The problem you're running into is a bug in the Max OS (talked to Ascend
about this one). Someone else here had a similar problem recently and I
replied with the steps to fix it. I attatched it below...
As for the V.90, on our Max 4004 and 4048, we're using 6.1.0 and have had
no problems at all with it. So far, I think the highest speed that I've
seen connect to us is 50666 or 50667 (something like that). Our 4002 is is
only V.34 and gets hit only if the other two Maxes are full--that still has
6.0.4 on it.
Anyway, here's the fix for your Max:
Check the following:
Ethernet --> Mod Config --> WAN options
Make sure that you have as many IPs in your pool(s) as you have
modems.
Go into your Max and change the following setting:
Ethernet --> Mod Config --> Auth
Auth Pool =
Change it to "No"
You will have to reset the Max after this or you'll never free up the IPs.
After setting and saving the above option:
Hit ^d then d (diagnostics)
Type in:
fsave
fclear
nvr (this resets the max)
Josh Hillman
hillman@talstar.com
> From: Pat Augustine <pata@ilnk.com>
> We've recently installed RadiusNT 2.5 on our backup Radius server, still
> running 2.2 on the primary. Users are signing on (mostly) ok.
> Checking Radlogs (and populating RadLogMsg per your previous message), I
> get the occasional user error. However, the table is also getting 1 entry
> per minute from each of our Maxes that basically says the Max itself
tried
> to login and failed (user not found). The Radlog entry looks like this:
>
> 10 Jun 15 1998 7:57PM pools-Max2.ilnk.com
> ascend (null)
(null)
> (null)
>
> Max2.ilnk.com is the name of the 4004. I see them go by in -x15 mode, as
> well, but I don't know what's causing it. We've upgraded the Max to
6.1.0+
> recently and have been advised by Ascend that it has an IP Pool leak that
> they hope to fix soon, but I don't know if that's related or not.
> Does this make any sense to you at all? Any known issues with Ascend
6.1.0+
> (the first release with v.90)?
>
> In addition, I seem to get little bubbles of unsuccessful logins. During
> the day, suddenly for about 15 minutes it seems everybody is getting a
LAN
> SECURITY ERROR and being rejected and then 15 minutes later, everybody
gets
> in just fine with no changes anywhere. Most of the rejects don't show up
in
> the Radius Logs (either the logfile on disk or the Radlogs table in
Emerald)
> at all, so the Max may just be rejecting them outright, but I don't know
why
> it would do that (I've got excess IP Pools allocated so that the Max
> shouldn't be out of IP addresses and did a Show Pool to verify that I had
at
> least SOME available IP addresses).