Re: [Emerald] Ascend Disconnect Reason

Josh Hillman ( (no email) )
Wed, 31 May 2000 20:45:37 -0400

> Scottie Arnett wrote:
> > I dug this out of the archives and was wanting to know how I go about
doing
> > this? I am using a Lucent Max 6048 with Radius NT 2.5.331.
> >
> > "Ascend reports disconnect using Ascend-Disconnect-Reason, not the RFC
> > standard Acct-Terminate-Cause. You can rename ADR to ATC, to allow it
> > to record, but the numbers will not be translated in Emerald."
> >
> > Is it a process with an SQL statement? In radiusNT?
> >
> > How do I translate the numbers to what they will mean? Is it in the
> > Ascend/Lucent manual?

The Ascend-Disconnect-Cause gives MUCH more detail about the various
disconnect causes than the RFC one does. Acct-Terminate-Cause lists 18
disconnect causes. Ascend-Disconnect-Cause has at least 74 documented ones.
See the files below. I'll keep them there at least for a few days. After
that, if someone wants them and they're not there, shoot me an email.

http://support.yourvillage.com/Ascend-Disconnect-Cause.txt
http://support.yourvillage.com/Ascend-Connect-Progress.txt
http://support.yourvillage.com/causeprogress.txt

Here are the most common disconnect causes (verbose) associated with PPP
sessions. The references to "20 minutes" and "several hours" are simply
what we have our stuff set for. I send the following information to people
anytime I send their call stats to them (have a handy script that I wrote
that outputs all of the useful connection info).

Normal disconnects:
45 - Normal disconnect initiated by the user or user's software.
100 - Session was disconnected by our equipment because no data was
being sent or received for 20 consecutive minutes (idle-
timeout). OR, the maximum session duration (several hours)
has been reached.
151 & 152 - Disconnected by ISP administrator.
Abnormal disconnects:
11 - Loss of carrier. This is usually caused by line noise
problems between the user's modem and the phone company
(the part of the connection that is analog). Call waiting
tones are interpreted as line noise also.
Windows NT sessions do not terminate correctly and will
always produce this value.
185 - Loss of signal from remote (customer) end:
Similar to 11, but can also be more specific to a modem
problem on the user's side, too many modem retrains (usually
from poor line conditions between the user and the phone
company), computer or software crashed while the user was
online, the phoneline was unplugged while the user was online.

Josh

For more information about this list (including removal) go to:
http://www.iea-software.com/support/maillists/liststart