Ascend Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: (ASCEND) 56K modem disconnects

Richard P. Eldridge wrote:

> Has anyone had any complaints with the 56K Flex modems disconnecting at
> random?  I have three Max 4048 boxes with 2 - 24 count modem cards in each.
>  I've had quite a bit of trouble here and starting to hear some complaints
> with my customers.  Do you have any suggestions?
>

Our Ascend equipment is the following:

    MAX 4000
    Serial No. 6494511
    Software Revision 5.0Ap36
    K56flex Mod-8 Modem Cards
    A single Primary Rate Interface (PRI)

About two to three weeks ago, we contacted Ascend technical support about this
issue, because a number of subscribers were complaining about random
disconnects. After going through the usual modem checks with the problematic
subscribers, we could not solve their problems nor did we see any correlations
such as a particular brand of modem (Hayes, USR, Motorola), technology
(K56flex, 33.6, etc.), or customer location (different cities) being the
culprit. The ascend-disconnect-cause-code for the random disconnects is either
11 (lost carrier) or 185 (remote end hung up). Here is what has transpired so
far:

1. Our first call into Ascend's technical support resulted in the
recommendation that we upgrade to 5.0Ap46. We performed the upgrade, but it
made no difference.

2. Our second call into Ascend technical support resulted in the suspicion that
we may have early revisions (A, B, or C) of K56flex modem cards which were
known to have problems. We checked each of our modem cards and found that all
were revision level F (the latest).

3. Our third call into Ascend technical support resulted in the recommendation
that we contact our telco (Pacific Bell) to have our PRI line tested. PacBell
tested our line at the central office switch of origin and deemed it to be
'pristine'. PacBell suggests it may be our CSU/DSU.

4. We moved our PRI from the WAN1 port to the WAN2 port on the MAX in order to
use a different CSU/DSU. Our customers were still being randomly disconnected.

5. We again contact PacBell. This time, they bring test equipment (DynaTech TTC
T-BERD Model 224) and connect it to the PairGain termination box of the PRI. It
has been monitoring the PRI for 48 hours now with only 4 CRC errors...not bad.
During this same period, one customer is randomly disconnected 15 out of 16
sessions.

6. We moved our PRI from the WAN2 port to the WAN3 port on the MAX...it made no
difference.

7. At the behest of the PacBell maintenance engineer, we contact the customer
and request he contact 611 repair and have them test his DS0 line. PacBell
tests the line and says it is clean.

8. Next, we have the customer contact his modem manufacturer to ensure he is
running the latest code in his modem. He is told by USR that he has the latest
33.6 code, revision level 2.0.

9. During all this, we have been performing our own dial-up tests using a Hayes
K56flex modem with Rockwell code version 1.120 and a standard DS0 line. We get
randomly disconnected as well!!! Our DS0 and PRI come from the same central
office. Hmm....

Okay...so now what do we do? The commonality here is the MAX and the telco, but
which one? It appears PacBell has shown the PRI to be stable (although we're
still suspicious) and that the MAX is the culprit. What is even more
interesting is that others on this list are starting to report the same
problem. Any suggestions?

Kim W. Premuda
FastWave Internet Services
San Diego, CA
kim@fastwave.net


++ Ascend Users Mailing List ++
To unsubscribe:	send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd:	<http://www.nealis.net/ascend/faq>
Follow-Ups: References: