Thanks Brett. You were right. syslog was picking up local0. I added '-' before /var/log/messages and that fixed my problem. I can go up to 1000 calls per second with logging.
Thanks again, Ali Pey On Wed, Jun 6, 2012 at 3:16 PM, Brett Nemeroff <[email protected]> wrote: > It's possible that your default route in syslog also is picking up local0. > For good measure I'd also stick a "-" before you /var/log/messages > > If turning messages off fixes your problem, then the problem is in syslog. > > BTW, This is a common problem. I've seen it many times. Sync writes in > syslog will take a 1000+ cps server down to 1 cps any day. > > -Brett > > > On Wed, Jun 6, 2012 at 2:14 PM, Ali Pey <[email protected]> wrote: > >> Yes, as I already mentioned in my email, I did restart syslog. >> >> >> On Wed, Jun 6, 2012 at 3:06 PM, Zahid Mehmood <[email protected]> wrote: >> >>> Did you restart syslog after the change?**** >>> >>> ** ** >>> >>> *From:* [email protected] [mailto: >>> [email protected]] *On Behalf Of *Ali Pey >>> *Sent:* Wednesday, June 06, 2012 3:05 PM >>> *To:* OpenSIPS users mailling list >>> *Subject:* Re: [OpenSIPS-Users] Call per second - OpenSIPS performance** >>> ** >>> >>> ** ** >>> >>> Brett, Duane, Thank you for your replies.**** >>> >>> ** ** >>> >>> So when I disable logging all together I can easily do 1000 call per >>> seconds, but even with asynchronous syslog I can only do 7 or 8 calls per >>> second.**** >>> >>> ** ** >>> >>> For asynchronous syslog I have this in syslog.conf (and have restarted >>> syslog service):**** >>> >>> local0.* -/var/log/opensips.log**** >>> >>> ** ** >>> >>> ** ** >>> >>> On Wed, Jun 6, 2012 at 1:48 PM, Brett Nemeroff <[email protected]> >>> wrote:**** >>> >>> Ensure you are using asynchronous writes with syslog. Disable logging >>> altogether and see if that fixes it. With JUST drouting your server >>> should scream >>> >>> -Brett >>> >>> On Jun 6, 2012, at 12:46 PM, "[email protected]"**** >>> >>> <[email protected]> wrote: >>> >>> > Wow 7 or 8 calls per second is horrible. >>> > >>> > http://www.opensips.org/Resources/PerformanceTests >>> > >>> > >>> > >>> > On , Ali Pey <[email protected]> wrote: >>> > > Hello Everyone, >>> > > >>> > > I have OpenSIPS setup on a fairly powerful server in trunking mode >>> and am using SIPp to do performance testing. With basic call scenarios I >>> can only get 7 or 8 call per second which seems very slow. When I look at >>> the network captures, I see the opensips is running behind which results to >>> re transmissions, etc. >>> > > >>> > > >>> > > >>> > > What call per second number should I expect? Currently I am only >>> using the Dynamic Routing module. >>> > > What can I do to improve the performance? >>> > > >>> > > >>> > > I have increased the number of children process or memory but didn't >>> make a difference. The CPU is not too busy either, it's around 80% idle. >>> > > >>> > > >>> > > >>> > > I am running out of ideas. Anyone has any experience with this that >>> they can share? >>> > > >>> > > >>> > > Regards, >>> > > Ali Pey >>> > > >>> > >**** >>> >>> > _______________________________________________ >>> > Users mailing list >>> > [email protected] >>> > http://lists.opensips.org/cgi-bin/mailman/listinfo/users >>> >>> _______________________________________________ >>> Users mailing list >>> [email protected] >>> http://lists.opensips.org/cgi-bin/mailman/listinfo/users**** >>> >>> ** ** >>> >>> _______________________________________________ >>> Users mailing list >>> [email protected] >>> http://lists.opensips.org/cgi-bin/mailman/listinfo/users >>> >>> >> >> _______________________________________________ >> Users mailing list >> [email protected] >> http://lists.opensips.org/cgi-bin/mailman/listinfo/users >> >> > > _______________________________________________ > Users mailing list > [email protected] > http://lists.opensips.org/cgi-bin/mailman/listinfo/users > >
_______________________________________________ Users mailing list [email protected] http://lists.opensips.org/cgi-bin/mailman/listinfo/users
