[asterisk-bugs] [Asterisk 0012595]: Asterisk stops accepting calls from Zap and responds with fast busy

[noreply at bugs.digium.com]

A NOTE has been added to this issue. 
====================================================================== 
http://bugs.digium.com/view.php?id=12595 
====================================================================== 
Reported By:                dawebber
Assigned To:                
====================================================================== 
Project:                    Asterisk
Issue ID:                   12595
Category:                   Channels/chan_zap
Reproducibility:            sometimes
Severity:                   minor
Priority:                   normal
Status:                     new
Asterisk Version:           1.4.19 
SVN Branch (only for SVN checkouts, not tarball releases): N/A 
SVN Revision (number only!):  
Disclaimer on File?:        N/A 
Request Review:              
====================================================================== 
Date Submitted:             05-06-2008 14:49 CDT
Last Modified:              05-12-2008 08:52 CDT
====================================================================== 
Summary:                    Asterisk stops accepting calls from Zap and responds
with fast busy
Description: 
This issue occurs approximately every 2 weeks on a slightly loaded box
(average 6 simultaneous incoming calls 24/7), but the issue once occurs on
a channel, will reoccur at any time, even with one incoming call. This box
doesn't place any outgoing calls. Periodic automatic Zap channel clearing
is set to 1800 seconds. The error in the asterisk logs looks like this:

<snip>
[May  5 09:17:45] DEBUG[17999] chan_zap.c: Ring requested on channel 0/1
already in use or previously requested on span 1.  Attempting to
renegotiating channel.
[May  5 09:17:45] DEBUG[17999] chan_zap.c: Ring requested on channel 0/1
already in use or previously requested on span 1.  Attempting to
renegotiating channel.
<snip>


The errors always occur in pairs as shown above. This error is similar to
the one described here:
http://www.trixbox.org/forums/trixbox-forums/help/inbound-calls-over-pri-get-fast-busy

I've also noticed that the error always occurs on channel 1 of the span.
After the error had occured several times, the calls might still go through
on the channel.

Eventually, the asterisk process will crash and restarting Asterisk takes
care of the issue. the snippet of log data above occured right before
crash.

Memory consumption and CPU level stay relatively low. No IRQ conflicts
with Digium Card are detected.


====================================================================== 

---------------------------------------------------------------------- 
 dawebber - 05-12-08 08:52  
---------------------------------------------------------------------- 
I was able to reproduce this issue today. On my PRI (standard setup: 23 B
channels + D), channel 3 appeared to be in locked state. So, I ran PRI
DEBUG. the output is attached. You can probably ignore the trailing set of
statements "Unable to handle return result on switchtype 3!". This is a
result of 2B-channel transfer on a 5ESS switch. It's not relevant to this
issue.

Here's the sequence of events (I didn't obfuscate any phone numbers or
caller ID info in the debug file). I made 2 phone calls to the PRI, so now
the first 3 channels appeared busy (3rd one was already locked up). I then
proceeded to make a third call, at which point I got busy signal back. I
then hung up all calls and stopped debugging. 

Issue History 
Date Modified   Username       Field                    Change               
====================================================================== 
05-12-08 08:52  dawebber       Note Added: 0086709                          
======================================================================