[Home]

Summary:ASTERISK-20718: Asterisk crashes or locks on 'queue show'
Reporter:Roberto (rsanchez@iniciasoluciones.es)Labels:
Date Opened:2012-11-22 02:27:43.000-0600Date Closed:2013-01-08 13:23:08.000-0600
Priority:MajorRegression?
Status:Closed/CompleteComponents:Applications/app_queue
Versions:1.8.15.0 Frequency of
Occurrence
Constant
Related
Issues:
Environment:Asterisk 1.8.11 cert8/Centos 5.8 on VirtualBOX, Asterisk 1.8.11 cert8/Debian 6 on Virtual BOX, Asterisk 1.8.11 cert8/Centos 6.3, Asterisk 1.8.15/Centos 5.8 on VirtualBOX, Asterisk 1.8.15/Debian 6 on Virtual BOX, Asterisk 1.8.15/Centos 6.3, Asterisk 1.8.15/Hyper-V, Asterisk 1.8.11/Hyper-VAttachments:( 0) backtrace-threads.rar
Description:When invoke QueueAdd function through telnet by 5038 port Asterisk it doesn't response and then, if invoke "queue show" on Asterisk CLI it crashes and no more actions can be made. Don't know if the problem occurs only  on virtual machines but it occurs on every development enviroment i've tried.
Comments:By: Dmitry Pavlenko (pavlenkoda) 2012-11-22 04:13:57.921-0600

I have the same issue: I issue "queue show" and asterisk freezes. Nothing is written into /var/log/messages.
I am on a real server (not virtual).
I upgraded from previous asterisk 1.8.11-cert7 to this new version (asterisk 1.8.11-cert8). asterisk 1.8.11-cert7 works perfectly in my call center.

By: Roberto (rsanchez@iniciasoluciones.es) 2012-11-22 04:26:04.707-0600

But i'm already using 1.8.11-cert8. I think the problem is not on version because i've reproduce the same behavior on 1.6.2.24 and 1.8.15 versions, don't know what could be...i guess it's some rpm installed on my system because all enviroments have been made by installing  all without any distribution.
Could you tell me which modules use no make menuselect? Thanks

By: Dmitry Pavlenko (pavlenkoda) 2012-11-22 05:15:21.708-0600

I am on Centos 6.3 and Linux asterisk 2.6.32-279.14.1.el6.x86_64. Do you need rpm -qa?

I returned 1.8.11-cert7 - it works perfectly. If it was a module to blame then I could not compile 1.8.11-cert7 either.

This is written into /var/log/asterisk/messages after I call into the Queue()
chan_sip.c: Autodestruct on dialog '1425428265@192.168.1.160' with owner in place (Method: BYE)
=======================================================================

I must say that I did yum upgrade yesterday. I use epel and rpmforge repositories as well. So may be it is something about it.
I will recompile once again 1.8.11-cert7 to see if it is true.




By: Roberto (rsanchez@iniciasoluciones.es) 2012-11-22 07:56:54.124-0600

there must be anymore, with Asterisk version 11 works...ther must be anything wrong on versions earlier than 11, don't know what...

By: Matt Jordan (mjordan) 2012-11-22 09:23:49.502-0600

Thank you for your bug report. In order to move your issue forward, we require a backtrace[1] from the core file produced after the crash. Also, be sure you have DONT_OPTIMIZE enabled in menuselect within the Compiler Flags section, then:

make install

After enabling, reproduce the crash, and then execute the backtrace[1] instructions. When complete, attach that file to this issue report.

[1] https://wiki.asterisk.org/wiki/display/AST/Getting+a+Backtrace



By: Dmitry Pavlenko (pavlenkoda) 2012-11-25 22:54:31.271-0600

I used DONT OPTIMIZE, DEBUG THREADS and BETTER BACKTRACES compiler flags.

The attachment containes the result  of: gdb -ex "thread apply all bt" --batch /usr/sbin/asterisk `pidof asterisk` > /tmp/backtrace-threads.txt

another file in the attachment is what I see on the screen while performing : asterisk -rx "core show locks" > /tmp/core-show-locks.txt

Test Scenario:

my Perl daemon once in 3 seconds sends QueueStatus AMI reques and receives no reply. In fact I can issue the same Queue show command in the console but I get no reply either


By: Dmitry Pavlenko (pavlenkoda) 2012-12-10 00:26:47.668-0600

I reinstalled the Centos 6.3. This time I did not use neither epel nor RPMFORGE repositories. First I checked everything with certified-asterisk-1.8.11-cert7. There were no problems with app_queue module.
After that I installed certified-asterisk-1.8.11-cert9. There are no problems with app_queue either.

It seems that some unappropriate rpms were updated from either epel or rpmforge repositories when I encountered this issue.

With base Centos repositories - there are no problems with app_queue.

I do not have a certified-asterisk-1.8.11-cert8 copy so I did not test it in this new environment.


By: Rusty Newton (rnewton) 2013-01-08 13:23:08.498-0600

This doesn't sound like an issue with Asterisk, as evidenced by Dmitry's last comment. Please contact a bug marshal in #asterisk-bugs if you feel differently and want to look at re-opening the issue.