[OpenSIPS-Users] Strange performance problem

Vlad Paiu vladpaiu at opensips.org
Wed Sep 14 10:49:44 CEST 2011


Hello,

Please check, as Brett said, if your logging is not the issue - if 
syslog is in async mode.
Further more, if logging turns out not to be the problem, you could use 
the time threshold parameters from the core for DNS and overall message 
processing [1], or the time threshold in the DB_MYSQL module, for 
queries [2], to check if DNS or MySQL is causing this performance issues.

[1] http://www.opensips.org/Resources/DocsCoreFcn#toc49
[2] http://www.opensips.org/html/docs/modules/devel/db_mysql.html#id249058

Regards,

Vlad Paiu
OpenSIPS Developer


On 09/14/2011 03:55 AM, Brett Nemeroff wrote:
> 2011/9/13 Alejandro Ríos <alerios at debian.org <mailto:alerios at debian.org>>
>
>     Hi all,
>
>     I'm running a sipp load test to an opensips instance on a 16 core,
>     8GB ram blade server.  It's only 100 simultaneous calls, at 10
>     calls per second, and I'm getting lots of retransmissions from the
>     sipp UAC.
>
>
> I'll give you the same advice I gave someone else just a sec ago. 
> Check your syslog config and see if you are set up for async writes. 
> Turn logging off completely on opensips and see if that changes 
> anything. I've seen bad logging configs bring perfectly capable 
> systems to it's knees.
> -Brett
>
>
> _______________________________________________
> Users mailing list
> Users at lists.opensips.org
> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensips.org/pipermail/users/attachments/20110914/e0747d6a/attachment.htm>


More information about the Users mailing list