[Asterisk-Users] Two Zap Problems with 1.0.2 that appeared at the same time: choppyness and squealing

[Ted Cabeen]

I've got an * system that is having some real problems with 1.0.2.
The biggest problem is that calls going through my T100P get choppy
for about 10 seconds every 1 or 2 minutes.  Asterisk is running on a
debian stable system with current packages.  The T100P is plugged into
a Adit Channelbank with 8 POTS lines hooked up to the Channelbank.
I've watched the vritual memory and CPU status on the * box during the
call and the system is totally idle.  Looking at the verbose logging
in *, there isn't any obvious activity on the console that corresponds
with the choppyness.  It sounds like this is the echo canceller having
problems, but regular phones plugged into the lines sound fine.  In
early November, this system was running fine on 1.0, but even
downgrading back to 1.0 now doesn't fix the problem.  Any thoughts?

The other problem is related to our legacy PBX system, which we
route through * when the * is being tested.  To do so, we run the 7
PBX lines into a FXS card on the Channelbank and configure * to bridge
the PBX lines to the POTS lines as necessary.  The problem is that
when the employee talking through * on the PBX hangs up the line, the
caller coming in through the POTS line hears a very loud squeal for a
second until * hangs up the line.  I've run the channelbank with a
loopbacked T1 replicating the same problem, and it doesn't happen, so
* is definitely involved somehow.  Is there a setting to determine how
asterisk detects hangups on FXS Zap channels?

-- 
Ted Cabeen           http://www.pobox.com/~secabeen            ted at impulse.net 
Check Website or Keyserver for PGP/GPG Key BA0349D2         secabeen at pobox.com
"I have taken all knowledge to be my province." -F. Bacon  secabeen at cabeen.org
"Human kind cannot bear very much reality."-T.S.Eliot       secabeen at gmail.com