Summary:ASTERISK-20262: Asterisk console becomes unresponsive under heavy call load
Reporter:Jeremy Pepper (jpepper)Labels:
Date Opened:2012-08-20 13:01:13Date Closed:2012-08-30 16:53:21
Versions:11.0.0-beta1 Frequency of
Environment:Ubuntu 10.04.1 LTSAttachments:
Description:I've only managed to trigger this a couple times so far. Each time, I've held up 100 simultaneous SIP calls via the manager between a queue and another test server. It works for a while, but eventually I stop seeing verbose messages on the console, and when I run {{core show channels}} to confirm there are still calls up, the console hangs.

The last time I triggered it, I had passed the {{-p}} option to Asterisk. I don't remember if I had the first time as well. There are still free CPU cycles to be had, though—Apache and SSH remained responsive on the same box.

I have triggered this issue with both {{-r}} and {{-c}}.
Comments:By: Richard Mudgett (rmudgett) 2012-08-20 14:46:04.799-0500

Debugging deadlocks: Please select DEBUG_THREADS and DONT_OPTIMIZE in the Compiler Flags section of menuselect. Recompile and install Asterisk (i.e. make install).  This will then give you the console command "core show locks." When the symptoms of the deadlock present themselves again, please provide output of the deadlock via:

# asterisk -rx "core show locks" | tee /tmp/core-show-locks.txt
# gdb -se "asterisk" <pid of asterisk> | tee /tmp/backtrace.txt
gdb> bt
gdb> bt full
gdb> thread apply all bt

Then attach the core-show-locks.txt and backtrace.txt files to this issue. Thanks!

By: Jeremy Pepper (jpepper) 2012-08-30 15:49:04.325-0500

Terribly sorry about the delayed response!

This turned out to be caused by a patch we had applied and not a problem I could reproduce on completely stock Asterisk.

Sorry for jumping the gun reporting this!

By: Matt Jordan (mjordan) 2012-08-30 16:53:21.839-0500

No problem - thanks for letting us know!