[SR-Users] Kamailio fails to generate core dump

[Ovidiu Sas]

IIRC, you get that message if core dumping is not properly enabled at
system level. Check that you have those settings and kill kamailio with a
SIGSEGV to test.
It always worked for me. I can't afford to install a server without core
dumping enabled.

-ovidiu
On Jan 3, 2015 5:23 PM, "Alex Balashov" <abalashov at evaristesys.com> wrote:

> Thanks, Ovidiu. But since the message "core was not generated" comes from
> handle_sigs() in Kamailio, what I am really interested is in what reason
> Kamailio itself would have for not dumping core. I assume that the
> situation would look different if Kamailio tried to dump core, but was
> restrained from doing so by operating system factors such as ulimits.
>
> From main.c:handle_sigs(), it appears that the log message came from here:
>
> #ifdef WCOREDUMP
> ... LM_ALERT("core was %sgenerated\n", WCOREDUMP(chld_status) ?  "" : "not
> " );
> #endif
>
> WCOREDUMP() as I understand it allows one to examine the return code of a
> dead child to determine if it returned a core dump. Is further information
> available in such a case, like a kind of errno for such cases?
>
> -- Alex
>
> On 01/03/2015 05:00 PM, Ovidiu Sas wrote:
>
>  Pretty annoying problem :(
>>
>> Here's how I enable core dumps on linux installs in sysctl (reboot
>> required):
>> fs.suid_dumpable = 1
>> kernel.core_pattern = /tmp/core.%e.%u.%t
>> kernel.core_uses_pid = 1
>>
>> On a live system (no need to restart):
>> echo 1 > /proc/sys/fs/suid_dumpable
>> echo 1 > /proc/sys/kernel/core_uses_pid
>> echo "/tmp/core.%e.%u.%t" > /proc/sys/kernel/core_pattern
>>
>> Hope that now you will be able to "enjoy" the core dumps :-/
>>
>> Regards,
>> Ovidiu Sas
>>
>> On Sat, Jan 3, 2015 at 4:38 PM, Alex Balashov <abalashov at evaristesys.com>
>> wrote:
>>
>>> Hi,
>>>
>>> I recently had a simultaneous crash on three production instances of
>>> Kamailio. Unfortunately, I have not been able to get to the bottom of the
>>> issue because Kamailio refused to produce a core dump:
>>>
>>> Dec 30 15:37:29 xxxxxx /usr/local/sbin/kamailio[13743]: ALERT: <core>
>>> [main.c:784]: handle_sigs(): child process 13763 exited by a signal 11
>>> Dec 30 15:37:29 xxxxxx /usr/local/sbin/kamailio[13743]: ALERT: <core>
>>> [main.c:787]: handle_sigs(): core was not generated
>>>
>>> I do not have disable_core_dump=yes (default is no), and checked ulimits
>>> for
>>> limits on core file size. Kamailio was running as root, and there was no
>>> limit.
>>>
>>> What are the other reasons why Kamailio may not generate a core dump in
>>> such
>>> a case?
>>>
>>> Thanks!
>>>
>>> -- Alex
>>>
>>> --
>>> Alex Balashov - Principal
>>> Evariste Systems LLC
>>> 235 E Ponce de Leon Ave
>>> Suite 106
>>> Decatur, GA 30030
>>> United States
>>>
>>> Tel: +1-678-954-0670
>>> Web: http://www.evaristesys.com/, http://www.alexbalashov.com/
>>>
>>> _______________________________________________
>>> SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing list
>>> sr-users at lists.sip-router.org
>>> http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users
>>>
>>
>>
>>
>>
>
> --
> Alex Balashov - Principal
> Evariste Systems LLC
> 235 E Ponce de Leon Ave
> Suite 106
> Decatur, GA 30030
> United States
>
> Tel: +1-678-954-0670
> Web: http://www.evaristesys.com/, http://www.alexbalashov.com/
>
> _______________________________________________
> SIP Express Router (SER) and Kamailio (OpenSER) - sr-users mailing list
> sr-users at lists.sip-router.org
> http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.sip-router.org/pipermail/sr-users/attachments/20150103/285e2c9d/attachment.html>