[OpenSIPS-Users] Help to analyze a memory dump after crash

[Mariana Arduini]

Hey!

I´ve uploaded the zipped file here:
http://www.4shared.com/zip/IgrX6IQ5/memory_dump.html

I appreciate any help on this!

Thanks,
Mariana.


On Mon, Apr 22, 2013 at 9:38 AM, Mariana Arduini <marianarduini at gmail.com>wrote:

> Hi Razvan!
>
> Thanks for the hint, but the memory dump logged more than 174000 lines.
> The zipped file is 1 MB big (18 MB unzipped) and a free account on pastebin
> won´t let me paste a file bigger than 500 KB.
>
> Is there any email I can sent it privately, or could you give us any
> guidance on how to analyse that info and get to any conclusion?
>
> Thanks and sorry for the huge dump...
>
> Mariana.
>
>
> On Mon, Apr 22, 2013 at 9:08 AM, Răzvan Crainea <razvan at opensips.org>wrote:
>
>> Hi, Mariana!
>>
>> Please upload the files on one of the online file storage services (like
>> [1]), and paste the links here.
>>
>> [1] http://pastebin.com/
>>
>> Best regards,
>>
>> Razvan Crainea
>> OpenSIPS Core Developer
>> http://www.opensips-solutions.**com <http://www.opensips-solutions.com>
>>
>>
>> On 04/22/2013 01:58 PM, Mariana Arduini wrote:
>>
>>> Hello all!
>>>
>>> We are running a suite of automated tests using Opensips 1.8.1.
>>>
>>> Once in a while, we get this crash, apparently related to a call where
>>> we have to update Session Timers. To try to get more info, we did this:
>>>
>>> - started opensips with Valgrind: valgrind --leak-check=full
>>> --show-reachable=yes --tool=memcheck --track-origins=yes
>>> /usr/local/sbin/opensips -P /var/run/opensips.pid -f
>>> /usr/local/etc/opensips/**opensips.cfg -m 64 2> valgrind.output
>>> - in the Makefile.conf, we enabled DEFS+= -DDBG_F_MALLOC and disabled
>>> DEFS+= -DF_MALLOC
>>> - in opensips.cfg file, we added lines: memlog=6 and memdump=1
>>>
>>> At some point in the tests, process 60939 exited with signal 11. A core
>>> was not generated, but we could get some info from Valgrind, and also
>>> the dump in syslog file.
>>>
>>> I´ve attached a file containning the syslog and valgrind report to this
>>> same message 3 days ago, but it was bigger than the max 40KB and I guess
>>> the monitoring blocked it. Please let me know one email address I can
>>> sent it to.
>>>
>>> Hopefully you can help us to detect what may be corrupting the memory.
>>>
>>> Thanks!
>>> Mariana.
>>>
>>>
>>> ______________________________**_________________
>>> Users mailing list
>>> Users at lists.opensips.org
>>> http://lists.opensips.org/cgi-**bin/mailman/listinfo/users<http://lists.opensips.org/cgi-bin/mailman/listinfo/users>
>>>
>>>
>> ______________________________**_________________
>> Users mailing list
>> Users at lists.opensips.org
>> http://lists.opensips.org/cgi-**bin/mailman/listinfo/users<http://lists.opensips.org/cgi-bin/mailman/listinfo/users>
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensips.org/pipermail/users/attachments/20130422/37669eb7/attachment-0001.htm>