Re: (PM) 3.8b15 and no authentication

Alex Weeks (awpmmail@ilinkusa.net)
Wed, 6 May 1998 13:03:55 -0500

Our problems were confined to Rockwell and USR chipsets.
The Rockwell 56k modems seemed to be unaffected, however the Rockwell 33.6
and 28.8 modems would sometimes require several attempts before a connection
was established. Then these same users were reporting more frequent
disconnects. I didn't check code or chip revisions on these modems.

On the USR side we only got reports on Sportsters and Winmodems. Those
modems with the V.90 upgrade were receiving good connection rates (around
48000), the 33.6 and 28.8s were down in their normal range. The complaint
from these users was just very frequent disconnects (30 second to 2 mintue
connect time).

Neither of these modem problems was accompanied by an excessive number of
retrains. Everything else in modem stats looked normal.

One of my USR customers with the V.90 ugrade called to ask if I could tell
him how to disable V.90 because his throughput seemed faster at 33.6.
Unfortunately he lost interest before this could be accomplished.

At any rate, I had to bail and get back to 3.7.2 to make the phone stop
ringing, so I was not able to properly document all the problems. I
wouldn't regard the above as strong fact, it's just my observations.

Oh, and the Zoltrix 56K modems just don't seem to work well at all. Can't
get those to negotiate at anything better than 4800 (no not a typo) on 3.7.2
and on 3.8b15 they won't negotiate at all.

Alex Weeks
I-Link USA

-----Original Message-----
From: Josh Richards <jrichard@livingston.com>
To: portmaster-users@livingston.com <portmaster-users@livingston.com>
Date: Tuesday, May 05, 1998 2:44 PM
Subject: Re: (PM) 3.8b15 and no authentication

>On 4 May 1998, Alex Weeks wrote:
>> Ironically, it was 3.8b15 that did it. Here's the issue, a couple of
users
>> can call, the modems negotiate, but then they are not authenticated.
Change
>> the phone number to a box that is running 3.7.2 and everything is fine
>> again. With some users this problem is repeatable 100% of the time, with
>> others it happens 3 or 4 out of 5 times. Then, of course, some not
affected
>> by it at all.
>>
>> Both boxes authenticate with the same RADIUS server. On one of the boxes
>> where we were having trouble we knocked it back to 3.7.2 and everything
was
>> fixed. In other words, the Portmaster configuration seems to be fine.
>>
>> Looking at the debug output it looks as if PPP starts and then nothing.
>> Eventually the modems give up.
>>
>> Anyone else seen this? Any ideas what is going on?
>
>Any pattern to what type of modems these users have? A certain client
>chipset? Certain rev. of client chipset? Etc. What types of connections
>are these and what sort of speeds are they seeing? Under a "show mX"
>(where X == the number shown under a "show modems" for their port), what
>do the stats for the call look like?
>
>--jr
>
>----
>Josh Richards - <jrichard@livingston.com> - [Beta Engineer]
>LUCENT Technologies - Remote Access Business Unit
>(formerly Livingston Enterprises, Inc.)
>http://www.livingston.com/
>
>-
>To unsubscribe, email 'majordomo@livingston.com' with
>'unsubscribe portmaster-users' in the body of the message.
>Searchable list archive: <URL:http://www.livingston.com/Tech/archive/>
>

-
To unsubscribe, email 'majordomo@livingston.com' with
'unsubscribe portmaster-users' in the body of the message.
Searchable list archive: <URL:http://www.livingston.com/Tech/archive/>