-2110 megazone@livingston.com > For support requests: support@livingston.com <http://www.livingston.com/> > Snail mail: 4464 Willow Road, Pleasanton, CA 94588 > >

One of my client wanted to know this. Made login user like this way. set user xx password xx set user host 1.1.1.1 (Netserver's ip) Then he wanted to login Netserver with account xx. i.e. he tried to telnet Netserver via LAN not dialup(modem). I think our Netserver doesn't provide this thing. Do you think so? And he also wanted to know Radius in the same situation. i.e. he wanted Radius to authenticate user in LAN not dialup(modem). I think Radius only for authenticating dialup user not LAN user. But he believed Radius also supported that thing because 3com's AB8000 could do that. Does anyone give me clear definition on Radius? I don't think so Radius also has a feature to authenticate LAN user. And if only made user and password in Users file(Radius) and there was no other statement suah as Framed-User or Login-User, bler, bler... what will happen? I mean Netserver has default setting( maybe he can log as login user or network user) or reject when Radius only authenticate user id and password and doesn't provide user info to Netserver. Weird question, though. :)

--IMA.Boundary.928038468 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Description: cc:Mail note part Mark, You're right. It sounds old. It should have a 486sx on the board, right? It's been a while since I've seen one of those. You won't have any problems with a memory upgrade, but you should use either a 4Meg or a 16Meg SIMM. We've seen compatability problems with several vendors' 8Meg SIMMs. The other thing you may want to consider is to get the board back to USR Service so that we can apply some of the hardware ECOs that have been made over the years and incorporated into the design. You'll have fewer problems as you increase the throughput on the board. Kurtiss ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~ __ __ ~ ~ Kurtiss Johnson | | | | RRRRR ~ ~ Product Manager | | | | ***RR RR ~ ~ US Robotics | \_/ |*** RRRRR ~ ~ kjohnson@usr.com \___/ RR RR ~ ~ See us at www.usr.com! ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ______________________________ Reply Separator _________________________________ Author: mark@vielle.datasys.net (Mark R. Lindsey) at Internet We have an old Netserver with 1 meg of RAM soldered on the board; this sounds antique compared to the 4 meg that seem to be the standard for quite some time now, and I wanted to be sure that I could just add what the rest of you are adding: I'd like to plug in a 4 or 8-meg non-parity 72-pin SIMM. Will that work? (Kurtiss?) Thanks. --- Mark R. Lindsey, mark@datasys.net DSS Online Network Engineer (912) 241-0607, Fax: 241-0190 --IMA.Boundary.928038468 Content-Type: text/plain; charset=US-ASCII; name="RFC822 message headers" Content-Transfer-Encoding: 7bit Content-Description: cc:Mail note part Content-Disposition: attachment; filename="RFC822 message headers" Received: from usr.com (mailgate.usr.com) by robogate2.usr.com with SMTP (IMA Internet Exchange 2.02 Enterprise) id 38ADA000; Tue, 27 May 97 07:56:32 -0500 Received: from mail.xmission.com by usr.com (8.7.5/3.1.090690-US Robotics) id HAA02904; Tue, 27 May 1997 07:35:15 -0500 (CDT) Received: from domo by mail.xmission.com with local (Exim 1.62 #1) id 0wWLJY-0000sM-00; Tue, 27 May 1997 06:25:20 -0600 Received: from vielle.datasys.net [204.252.164.2] (mark) by mail.xmission.com with smtp (Exim 1.62 #1) id 0wWLJU-0000rz-00; Tue, 27 May 1997 06:25:17 -0600 Received: (from mark@localhost) by vielle.datasys.net (8.6.11/8.6.9.MRL) id IAA13801 for usr-tc@mail.xmission.com; Tue, 27 May 1997 08:32:12 -0400 Message-Id: <199705271232.IAA13801@vielle.datasys.net> X-Pgp-Fingerprint: DB 6B 9E 9F B5 F2 9B 6D A0 BE D8 10 6B 22 8F 06 X-Tom-Swiftie: Eating Uranium makes me feel funny, said Tom glowingly. X-Mailer: Mail User's Shell (7.2.6 1995-03-03) Sender: owner-usr-tc@xmission.com Precedence: bulk Reply-To: usr-tc@mail.xmission.com --IMA.Boundary.928038468--

--IMA.Boundary.138038468 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Description: cc:Mail note part Seob, On each of the NETServers, issue the command "set proxyarp on". It sounds to me like the requests are making their way to the network but can't find a return path for users connected to two of the NETServers. This command tells the NETServer to repond with it's own MAC address when a host on the segment tries to find the return path. Kurtiss ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~ __ __ ~ ~ Kurtiss Johnson | | | | RRRRR ~ ~ Product Manager | | | | ***RR RR ~ ~ US Robotics | \_/ |*** RRRRR ~ ~ kjohnson@usr.com \___/ RR RR ~ ~ See us at www.usr.com! ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ______________________________ Reply Separator _________________________________ Author: JungSeob Lee <audience@soback.kornet.nm.kr> at Internet I set up four tc with four netserver pri (3.3.28). Two of each share same subnet. Say Tc1,Tc2's network is 1.1.1.0 and Tc3,Tc4 is 1.1.2.0. (netmask is 255.255.255.0) i.e. Tc 1 Netserver 1.1.1.2 Assigned ip adress starts 1.1.1.10. limit is 40 Tc 2 Netserver 1.1.1.4 " " 1.1.1.60 " Tc 3 Netserver 1.1.2.2 " " 1.1.2.10 " Tc 4 Netserver 1.1.2.4 " " 1.1.2.60 " Then these boxes connect router with ip. i.e. router just has each ip and there is no routing protocol between Netserver and router.(net0 routing off) Anyway I can ping each Netserver from router vice versa. So I wanted to ping via dial-up user. I could ping anywhere when connected via Tc1, Tc3 dialup user but couldn't ping router ,of course, other subnet when connected via Tc2, Tc4 but I could ping Tc2 and Tc4. Does anyone know why? Do I have to use RIP or something in router and Netserver? Thanks you. From Seoul, Seob --IMA.Boundary.138038468 Content-Type: text/plain; charset=US-ASCII; name="RFC822 message headers" Content-Transfer-Encoding: 7bit Content-Description: cc:Mail note part Content-Disposition: attachment; filename="RFC822 message headers" Received: from usr.com (mailgate.usr.com) by robogate2.usr.com with SMTP (IMA Internet Exchange 2.02 Enterprise) id 38A8FA60; Tue, 27 May 97 02:39:18 -0500 Received: from mail.xmission.com by usr.com (8.7.5/3.1.090690-US Robotics) id CAA27917; Tue, 27 May 1997 02:18:00 -0500 (CDT) Received: from domo by mail.xmission.com with local (Exim 1.62 #1) id 0wWDcf-0002Ph-00; Mon, 26 May 1997 22:12:33 -0600 Received: from slack.xmission.com [199.104.120.18] (pashdown) by mail.xmission.com with smtp (Exim 1.62 #1) id 0wWDcc-0002PR-00; Mon, 26 May 1997 22:12:30 -0600 Received: (from pashdown@localhost) by slack.xmission.com (8.8.5/8.7.5) id WAA20710 for usr-tc@xmission.com; Mon, 26 May 1997 22:12:29 -0600 (MDT) Received: from soback.kornet.nm.kr [168.126.3.4] by mail.xmission.com with smtp (Exim 1.62 #1) id 0wVyT7-0005h9-00; Mon, 26 May 1997 06:01:41 -0600 Received: from localhost (audience@localhost) by soback.kornet.nm.kr (8.6.12h2/8.6.9) with SMTP id VAA09492; Mon, 26 May 1997 21:00:16 +0900 X-Sender: audience@soback4 cc: jslee@usr.co.kr In-Reply-To: <Pine.LNX.3.95.970515104827.21789B-100000@ns.usr.co.kr> Message-ID: <Pine.GSO.3.94K.970526204455.894A-100000@soback4> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-usr-tc@xmission.com Precedence: bulk Reply-To: usr-tc@mail.xmission.com --IMA.Boundary.138038468--

--IMA.Boundary.338038468 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Description: cc:Mail note part Haven't. No plans that I'm aware of to produce anything for "FreeBSD", and anything we do with BSDi would most likely be available only in the bundle with the EdgeServer. Kurtiss ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~ __ __ ~ ~ Kurtiss Johnson | | | | RRRRR ~ ~ Product Manager | | | | ***RR RR ~ ~ US Robotics | \_/ |*** RRRRR ~ ~ kjohnson@usr.com \___/ RR RR ~ ~ See us at www.usr.com! ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ______________________________ Reply Separator _________________________________ Author: "Webmaster" <CTHOMPSON@wingnet.net> at Internet I've heard through the grapevine that USR has developed this software package for use with FreeBSD. Is this true? There are a load of BSDI users out here that could make use of such an animal. Since BSDI & USR are working together on the EdgeServer, it makes sense that USR would do this if they haven't. Comments? Webmaster http://www.wingnet.net --IMA.Boundary.338038468 Content-Type: text/plain; charset=US-ASCII; name="RFC822 message headers" Content-Transfer-Encoding: 7bit Content-Description: cc:Mail note part Content-Disposition: attachment; filename="RFC822 message headers" Received: from usr.com (mailgate.usr.com) by robogate2.usr.com with SMTP (IMA Internet Exchange 2.02 Enterprise) id 385CCD10; Fri, 23 May 97 11:58:57 -0500 Received: from mail.xmission.com by usr.com (8.7.5/3.1.090690-US Robotics) id LAA20229; Fri, 23 May 1997 11:37:42 -0500 (CDT) Received: from domo by mail.xmission.com with local (Exim 1.62 #1) id 0wUx8Q-0003zz-00; Fri, 23 May 1997 10:24:06 -0600 Received: from agape.wingnet.net [206.30.215.2] (root) by mail.xmission.com with smtp (Exim 1.62 #1) id 0wUx8L-0003zk-00; Fri, 23 May 1997 10:24:02 -0600 Received: from makrothumia.wingnet.net (makrothumia.wingnet.net [206.30.215.5]) by agape.wingnet.net (8.8.5/8.7.3) with SMTP id MAA11221 for <usr-tc@xmission.com>; Fri, 23 May 1997 12:24:00 -0400 (EDT) Message-Id: <199705231624.MAA11221@agape.wingnet.net> Comments: Authenticated sender is <craig@mail.wingnet.net> Organization: WingNET Internet Services X-mailer: Pegasus Mail for Windows (v2.53/R1) Sender: owner-usr-tc@xmission.com Precedence: bulk Reply-To: usr-tc@mail.xmission.com --IMA.Boundary.338038468--

In order to do multilink between chassis, you need an MPIP daemon running with Netserver code higher than 3.4.23. I asked a while back where we are supposed to get the MPIP daemon, since it isn't on USR's site. Nobody seemed to know.

I'm exploring installing a PRI connection to the total control unit. Do I have to have Voice and Data service or can I live with data only lines? USWorst insists that I need Voice/Data service which costs $80 per month per line here v. $53 per month per line for data only. Any of you have any experience with this? Thanks, Greg Coffey, CoffeyNet ~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~ 142 S. Center St. 307-234-5443 307-234-5446 Fax Casper, WY 82601 Local Internet for Casper, Rawlins, Douglas, http://www.coffey.com Wheatland, Pinedale, Lander & Lusk, WY ~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~

--IMA.Boundary.438158468 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Description: cc:Mail note part These are Ascend proprietary attributes that exist in the Experimental and Reserved sections of the RADIUS attribute space. The NETServer and NMC don't understand those attributes and may choke on them (depending on code you're running). As far as what you NEED to configure, there isn't anything. You CAN configure a Max-Port-Limit for the user. Kurtiss ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~ __ __ ~ ~ Kurtiss Johnson | | | | RRRRR ~ ~ Product Manager | | | | ***RR RR ~ ~ US Robotics | \_/ |*** RRRRR ~ ~ kjohnson@usr.com \___/ RR RR ~ ~ See us at www.usr.com! ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ______________________________ Reply Separator _________________________________ Author: Laszlo Vecsey <master@internexus.net> at Internet What needs to be configured in radius for a dialup user to establish a Multilink PPP session? I'm using unix merit radius, and the only related items I found in the dictionary are these: ATTRIBUTE Ascend-Multilink-ID 187 integer ATTRIBUTE Ascend-Num-In-Multilink 188 integer - lv --IMA.Boundary.438158468 Content-Type: text/plain; charset=US-ASCII; name="RFC822 message headers" Content-Transfer-Encoding: 7bit Content-Description: cc:Mail note part Content-Disposition: attachment; filename="RFC822 message headers" Received: from usr.com (mailgate.usr.com) by robogate2.usr.com with SMTP (IMA Internet Exchange 2.02 Enterprise) id 38C1FE90; Wed, 28 May 97 07:07:05 -0500 Received: from mail.xmission.com by usr.com (8.7.5/3.1.090690-US Robotics) id GAA15388; Wed, 28 May 1997 06:45:45 -0500 (CDT) Received: from domo by mail.xmission.com with local (Exim 1.62 #1) id 0wWTId-0002JR-00; Tue, 27 May 1997 14:56:55 -0600 Received: from micro.internexus.net [206.152.14.2] (R-Foo) by mail.xmission.com with esmtp (Exim 1.62 #1) id 0wWTIb-0002JH-00; Tue, 27 May 1997 14:56:53 -0600 Received: from localhost (master@localhost) by micro.internexus.net (8.8.5/8.8.4) with SMTP id QAA23019 for <usr-tc@xmission.com>; Tue, 27 May 1997 16:56:46 -0400 Message-ID: <Pine.LNX.3.96.970527165420.22939A-100000@micro.internexus.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-usr-tc@xmission.com Precedence: bulk Reply-To: usr-tc@mail.xmission.com --IMA.Boundary.438158468--

On Wed, 28 May 1997, Greg Coffey wrote: > I'm exploring installing a PRI connection to the total control unit. Do I > have to have Voice and Data service or can I live with data only lines? > USWorst insists that I need Voice/Data service which costs $80 per month per > line here v. $53 per month per line for data only. Any of you have any > experience with this? You'll need the voice component if you wish to accept any analog modem calls including X2. What you shouldn't need is any DID numbers which US West may force you to have. Because US West is so expensive, I went with MCI who isn't forcing us to have DID. They just have one phone # for the entire hunt group. Brian

--IMA.Boundary.638158468 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Description: cc:Mail note part That could be easily done with Max-Port-Limit (by setting it to one port, the user can negotiate MP but won't get the added bandwidth). Kurtiss ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~ __ __ ~ ~ Kurtiss Johnson | | | | RRRRR ~ ~ Product Manager | | | | ***RR RR ~ ~ US Robotics | \_/ |*** RRRRR ~ ~ kjohnson@usr.com \___/ RR RR ~ ~ See us at www.usr.com! ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ______________________________ Reply Separator _________________________________ Author: Laszlo Vecsey <master@internexus.net> at Internet What about restricting it by default for most users? On Tue, 27 May 1997, MegaZone wrote: > Once upon a time Laszlo Vecsey shaped the electrons to say... > >What needs to be configured in radius for a dialup user to establish a > >Multilink PPP session? I'm using unix merit radius, and the only related > > Nothing. There is nothing at all special in RADIUS for this. It should > be on by default. > > > ATTRIBUTE Ascend-Multilink-ID 187 integer > > ATTRIBUTE Ascend-Num-In-Multilink 188 integer > > These are Ascend specific attributes for some extensions they alone > created. > > -MZ > -- > Livingston Enterprises - Chair, Department of Interstitial Affairs > Phone: 800-458-9966 510-737-2100 FAX: 510-737-2110 megazone@livingston.com > For support requests: support@livingston.com <http://www.livingston.com/> > Snail mail: 4464 Willow Road, Pleasanton, CA 94588 > > --IMA.Boundary.638158468 Content-Type: text/plain; charset=US-ASCII; name="RFC822 message headers" Content-Transfer-Encoding: 7bit Content-Description: cc:Mail note part Content-Disposition: attachment; filename="RFC822 message headers" Received: from usr.com (mailgate.usr.com) by robogate2.usr.com with SMTP (IMA Internet Exchange 2.02 Enterprise) id 38B69910; Tue, 27 May 97 18:09:05 -0500 Received: from mail.xmission.com by usr.com (8.7.5/3.1.090690-US Robotics) id RAA00617; Tue, 27 May 1997 17:47:47 -0500 (CDT) Received: from domo by mail.xmission.com with local (Exim 1.62 #1) id 0wWUvz-0001Gq-00; Tue, 27 May 1997 16:41:39 -0600 Received: from micro.internexus.net [206.152.14.2] (R-Foo) by mail.xmission.com with esmtp (Exim 1.62 #1) id 0wWUvu-0001GX-00; Tue, 27 May 1997 16:41:34 -0600 Received: from localhost (master@localhost) by micro.internexus.net (8.8.5/8.8.4) with SMTP id SAA25463 for <usr-tc@mail.xmission.com>; Tue, 27 May 1997 18:41:32 -0400 In-Reply-To: <199705272148.OAA01319@server.livingston.com> Message-ID: <Pine.LNX.3.96.970527184051.25446A-100000@micro.internexus.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-usr-tc@xmission.com Precedence: bulk Reply-To: usr-tc@mail.xmission.com --IMA.Boundary.638158468--

Pete Ashdown <pashdown@xmission.com> writes: > In order to do multilink between chassis, you need an MPIP daemon running > with Netserver code higher than 3.4.23. I asked a while back where we are > supposed to get the MPIP daemon, since it isn't on USR's site. Nobody > seemed to know. You can also use a NETServer itself as an MPIP server rather than an off-node Unix system. I forget how many individual NETServers can participate this way (it's a bit more bounded than a Unix server, and somewhat dependent on what else you are using the NETServer for in terms of services), but it may be suitable for a particular configuration. -- David /-----------------------------------------------------------------------\ \ David Bolen \ Internet: db3l@ans.net / | ANS Communications \ Phone: (914) 789-5327 | / 100 Clearbrook Road, Elmsford, NY 10523 \ Fax: (914) 789-5310 \ \-----------------------------------------------------------------------/

Hi, We have four USR rack. We found two of them with software 3.1.7. This rack netserver card initially and randomly drop user dail-up lines. After contacting with USR, they suggest to upgrade 3.3.3. We find that it doesn't work our existing radius 2.0 version. The radiusd deamon even dies out unreasonably after upgrading. Can anyone suggest this problem? Kenneth Tse

On Wed, 28 May 1997, Brian Elfert wrote: > > On Wed, 28 May 1997, Greg Coffey wrote: > > > I'm exploring installing a PRI connection to the total control unit. Do I > > have to have Voice and Data service or can I live with data only lines? > > USWorst insists that I need Voice/Data service which costs $80 per month per > > line here v. $53 per month per line for data only. Any of you have any > > experience with this? > > You'll need the voice component if you wish to accept any analog modem > calls including X2. What you shouldn't need is any DID numbers which US > West may force you to have. > > Because US West is so expensive, I went with MCI who isn't forcing us to > have DID. They just have one phone # for the entire hunt group. US West also gave us a big block of DID numbers when we got "advanced T1" for dialup. I just picked the easiest number for our customers to remember and advertised it. Before, when we had only "basic T1", they wouldn't do x2 because US West had provisioned them through back-to-back D4 channel banks. Since the basic t1 circuits were cheaper, we kept them and just made them overflow (hunt) to our "advanced T1" and gave the new phone number to the x2 customers. If I had it to do over we would have ordered PRI from TCG. The problem was US West couldn't port our phone number without charging by the minute for calls forwarded to our TCG number. So, instead of changing phone numbers on hundreds of customers, we stuck with US West. Does anybody else have nightmare stories about having to switch dialup phone numbers? I have had to do it in 4 cities now in order to save money. There are always a few stubborn customers who won't change until you turn the old number completely off. . . no matter how much e-mail you send. . . and then there are those users who never check their local mail. -CH

On Wed, 28 May 1997, Charles Hill wrote: > If I had it to do over we would have ordered PRI from TCG. The problem > was US West couldn't port our phone number without charging by the minute > for calls forwarded to our TCG number. So, instead of changing phone > numbers on hundreds of customers, we stuck with US West. I thought US West would charge by the minute for forwarded calls only if the line was measured. One problem with call forwarding is that the 5ESS and 1AESS switches can only forward 99 calls before giving busy signals. Also, a 1A switch has what can be a major problem for most ISPs. When a call is forwarded from a 1A switch, no more calls can be forwarded until the line is answered on the other end. Any other calls during this time will get busy signals. > Does anybody else have nightmare stories about having to switch dialup > phone numbers? I have had to do it in 4 cities now in order to save > money. There are always a few stubborn customers who won't change until > you turn the old number completely off. . . no matter how much e-mail you > send. . . and then there are those users who never check their local mail. I don't have any nightmare stories so far. I'm currently forwarding a 40 line residential hunt group to a 40 line business hunt group. With residential service, you can do unlimited forwarding with no per minute costs. I plan to abandon the residential hunt group this fall because of the forwarding problems. Brian

Brian Elfert said once upon a time: >I thought US West would charge by the minute for forwarded calls only if >the line was measured. They don't. >One problem with call forwarding is that the 5ESS and 1AESS switches can >only forward 99 calls before giving busy signals. Also, a 1A switch has >what can be a major problem for most ISPs. When a call is forwarded from >a 1A switch, no more calls can be forwarded until the line is answered on >the other end. Any other calls during this time will get busy >signals. Aye on that one. We have four lines that forward to 350, which we still get busy signals on because people use just one of them. We're eventually going to abandon this scheme. >> Does anybody else have nightmare stories about having to switch dialup >> phone numbers? I have had to do it in 4 cities now in order to save >> money. There are always a few stubborn customers who won't change until >> you turn the old number completely off. . . no matter how much e-mail you >> send. . . and then there are those users who never check their local mail. It hasn't been bad for us. We switched from an analog US West number to a digital ELI USR-TC based number. Once you tell people that they'll get better results with the new number, they're happy to switch. The only problem is with the Quake players (due to the TC bug). They're still using the old number. However the old number has been cut back from 250 lines to 36 now.

Some of our usr tc's seem to be spontaneously rebooting themselves. Has anyone else observed this feature? -- Stephen Corbesero This message has been brought to you by electrons. flash@early.com Electrons -- The other charged particle.