compare the quality of the motherboards. try enable-clock-nanosleep=false in switch.conf.xml also try -heavy_timer (last resort) you may want to explore a newer kernel that supports timerfd (centos 6 for instance) On Tue, Mar 27, 2012 at 9:16 AM, Stephen Wilde <wstephen80@gmail.com> wrote:

Yes, it's the original kernel. I start linux without any additional parameter.

I have this problem in 4 of my server and no issue with only 1 server:

Problem with timer_test on:

Server.1, dual Xeon X5660 (12 core in HT = 24), CentOS 5.7, kernel 2.6.18-274.3.1.el5, FS commit 2012-03-26, cpu load of 35% Server.2, dual Xeon X5520 (8 core in HT = 16), CentOS 5.5, kernel 2.6.18-194.17.4.el5, FS commit  2012-03-26, cpu load of 50% Server.3, dual Xeon E5606 (8 core), CentOS 5.7, kernel 2.6.18-274.17.1.el5, FS commit  2012-02-24, cpu load of 35% Server.4, dual Xeon  X5670 (12 core in HT = 24), CentOS 5.7, kernel 2.6.18-274.12.1.el5, FS commit  2012-02-24, cpu load of 25%

No problem on this server:

Server.5, dual Xeon  X5520 (8 core in HT = 16), CentOS 5.5, kernel  2.6.18-194.3.1.el5, FS commit  2011-10-24, cpu load of 50%

Stephen

On Tue, Mar 27, 2012 at 3:48 PM, Peter Olsson <peter.olsson@visionutveckling.se> wrote:

...

Is this the original CentOS kernel? Also, did you set any strange kernel parameters during startup? I think CentOS kernel out-of-the-box usually handles this just fine.

/Peter

Från: freeswitch-users-bounces@lists.freeswitch.org [mailto:freeswitch-users-bounces@lists.freeswitch.org] För Stephen Wilde Skickat: den 27 mars 2012 15:41

Till: FreeSWITCH Users Help Ämne: Re: [Freeswitch-users] Strange "timer_test" result

With "1ms-timer" the behaviour is changed but the "timer_test" remain strange:

2012-03-27 15:36:49.609946 [CONSOLE] mod_commands.c:559 Timer Test: 1 sleep 20 1

2012-03-27 15:36:49.630810 [CONSOLE] mod_commands.c:559 Timer Test: 2 sleep 20 21225

2012-03-27 15:36:49.671927 [CONSOLE] mod_commands.c:559 Timer Test: 3 sleep 20 40716

2012-03-27 15:36:49.671927 [CONSOLE] mod_commands.c:559 Timer Test: 4 sleep 20 29

2012-03-27 15:36:49.690968 [CONSOLE] mod_commands.c:559 Timer Test: 5 sleep 20 19451

2012-03-27 15:36:49.714993 [CONSOLE] mod_commands.c:559 Timer Test: 6 sleep 20 23380

2012-03-27 15:36:49.714993 [CONSOLE] mod_commands.c:559 Timer Test: 7 sleep 20 35

2012-03-27 15:36:49.752814 [CONSOLE] mod_commands.c:559 Timer Test: 8 sleep 20 38474

2012-03-27 15:36:49.752814 [CONSOLE] mod_commands.c:559 Timer Test: 9 sleep 20 41

2012-03-27 15:36:49.776956 [CONSOLE] mod_commands.c:559 Timer Test: 10 sleep 20 23408

2012-03-27 15:36:49.798632 [CONSOLE] mod_commands.c:559 Timer Test: 11 sleep 20 22079

2012-03-27 15:36:49.838054 [CONSOLE] mod_commands.c:559 Timer Test: 12 sleep 20 39021

2012-03-27 15:36:49.838054 [CONSOLE] mod_commands.c:559 Timer Test: 13 sleep 20 35

2012-03-27 15:36:49.857783 [CONSOLE] mod_commands.c:559 Timer Test: 14 sleep 20 19946

2012-03-27 15:36:49.887887 [CONSOLE] mod_commands.c:559 Timer Test: 15 sleep 20 29935

2012-03-27 15:36:49.906809 [CONSOLE] mod_commands.c:559 Timer Test: 16 sleep 20 18774

2012-03-27 15:36:49.946751 [CONSOLE] mod_commands.c:559 Timer Test: 17 sleep 20 40290

2012-03-27 15:36:49.946751 [CONSOLE] mod_commands.c:559 Timer Test: 18 sleep 20 32

2012-03-27 15:36:49.966757 [CONSOLE] mod_commands.c:559 Timer Test: 19 sleep 20 19813

2012-03-27 15:36:50.005781 [CONSOLE] mod_commands.c:559 Timer Test: 20 sleep 20 39099

2012-03-27 15:36:50.005781 [CONSOLE] mod_commands.c:559 Timer Test: 21 sleep 20 24

2012-03-27 15:36:50.023894 [CONSOLE] mod_commands.c:559 Timer Test: 22 sleep 20 18116

2012-03-27 15:36:50.046022 [CONSOLE] mod_commands.c:559 Timer Test: 23 sleep 20 22046

2012-03-27 15:36:50.074716 [CONSOLE] mod_commands.c:559 Timer Test: 24 sleep 20 29054

2012-03-27 15:36:50.074716 [CONSOLE] mod_commands.c:559 Timer Test: 25 sleep 20 25

2012-03-27 15:36:50.111794 [CONSOLE] mod_commands.c:559 Timer Test: 26 sleep 20 37413

2012-03-27 15:36:50.131778 [CONSOLE] mod_commands.c:559 Timer Test: 27 sleep 20 19684

2012-03-27 15:36:50.155846 [CONSOLE] mod_commands.c:559 Timer Test: 28 sleep 20 23480

2012-03-27 15:36:50.155846 [CONSOLE] mod_commands.c:559 Timer Test: 29 sleep 20 37

2012-03-27 15:36:50.179738 [CONSOLE] mod_commands.c:559 Timer Test: 30 sleep 20 23811

2012-03-27 15:36:50.196725 [CONSOLE] mod_commands.c:559 Timer Test: 31 sleep 20 18828

2012-03-27 15:36:50.220604 [CONSOLE] mod_commands.c:559 Timer Test: 32 sleep 20 22205

2012-03-27 15:36:50.250696 [CONSOLE] mod_commands.c:559 Timer Test: 33 sleep 20 30868

2012-03-27 15:36:50.275841 [CONSOLE] mod_commands.c:559 Timer Test: 34 sleep 20 24303

2012-03-27 15:36:50.275841 [CONSOLE] mod_commands.c:559 Timer Test: 35 sleep 20 57

2012-03-27 15:36:50.299676 [CONSOLE] mod_commands.c:559 Timer Test: 36 sleep 20 24607

2012-03-27 15:36:50.319195 [CONSOLE] mod_commands.c:559 Timer Test: 37 sleep 20 18660

2012-03-27 15:36:50.336701 [CONSOLE] mod_commands.c:559 Timer Test: 38 sleep 20 18321

2012-03-27 15:36:50.377677 [CONSOLE] mod_commands.c:559 Timer Test: 39 sleep 20 40551

2012-03-27 15:36:50.377677 [CONSOLE] mod_commands.c:559 Timer Test: 40 sleep 20 33

2012-03-27 15:36:50.398692 [CONSOLE] mod_commands.c:559 Timer Test: 41 sleep 20 20552

2012-03-27 15:36:50.421692 [CONSOLE] mod_commands.c:559 Timer Test: 42 sleep 20 23208

2012-03-27 15:36:50.459684 [CONSOLE] mod_commands.c:559 Timer Test: 43 sleep 20 38668

2012-03-27 15:36:50.459684 [CONSOLE] mod_commands.c:559 Timer Test: 44 sleep 20 34

2012-03-27 15:36:50.478817 [CONSOLE] mod_commands.c:559 Timer Test: 45 sleep 20 18432

2012-03-27 15:36:50.519691 [CONSOLE] mod_commands.c:559 Timer Test: 46 sleep 20 41447

2012-03-27 15:36:50.519691 [CONSOLE] mod_commands.c:559 Timer Test: 47 sleep 20 29

2012-03-27 15:36:50.542590 [CONSOLE] mod_commands.c:559 Timer Test: 48 sleep 20 22699

2012-03-27 15:36:50.552674 [CONSOLE] mod_commands.c:559 Timer Test: 49 sleep 20 10034

2012-03-27 15:36:50.590676 [CONSOLE] mod_commands.c:559 Timer Test: 50 sleep 20 38060

2012-03-27 15:36:50.608681 [CONSOLE] mod_commands.c:559 Timer Test: 51 sleep 20 17448

2012-03-27 15:36:50.625755 [CONSOLE] mod_commands.c:559 Timer Test: 52 sleep 20 17767

2012-03-27 15:36:50.643693 [CONSOLE] mod_commands.c:559 Timer Test: 53 sleep 20 17263

2012-03-27 15:36:50.663677 [CONSOLE] mod_commands.c:559 Timer Test: 54 sleep 20 21429

2012-03-27 15:36:50.682676 [CONSOLE] mod_commands.c:559 Timer Test: 55 sleep 20 18072

2012-03-27 15:36:50.723740 [CONSOLE] mod_commands.c:559 Timer Test: 56 sleep 20 40933

2012-03-27 15:36:50.723740 [CONSOLE] mod_commands.c:559 Timer Test: 57 sleep 20 29

2012-03-27 15:36:50.758800 [CONSOLE] mod_commands.c:559 Timer Test: 58 sleep 20 35950

2012-03-27 15:36:50.758800 [CONSOLE] mod_commands.c:559 Timer Test: 59 sleep 20 30

2012-03-27 15:36:50.779674 [CONSOLE] mod_commands.c:559 Timer Test: 60 sleep 20 19772

The CPU load is low (20%) so what parameter can affect this? The CentOS version? The kernel versione? The FS version? The number of cpu cores?

Stephen

On Mon, Mar 26, 2012 at 4:40 PM, Stephen Wilde <wstephen80@gmail.com> wrote:

Ok, I'll try with this parameter, I think I have to do a Freeswitch restart to make this change...

Stephen

On Mon, Mar 26, 2012 at 4:31 PM, Peter Olsson <peter.olsson@visionutveckling.se> wrote:

You could try to enable the ”old” 1 ms-timer. It will be a little less efficient (a little more CPU load), and I’m not sure it helps, but it’s worth a try.

In switch.conf.xml (under autoload_configs), add this: <param name="1ms-timer" value="true"/> - within the settings-tags.

/Peter

Från: freeswitch-users-bounces@lists.freeswitch.org [mailto:freeswitch-users-bounces@lists.freeswitch.org] För Stephen Wilde Skickat: den 26 mars 2012 16:20

Till: FreeSWITCH Users Help Ämne: Re: [Freeswitch-users] Strange "timer_test" result

Thank you Peter for your reply!

I have already tried the upgrade to latest git, now I'm on FreeSWITCH Version 1.1.beta1 (git-c31a799 2012-03-24 14-11-49 -0700) and the result is the same.

I have tried with few load (during night) and in this case the timer_test is perfect.

Any suggestion?

On Mon, Mar 26, 2012 at 4:08 PM, Peter Olsson <peter.olsson@visionutveckling.se> wrote:

Yes, most likely.

It’s very strange though, that a machine like this gives so poor timing results.

The first thing I would do is to upgrade to latest GIT HEAD. A patch was commited about two weeks ago, that does the calculation for timer_test more properly (so time for logging is not calculated).

Have you tried to do the same test with no load at all?

/Peter

Från: freeswitch-users-bounces@lists.freeswitch.org [mailto:freeswitch-users-bounces@lists.freeswitch.org] För Stephen Wilde Skickat: den 26 mars 2012 15:52 Till: FreeSWITCH Users Help Ämne: Re: [Freeswitch-users] Strange "timer_test" result

Can this issue affect the voice quality?

On Fri, Mar 23, 2012 at 7:46 PM, Stephen Wilde <wstephen80@gmail.com> wrote:

Real hardware, a dedicate server with 2 Xeon X5670 (a total 24 core each one with 12Mb cache at 2.93GHz) that is running at 20% - 25% of load. OS is CentOS 5.7 64bit and FS is (git-0626c89 2012-02-29 14-45-39 -0600)

On Fri, Mar 23, 2012 at 6:33 PM, Brian Foster <bdfoster@endigotech.com> wrote:

Is this on virtualized or real hardware?

-BDF

On Fri, Mar 23, 2012 at 1:31 PM, Stephen Wilde <wstephen80@gmail.com> wrote:

I have run a "timer_test" in a dedicated FS server and I see strange result: it's normal?

Stephen

freeswitch@internal> timer_test 20 40

Avg: 19.866ms Total Time: 795.880ms

2012-03-23 18:25:54.157822 [CONSOLE] mod_commands.c:549 Timer Test: samplecount after init: 1

2012-03-23 18:25:54.178818 [CONSOLE] mod_commands.c:554 Timer Test: samplecount after first step: 2

2012-03-23 18:25:54.199833 [CONSOLE] mod_commands.c:563 Timer Test: 1 sleep 20 19568

2012-03-23 18:25:54.231890 [CONSOLE] mod_commands.c:563 Timer Test: 2 sleep 20 38231

2012-03-23 18:25:54.252816 [CONSOLE] mod_commands.c:563 Timer Test: 3 sleep 20 18847

2012-03-23 18:25:54.262818 [CONSOLE] mod_commands.c:563 Timer Test: 4 sleep 20 13982

2012-03-23 18:25:54.305813 [CONSOLE] mod_commands.c:563 Timer Test: 5 sleep 20 34793

2012-03-23 18:25:54.305813 [CONSOLE] mod_commands.c:563 Timer Test: 6 sleep 20 2

2012-03-23 18:25:54.326811 [CONSOLE] mod_commands.c:563 Timer Test: 7 sleep 20 23166

2012-03-23 18:25:54.347807 [CONSOLE] mod_commands.c:563 Timer Test: 8 sleep 20 16957

2012-03-23 18:25:54.357811 [CONSOLE] mod_commands.c:563 Timer Test: 9 sleep 20 17643

2012-03-23 18:25:54.378828 [CONSOLE] mod_commands.c:563 Timer Test: 10 sleep 20 18786

2012-03-23 18:25:54.399856 [CONSOLE] mod_commands.c:563 Timer Test: 11 sleep 20 25100

2012-03-23 18:25:54.420855 [CONSOLE] mod_commands.c:563 Timer Test: 12 sleep 20 18552

2012-03-23 18:25:54.441855 [CONSOLE] mod_commands.c:563 Timer Test: 13 sleep 20 18815

2012-03-23 18:25:54.462798 [CONSOLE] mod_commands.c:563 Timer Test: 14 sleep 20 19464

2012-03-23 18:25:54.484300 [CONSOLE] mod_commands.c:563 Timer Test: 15 sleep 20 22012

2012-03-23 18:25:54.494804 [CONSOLE] mod_commands.c:563 Timer Test: 16 sleep 20 13980

2012-03-23 18:25:54.515793 [CONSOLE] mod_commands.c:563 Timer Test: 17 sleep 20 19065

2012-03-23 18:25:54.556792 [CONSOLE] mod_commands.c:563 Timer Test: 18 sleep 20 39585

2012-03-23 18:25:54.556792 [CONSOLE] mod_commands.c:563 Timer Test: 19 sleep 20 2

2012-03-23 18:25:54.577790 [CONSOLE] mod_commands.c:563 Timer Test: 20 sleep 20 26255

2012-03-23 18:25:54.598790 [CONSOLE] mod_commands.c:563 Timer Test: 21 sleep 20 17872

2012-03-23 18:25:54.630794 [CONSOLE] mod_commands.c:563 Timer Test: 22 sleep 20 32191

2012-03-23 18:25:54.651790 [CONSOLE] mod_commands.c:563 Timer Test: 23 sleep 20 22634

2012-03-23 18:25:54.672788 [CONSOLE] mod_commands.c:563 Timer Test: 24 sleep 20 15483

2012-03-23 18:25:54.693783 [CONSOLE] mod_commands.c:563 Timer Test: 25 sleep 20 22813

2012-03-23 18:25:54.714783 [CONSOLE] mod_commands.c:563 Timer Test: 26 sleep 20 17099

2012-03-23 18:25:54.714783 [CONSOLE] mod_commands.c:563 Timer Test: 27 sleep 20 1

2012-03-23 18:25:54.734832 [CONSOLE] mod_commands.c:563 Timer Test: 28 sleep 20 29108

2012-03-23 18:25:54.755836 [CONSOLE] mod_commands.c:563 Timer Test: 29 sleep 20 11492

2012-03-23 18:25:54.776830 [CONSOLE] mod_commands.c:563 Timer Test: 30 sleep 20 20855

2012-03-23 18:25:54.808851 [CONSOLE] mod_commands.c:563 Timer Test: 31 sleep 20 32579

2012-03-23 18:25:54.818833 [CONSOLE] mod_commands.c:563 Timer Test: 32 sleep 20 18173

2012-03-23 18:25:54.850828 [CONSOLE] mod_commands.c:563 Timer Test: 33 sleep 20 22666

2012-03-23 18:25:54.871855 [CONSOLE] mod_commands.c:563 Timer Test: 34 sleep 20 23792

2012-03-23 18:25:54.892823 [CONSOLE] mod_commands.c:563 Timer Test: 35 sleep 20 26158

2012-03-23 18:25:54.913823 [CONSOLE] mod_commands.c:563 Timer Test: 36 sleep 20 13080

2012-03-23 18:25:54.934844 [CONSOLE] mod_commands.c:563 Timer Test: 37 sleep 20 24609

2012-03-23 18:25:54.934844 [CONSOLE] mod_commands.c:563 Timer Test: 38 sleep 20 1

2012-03-23 18:25:54.954766 [CONSOLE] mod_commands.c:563 Timer Test: 39 sleep 20 19413

2012-03-23 18:25:54.975764 [CONSOLE] mod_commands.c:563 Timer Test: 40 sleep 20 19820

freeswitch@internal> status

UP 0 years, 0 days, 16 hours, 10 minutes, 25 seconds, 539 milliseconds, 696 microseconds

FreeSWITCH is ready

955611 session(s) since startup

2192 session(s) 0/50

6000 session(s) max

min idle cpu 0.00/74.00

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

-- Brian D. Foster Endigo Computer LLC Email: bdfoster@endigotech.com Phone: 317-800-7876 Indianapolis, Indiana, USA

This message contains confidential information and is intended for those listed in the "To:", "CC:", and/or "BCC:" fields of the message header. If you are not the intended recipient you are notified that disclosing, copying, distributing or taking any action in reliance on the contents of this information is strictly prohibited. E-mail transmission cannot be guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. The sender therefore does not accept liability for any errors or omissions in the contents of this message, which arise as a result of e-mail transmission. If verification is required please request a hard-copy version.

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

!DSPAM:4f71c23b32762536418862!

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

-- Anthony Minessale II FreeSWITCH http://www.freeswitch.org/ ClueCon http://www.cluecon.com/ Twitter: http://twitter.com/FreeSWITCH_wire AIM: anthm MSN:anthony_minessale@hotmail.com GTALK/JABBER/PAYPAL:anthony.minessale@gmail.com IRC: irc.freenode.net #freeswitch FreeSWITCH Developer Conference sip:888@conference.freeswitch.org googletalk:conf+888@conference.freeswitch.org pstn:+19193869900

I don't think the timing code has changes since that october version really. The only other possibility is maybe something else on the box is using up the cpu. When I run the test on my box circa 2008 with centos 5.7 i get a completely clean score.

From my experience, the only boxes that act bad in this case are one of the following:

32 bit linux on 64 bit hw bad motherborard (may be fixed with the disable nanosleep) overloaded box freeswitch@DeathSTAR> timer_test 2012-03-27 10:41:33.788056 [CONSOLE] mod_commands.c:559 Timer Test: 1 sleep 20 19942 2012-03-27 10:41:33.808060 [CONSOLE] mod_commands.c:559 Timer Test: 2 sleep 20 20004 2012-03-27 10:41:33.828064 [CONSOLE] mod_commands.c:559 Timer Test: 3 sleep 20 20004 2012-03-27 10:41:33.848068 [CONSOLE] mod_commands.c:559 Timer Test: 4 sleep 20 20004 2012-03-27 10:41:33.868072 [CONSOLE] mod_commands.c:559 Timer Test: 5 sleep 20 20002 2012-03-27 10:41:33.888076 [CONSOLE] mod_commands.c:559 Timer Test: 6 sleep 20 20009 2012-03-27 10:41:33.908079 [CONSOLE] mod_commands.c:559 Timer Test: 7 sleep 20 20011 2012-03-27 10:41:33.928088 [CONSOLE] mod_commands.c:559 Timer Test: 8 sleep 20 19998 2012-03-27 10:41:33.948088 [CONSOLE] mod_commands.c:559 Timer Test: 9 sleep 20 20000 2012-03-27 10:41:33.968091 [CONSOLE] mod_commands.c:559 Timer Test: 10 sleep 20 20001 2012-03-27 10:41:33.988098 [CONSOLE] mod_commands.c:559 Timer Test: 11 sleep 20 20003 2012-03-27 10:41:34.008098 [CONSOLE] mod_commands.c:559 Timer Test: 12 sleep 20 20001 2012-03-27 10:41:34.028102 [CONSOLE] mod_commands.c:559 Timer Test: 13 sleep 20 20003 2012-03-27 10:41:34.048106 [CONSOLE] mod_commands.c:559 Timer Test: 14 sleep 20 20004 2012-03-27 10:41:34.068110 [CONSOLE] mod_commands.c:559 Timer Test: 15 sleep 20 20005 2012-03-27 10:41:34.088114 [CONSOLE] mod_commands.c:559 Timer Test: 16 sleep 20 20004 2012-03-27 10:41:34.108118 [CONSOLE] mod_commands.c:559 Timer Test: 17 sleep 20 20004 2012-03-27 10:41:34.128122 [CONSOLE] mod_commands.c:559 Timer Test: 18 sleep 20 20003 2012-03-27 10:41:34.148126 [CONSOLE] mod_commands.c:559 Timer Test: 19 sleep 20 20004 2012-03-27 10:41:34.168129 [CONSOLE] mod_commands.c:559 Timer Test: 20 sleep 20 20003 2012-03-27 10:41:34.188133 [CONSOLE] mod_commands.c:559 Timer Test: 21 sleep 20 20005 2012-03-27 10:41:34.208137 [CONSOLE] mod_commands.c:559 Timer Test: 22 sleep 20 20003 2012-03-27 10:41:34.228141 [CONSOLE] mod_commands.c:559 Timer Test: 23 sleep 20 20004 2012-03-27 10:41:34.248145 [CONSOLE] mod_commands.c:559 Timer Test: 24 sleep 20 20004 2012-03-27 10:41:34.268148 [CONSOLE] mod_commands.c:559 Timer Test: 25 sleep 20 20004 2012-03-27 10:41:34.288152 [CONSOLE] mod_commands.c:559 Timer Test: 26 sleep 20 20003 2012-03-27 10:41:34.308156 [CONSOLE] mod_commands.c:559 Timer Test: 27 sleep 20 20005 2012-03-27 10:41:34.328160 [CONSOLE] mod_commands.c:559 Timer Test: 28 sleep 20 20001 2012-03-27 10:41:34.348164 [CONSOLE] mod_commands.c:559 Timer Test: 29 sleep 20 20003 2012-03-27 10:41:34.368167 [CONSOLE] mod_commands.c:559 Timer Test: 30 sleep 20 20004 2012-03-27 10:41:34.388171 [CONSOLE] mod_commands.c:559 Timer Test: 31 sleep 20 20004 2012-03-27 10:41:34.408175 [CONSOLE] mod_commands.c:559 Timer Test: 32 sleep 20 20003 2012-03-27 10:41:34.428179 [CONSOLE] mod_commands.c:559 Timer Test: 33 sleep 20 20004 2012-03-27 10:41:34.448183 [CONSOLE] mod_commands.c:559 Timer Test: 34 sleep 20 20004 2012-03-27 10:41:34.468186 [CONSOLE] mod_commands.c:559 Timer Test: 35 sleep 20 20004 2012-03-27 10:41:34.488190 [CONSOLE] mod_commands.c:559 Timer Test: 36 sleep 20 20004 2012-03-27 10:41:34.508194 [CONSOLE] mod_commands.c:559 Timer Test: 37 sleep 20 20004 2012-03-27 10:41:34.528198 [CONSOLE] mod_commands.c:559 Timer Test: 38 sleep 20 20004 2012-03-27 10:41:34.548202 [CONSOLE] mod_commands.c:559 Timer Test: 39 sleep 20 20003 2012-03-27 10:41:34.568205 [CONSOLE] mod_commands.c:559 Timer Test: 40 sleep 20 20004 2012-03-27 10:41:34.588209 [CONSOLE] mod_commands.c:559 Timer Test: 41 sleep 20 20005 2012-03-27 10:41:34.608213 [CONSOLE] mod_commands.c:559 Timer Test: 42 sleep 20 20003 2012-03-27 10:41:34.628217 [CONSOLE] mod_commands.c:559 Timer Test: 43 sleep 20 20004 2012-03-27 10:41:34.648221 [CONSOLE] mod_commands.c:559 Timer Test: 44 sleep 20 20004 2012-03-27 10:41:34.668225 [CONSOLE] mod_commands.c:559 Timer Test: 45 sleep 20 20004 2012-03-27 10:41:34.688228 [CONSOLE] mod_commands.c:559 Timer Test: 46 sleep 20 20003 2012-03-27 10:41:34.708339 [CONSOLE] mod_commands.c:559 Timer Test: 47 sleep 20 20107 2012-03-27 10:41:34.728236 [CONSOLE] mod_commands.c:559 Timer Test: 48 sleep 20 19902 2012-03-27 10:41:34.748240 [CONSOLE] mod_commands.c:559 Timer Test: 49 sleep 20 20024 2012-03-27 10:41:34.768244 [CONSOLE] mod_commands.c:559 Timer Test: 50 sleep 20 20043 Avg: 20.003ms Total Time: 1000.182ms On Tue, Mar 27, 2012 at 10:37 AM, Stephen Wilde <wstephen80@gmail.com> wrote:

Thank you for your reply. The motherboard cannot be the issue because 2 of these servers (server.2 and server.5) are exactly the same hardware (HP DL380 G6). I'll try with enable-clock-nanosleep=false and I'll try also with -heavy_timer cmd line parameter. More difficult to test timerfd (these server are in production and it's not simple to do this upgrade). In the working server is running a very old FS version (2011-10-24), I'll try also to run this version in a not working server.

Stephen

On Tue, Mar 27, 2012 at 5:21 PM, Anthony Minessale <anthony.minessale@gmail.com> wrote:

...

compare the quality of the motherboards.

try enable-clock-nanosleep=false in switch.conf.xml

also try -heavy_timer (last resort)

you may want to explore a newer kernel that supports timerfd (centos 6 for instance)

On Tue, Mar 27, 2012 at 9:16 AM, Stephen Wilde <wstephen80@gmail.com> wrote:

...

Yes, it's the original kernel. I start linux without any additional parameter.

I have this problem in 4 of my server and no issue with only 1 server:

Problem with timer_test on:

Server.1, dual Xeon X5660 (12 core in HT = 24), CentOS 5.7, kernel 2.6.18-274.3.1.el5, FS commit 2012-03-26, cpu load of 35% Server.2, dual Xeon X5520 (8 core in HT = 16), CentOS 5.5, kernel 2.6.18-194.17.4.el5, FS commit  2012-03-26, cpu load of 50% Server.3, dual Xeon E5606 (8 core), CentOS 5.7, kernel 2.6.18-274.17.1.el5, FS commit  2012-02-24, cpu load of 35% Server.4, dual Xeon  X5670 (12 core in HT = 24), CentOS 5.7, kernel 2.6.18-274.12.1.el5, FS commit  2012-02-24, cpu load of 25%

No problem on this server:

Server.5, dual Xeon  X5520 (8 core in HT = 16), CentOS 5.5, kernel  2.6.18-194.3.1.el5, FS commit  2011-10-24, cpu load of 50%

Stephen

On Tue, Mar 27, 2012 at 3:48 PM, Peter Olsson <peter.olsson@visionutveckling.se> wrote:

...

Is this the original CentOS kernel? Also, did you set any strange kernel parameters during startup? I think CentOS kernel out-of-the-box usually handles this just fine.

/Peter

Från: freeswitch-users-bounces@lists.freeswitch.org [mailto:freeswitch-users-bounces@lists.freeswitch.org] För Stephen Wilde Skickat: den 27 mars 2012 15:41

Till: FreeSWITCH Users Help Ämne: Re: [Freeswitch-users] Strange "timer_test" result

With "1ms-timer" the behaviour is changed but the "timer_test" remain strange:

2012-03-27 15:36:49.609946 [CONSOLE] mod_commands.c:559 Timer Test: 1 sleep 20 1

2012-03-27 15:36:49.630810 [CONSOLE] mod_commands.c:559 Timer Test: 2 sleep 20 21225

2012-03-27 15:36:49.671927 [CONSOLE] mod_commands.c:559 Timer Test: 3 sleep 20 40716

2012-03-27 15:36:49.671927 [CONSOLE] mod_commands.c:559 Timer Test: 4 sleep 20 29

2012-03-27 15:36:49.690968 [CONSOLE] mod_commands.c:559 Timer Test: 5 sleep 20 19451

2012-03-27 15:36:49.714993 [CONSOLE] mod_commands.c:559 Timer Test: 6 sleep 20 23380

2012-03-27 15:36:49.714993 [CONSOLE] mod_commands.c:559 Timer Test: 7 sleep 20 35

2012-03-27 15:36:49.752814 [CONSOLE] mod_commands.c:559 Timer Test: 8 sleep 20 38474

2012-03-27 15:36:49.752814 [CONSOLE] mod_commands.c:559 Timer Test: 9 sleep 20 41

2012-03-27 15:36:49.776956 [CONSOLE] mod_commands.c:559 Timer Test: 10 sleep 20 23408

2012-03-27 15:36:49.798632 [CONSOLE] mod_commands.c:559 Timer Test: 11 sleep 20 22079

2012-03-27 15:36:49.838054 [CONSOLE] mod_commands.c:559 Timer Test: 12 sleep 20 39021

2012-03-27 15:36:49.838054 [CONSOLE] mod_commands.c:559 Timer Test: 13 sleep 20 35

2012-03-27 15:36:49.857783 [CONSOLE] mod_commands.c:559 Timer Test: 14 sleep 20 19946

2012-03-27 15:36:49.887887 [CONSOLE] mod_commands.c:559 Timer Test: 15 sleep 20 29935

2012-03-27 15:36:49.906809 [CONSOLE] mod_commands.c:559 Timer Test: 16 sleep 20 18774

2012-03-27 15:36:49.946751 [CONSOLE] mod_commands.c:559 Timer Test: 17 sleep 20 40290

2012-03-27 15:36:49.946751 [CONSOLE] mod_commands.c:559 Timer Test: 18 sleep 20 32

2012-03-27 15:36:49.966757 [CONSOLE] mod_commands.c:559 Timer Test: 19 sleep 20 19813

2012-03-27 15:36:50.005781 [CONSOLE] mod_commands.c:559 Timer Test: 20 sleep 20 39099

2012-03-27 15:36:50.005781 [CONSOLE] mod_commands.c:559 Timer Test: 21 sleep 20 24

2012-03-27 15:36:50.023894 [CONSOLE] mod_commands.c:559 Timer Test: 22 sleep 20 18116

2012-03-27 15:36:50.046022 [CONSOLE] mod_commands.c:559 Timer Test: 23 sleep 20 22046

2012-03-27 15:36:50.074716 [CONSOLE] mod_commands.c:559 Timer Test: 24 sleep 20 29054

2012-03-27 15:36:50.074716 [CONSOLE] mod_commands.c:559 Timer Test: 25 sleep 20 25

2012-03-27 15:36:50.111794 [CONSOLE] mod_commands.c:559 Timer Test: 26 sleep 20 37413

2012-03-27 15:36:50.131778 [CONSOLE] mod_commands.c:559 Timer Test: 27 sleep 20 19684

2012-03-27 15:36:50.155846 [CONSOLE] mod_commands.c:559 Timer Test: 28 sleep 20 23480

2012-03-27 15:36:50.155846 [CONSOLE] mod_commands.c:559 Timer Test: 29 sleep 20 37

2012-03-27 15:36:50.179738 [CONSOLE] mod_commands.c:559 Timer Test: 30 sleep 20 23811

2012-03-27 15:36:50.196725 [CONSOLE] mod_commands.c:559 Timer Test: 31 sleep 20 18828

2012-03-27 15:36:50.220604 [CONSOLE] mod_commands.c:559 Timer Test: 32 sleep 20 22205

2012-03-27 15:36:50.250696 [CONSOLE] mod_commands.c:559 Timer Test: 33 sleep 20 30868

2012-03-27 15:36:50.275841 [CONSOLE] mod_commands.c:559 Timer Test: 34 sleep 20 24303

2012-03-27 15:36:50.275841 [CONSOLE] mod_commands.c:559 Timer Test: 35 sleep 20 57

2012-03-27 15:36:50.299676 [CONSOLE] mod_commands.c:559 Timer Test: 36 sleep 20 24607

2012-03-27 15:36:50.319195 [CONSOLE] mod_commands.c:559 Timer Test: 37 sleep 20 18660

2012-03-27 15:36:50.336701 [CONSOLE] mod_commands.c:559 Timer Test: 38 sleep 20 18321

2012-03-27 15:36:50.377677 [CONSOLE] mod_commands.c:559 Timer Test: 39 sleep 20 40551

2012-03-27 15:36:50.377677 [CONSOLE] mod_commands.c:559 Timer Test: 40 sleep 20 33

2012-03-27 15:36:50.398692 [CONSOLE] mod_commands.c:559 Timer Test: 41 sleep 20 20552

2012-03-27 15:36:50.421692 [CONSOLE] mod_commands.c:559 Timer Test: 42 sleep 20 23208

2012-03-27 15:36:50.459684 [CONSOLE] mod_commands.c:559 Timer Test: 43 sleep 20 38668

2012-03-27 15:36:50.459684 [CONSOLE] mod_commands.c:559 Timer Test: 44 sleep 20 34

2012-03-27 15:36:50.478817 [CONSOLE] mod_commands.c:559 Timer Test: 45 sleep 20 18432

2012-03-27 15:36:50.519691 [CONSOLE] mod_commands.c:559 Timer Test: 46 sleep 20 41447

2012-03-27 15:36:50.519691 [CONSOLE] mod_commands.c:559 Timer Test: 47 sleep 20 29

2012-03-27 15:36:50.542590 [CONSOLE] mod_commands.c:559 Timer Test: 48 sleep 20 22699

2012-03-27 15:36:50.552674 [CONSOLE] mod_commands.c:559 Timer Test: 49 sleep 20 10034

2012-03-27 15:36:50.590676 [CONSOLE] mod_commands.c:559 Timer Test: 50 sleep 20 38060

2012-03-27 15:36:50.608681 [CONSOLE] mod_commands.c:559 Timer Test: 51 sleep 20 17448

2012-03-27 15:36:50.625755 [CONSOLE] mod_commands.c:559 Timer Test: 52 sleep 20 17767

2012-03-27 15:36:50.643693 [CONSOLE] mod_commands.c:559 Timer Test: 53 sleep 20 17263

2012-03-27 15:36:50.663677 [CONSOLE] mod_commands.c:559 Timer Test: 54 sleep 20 21429

2012-03-27 15:36:50.682676 [CONSOLE] mod_commands.c:559 Timer Test: 55 sleep 20 18072

2012-03-27 15:36:50.723740 [CONSOLE] mod_commands.c:559 Timer Test: 56 sleep 20 40933

2012-03-27 15:36:50.723740 [CONSOLE] mod_commands.c:559 Timer Test: 57 sleep 20 29

2012-03-27 15:36:50.758800 [CONSOLE] mod_commands.c:559 Timer Test: 58 sleep 20 35950

2012-03-27 15:36:50.758800 [CONSOLE] mod_commands.c:559 Timer Test: 59 sleep 20 30

2012-03-27 15:36:50.779674 [CONSOLE] mod_commands.c:559 Timer Test: 60 sleep 20 19772

The CPU load is low (20%) so what parameter can affect this? The CentOS version? The kernel versione? The FS version? The number of cpu cores?

Stephen

On Mon, Mar 26, 2012 at 4:40 PM, Stephen Wilde <wstephen80@gmail.com> wrote:

Ok, I'll try with this parameter, I think I have to do a Freeswitch restart to make this change...

Stephen

On Mon, Mar 26, 2012 at 4:31 PM, Peter Olsson <peter.olsson@visionutveckling.se> wrote:

You could try to enable the ”old” 1 ms-timer. It will be a little less efficient (a little more CPU load), and I’m not sure it helps, but it’s worth a try.

In switch.conf.xml (under autoload_configs), add this: <param name="1ms-timer" value="true"/> - within the settings-tags.

/Peter

Från: freeswitch-users-bounces@lists.freeswitch.org [mailto:freeswitch-users-bounces@lists.freeswitch.org] För Stephen Wilde Skickat: den 26 mars 2012 16:20

Till: FreeSWITCH Users Help Ämne: Re: [Freeswitch-users] Strange "timer_test" result

Thank you Peter for your reply!

I have already tried the upgrade to latest git, now I'm on FreeSWITCH Version 1.1.beta1 (git-c31a799 2012-03-24 14-11-49 -0700) and the result is the same.

I have tried with few load (during night) and in this case the timer_test is perfect.

Any suggestion?

On Mon, Mar 26, 2012 at 4:08 PM, Peter Olsson <peter.olsson@visionutveckling.se> wrote:

Yes, most likely.

It’s very strange though, that a machine like this gives so poor timing results.

The first thing I would do is to upgrade to latest GIT HEAD. A patch was commited about two weeks ago, that does the calculation for timer_test more properly (so time for logging is not calculated).

Have you tried to do the same test with no load at all?

/Peter

Från: freeswitch-users-bounces@lists.freeswitch.org [mailto:freeswitch-users-bounces@lists.freeswitch.org] För Stephen Wilde Skickat: den 26 mars 2012 15:52 Till: FreeSWITCH Users Help Ämne: Re: [Freeswitch-users] Strange "timer_test" result

Can this issue affect the voice quality?

On Fri, Mar 23, 2012 at 7:46 PM, Stephen Wilde <wstephen80@gmail.com> wrote:

Real hardware, a dedicate server with 2 Xeon X5670 (a total 24 core each one with 12Mb cache at 2.93GHz) that is running at 20% - 25% of load. OS is CentOS 5.7 64bit and FS is (git-0626c89 2012-02-29 14-45-39 -0600)

On Fri, Mar 23, 2012 at 6:33 PM, Brian Foster <bdfoster@endigotech.com> wrote:

Is this on virtualized or real hardware?

-BDF

On Fri, Mar 23, 2012 at 1:31 PM, Stephen Wilde <wstephen80@gmail.com> wrote:

I have run a "timer_test" in a dedicated FS server and I see strange result: it's normal?

Stephen

freeswitch@internal> timer_test 20 40

Avg: 19.866ms Total Time: 795.880ms

2012-03-23 18:25:54.157822 [CONSOLE] mod_commands.c:549 Timer Test: samplecount after init: 1

2012-03-23 18:25:54.178818 [CONSOLE] mod_commands.c:554 Timer Test: samplecount after first step: 2

2012-03-23 18:25:54.199833 [CONSOLE] mod_commands.c:563 Timer Test: 1 sleep 20 19568

2012-03-23 18:25:54.231890 [CONSOLE] mod_commands.c:563 Timer Test: 2 sleep 20 38231

2012-03-23 18:25:54.252816 [CONSOLE] mod_commands.c:563 Timer Test: 3 sleep 20 18847

2012-03-23 18:25:54.262818 [CONSOLE] mod_commands.c:563 Timer Test: 4 sleep 20 13982

2012-03-23 18:25:54.305813 [CONSOLE] mod_commands.c:563 Timer Test: 5 sleep 20 34793

2012-03-23 18:25:54.305813 [CONSOLE] mod_commands.c:563 Timer Test: 6 sleep 20 2

2012-03-23 18:25:54.326811 [CONSOLE] mod_commands.c:563 Timer Test: 7 sleep 20 23166

2012-03-23 18:25:54.347807 [CONSOLE] mod_commands.c:563 Timer Test: 8 sleep 20 16957

2012-03-23 18:25:54.357811 [CONSOLE] mod_commands.c:563 Timer Test: 9 sleep 20 17643

2012-03-23 18:25:54.378828 [CONSOLE] mod_commands.c:563 Timer Test: 10 sleep 20 18786

2012-03-23 18:25:54.399856 [CONSOLE] mod_commands.c:563 Timer Test: 11 sleep 20 25100

2012-03-23 18:25:54.420855 [CONSOLE] mod_commands.c:563 Timer Test: 12 sleep 20 18552

2012-03-23 18:25:54.441855 [CONSOLE] mod_commands.c:563 Timer Test: 13 sleep 20 18815

2012-03-23 18:25:54.462798 [CONSOLE] mod_commands.c:563 Timer Test: 14 sleep 20 19464

2012-03-23 18:25:54.484300 [CONSOLE] mod_commands.c:563 Timer Test: 15 sleep 20 22012

2012-03-23 18:25:54.494804 [CONSOLE] mod_commands.c:563 Timer Test: 16 sleep 20 13980

2012-03-23 18:25:54.515793 [CONSOLE] mod_commands.c:563 Timer Test: 17 sleep 20 19065

2012-03-23 18:25:54.556792 [CONSOLE] mod_commands.c:563 Timer Test: 18 sleep 20 39585

2012-03-23 18:25:54.556792 [CONSOLE] mod_commands.c:563 Timer Test: 19 sleep 20 2

2012-03-23 18:25:54.577790 [CONSOLE] mod_commands.c:563 Timer Test: 20 sleep 20 26255

2012-03-23 18:25:54.598790 [CONSOLE] mod_commands.c:563 Timer Test: 21 sleep 20 17872

2012-03-23 18:25:54.630794 [CONSOLE] mod_commands.c:563 Timer Test: 22 sleep 20 32191

2012-03-23 18:25:54.651790 [CONSOLE] mod_commands.c:563 Timer Test: 23 sleep 20 22634

2012-03-23 18:25:54.672788 [CONSOLE] mod_commands.c:563 Timer Test: 24 sleep 20 15483

2012-03-23 18:25:54.693783 [CONSOLE] mod_commands.c:563 Timer Test: 25 sleep 20 22813

2012-03-23 18:25:54.714783 [CONSOLE] mod_commands.c:563 Timer Test: 26 sleep 20 17099

2012-03-23 18:25:54.714783 [CONSOLE] mod_commands.c:563 Timer Test: 27 sleep 20 1

2012-03-23 18:25:54.734832 [CONSOLE] mod_commands.c:563 Timer Test: 28 sleep 20 29108

2012-03-23 18:25:54.755836 [CONSOLE] mod_commands.c:563 Timer Test: 29 sleep 20 11492

2012-03-23 18:25:54.776830 [CONSOLE] mod_commands.c:563 Timer Test: 30 sleep 20 20855

2012-03-23 18:25:54.808851 [CONSOLE] mod_commands.c:563 Timer Test: 31 sleep 20 32579

2012-03-23 18:25:54.818833 [CONSOLE] mod_commands.c:563 Timer Test: 32 sleep 20 18173

2012-03-23 18:25:54.850828 [CONSOLE] mod_commands.c:563 Timer Test: 33 sleep 20 22666

2012-03-23 18:25:54.871855 [CONSOLE] mod_commands.c:563 Timer Test: 34 sleep 20 23792

2012-03-23 18:25:54.892823 [CONSOLE] mod_commands.c:563 Timer Test: 35 sleep 20 26158

2012-03-23 18:25:54.913823 [CONSOLE] mod_commands.c:563 Timer Test: 36 sleep 20 13080

2012-03-23 18:25:54.934844 [CONSOLE] mod_commands.c:563 Timer Test: 37 sleep 20 24609

2012-03-23 18:25:54.934844 [CONSOLE] mod_commands.c:563 Timer Test: 38 sleep 20 1

2012-03-23 18:25:54.954766 [CONSOLE] mod_commands.c:563 Timer Test: 39 sleep 20 19413

2012-03-23 18:25:54.975764 [CONSOLE] mod_commands.c:563 Timer Test: 40 sleep 20 19820

freeswitch@internal> status

UP 0 years, 0 days, 16 hours, 10 minutes, 25 seconds, 539 milliseconds, 696 microseconds

FreeSWITCH is ready

955611 session(s) since startup

2192 session(s) 0/50

6000 session(s) max

min idle cpu 0.00/74.00

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users

UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

-- Brian D. Foster Endigo Computer LLC Email: bdfoster@endigotech.com Phone: 317-800-7876 Indianapolis, Indiana, USA

This message contains confidential information and is intended for those listed in the "To:", "CC:", and/or "BCC:" fields of the message header. If you are not the intended recipient you are notified that disclosing, copying, distributing or taking any action in reliance on the contents of this information is strictly prohibited. E-mail transmission cannot be guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. The sender therefore does not accept liability for any errors or omissions in the contents of this message, which arise as a result of e-mail transmission. If verification is required please request a hard-copy version.

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users

UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users

UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users

UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

!DSPAM:4f71c23b32762536418862!

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users

UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

-- Anthony Minessale II

FreeSWITCH http://www.freeswitch.org/ ClueCon http://www.cluecon.com/ Twitter: http://twitter.com/FreeSWITCH_wire

AIM: anthm MSN:anthony_minessale@hotmail.com GTALK/JABBER/PAYPAL:anthony.minessale@gmail.com IRC: irc.freenode.net #freeswitch

FreeSWITCH Developer Conference sip:888@conference.freeswitch.org googletalk:conf+888@conference.freeswitch.org pstn:+19193869900

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

-- Anthony Minessale II FreeSWITCH http://www.freeswitch.org/ ClueCon http://www.cluecon.com/ Twitter: http://twitter.com/FreeSWITCH_wire AIM: anthm MSN:anthony_minessale@hotmail.com GTALK/JABBER/PAYPAL:anthony.minessale@gmail.com IRC: irc.freenode.net #freeswitch FreeSWITCH Developer Conference sip:888@conference.freeswitch.org googletalk:conf+888@conference.freeswitch.org pstn:+19193869900

You can also try mod_posix_timer if you have it built. My CentOS 5.4 server performs better under load using it with heavy-timer vs using the default timer. "timer_test 20 50 posix" On Tue, Mar 27, 2012 at 12:07 PM, Stephen Wilde <wstephen80@gmail.com>wrote:

Ok, I'll try disabling the nanosleep.

In server.2 the current load of the box is 35% and after FreeSwitch (at 750% of cpu), in top there is Sangoma Transcoding server (4% of cpu).

The result of "timer_test 20 50" is:

http://pastebin.freeswitch.org/18756

Here the result of "timer_test 10 50" :

http://pastebin.freeswitch.org/18757

It's normal that only the timer_test on 20 is bad where 10ms is good and also is good 40,60,120?:

Stephen

Please report crashes in jira, otherwise they'll never get fixed. On Tue, Mar 27, 2012 at 12:34 PM, Stephen Wilde <wstephen80@gmail.com>wrote:

I have tried to use mod_posix_timer but with this timer I had some FS crashes (segmentation fault) with core dump so I'm moved back to soft timers.

Stephen

On Tue, Mar 27, 2012 at 6:28 PM, Christopher Rienzo <cmrienzo@gmail.com>wrote:

...

You can also try mod_posix_timer if you have it built. My CentOS 5.4 server performs better under load using it with heavy-timer vs using the default timer.

"timer_test 20 50 posix"

On Tue, Mar 27, 2012 at 12:07 PM, Stephen Wilde <wstephen80@gmail.com>wrote:

...

Ok, I'll try disabling the nanosleep.

In server.2 the current load of the box is 35% and after FreeSwitch (at 750% of cpu), in top there is Sangoma Transcoding server (4% of cpu).

The result of "timer_test 20 50" is:

http://pastebin.freeswitch.org/18756

Here the result of "timer_test 10 50" :

http://pastebin.freeswitch.org/18757

It's normal that only the timer_test on 20 is bad where 10ms is good and also is good 40,60,120?:

Stephen

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

It should be, the timerfd has been around since kernel >= 2.6.25 and libc >= 2.8 On Tue, Mar 27, 2012 at 12:29 PM, Lloyd Aloysius <lloyd.aloysius@gmail.com> wrote:

Tony,

Your recommendation for using the new kernel, is stable for a production environment?

Thanks Lloyd

On Tue, Mar 27, 2012 at 1:14 PM, Christopher Rienzo <cmrienzo@gmail.com> wrote:

...

Please report crashes in jira, otherwise they'll never get fixed.

On Tue, Mar 27, 2012 at 12:34 PM, Stephen Wilde <wstephen80@gmail.com> wrote:

...

I have tried to use mod_posix_timer but with this timer I had some FS crashes (segmentation fault) with core dump so I'm moved back to soft timers.

Stephen

On Tue, Mar 27, 2012 at 6:28 PM, Christopher Rienzo <cmrienzo@gmail.com> wrote:

...

You can also try mod_posix_timer if you have it built.  My CentOS 5.4 server performs better under load using it with heavy-timer vs using the default timer.

"timer_test 20 50 posix"

On Tue, Mar 27, 2012 at 12:07 PM, Stephen Wilde <wstephen80@gmail.com> wrote:

...

Ok, I'll try disabling the nanosleep.

In server.2 the current load of the box is 35% and after FreeSwitch (at 750% of cpu), in top there is Sangoma Transcoding server (4% of cpu).

The result of "timer_test 20 50" is:

http://pastebin.freeswitch.org/18756

Here the result of "timer_test 10 50" :

http://pastebin.freeswitch.org/18757

It's normal that only the timer_test on 20 is bad where 10ms is good and also is good 40,60,120?:

Stephen

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

-- Anthony Minessale II FreeSWITCH http://www.freeswitch.org/ ClueCon http://www.cluecon.com/ Twitter: http://twitter.com/FreeSWITCH_wire AIM: anthm MSN:anthony_minessale@hotmail.com GTALK/JABBER/PAYPAL:anthony.minessale@gmail.com IRC: irc.freenode.net #freeswitch FreeSWITCH Developer Conference sip:888@conference.freeswitch.org googletalk:conf+888@conference.freeswitch.org pstn:+19193869900

Tony, Thank you. On Tue, Mar 27, 2012 at 2:24 PM, Stephen Wilde <wstephen80@gmail.com> wrote:

The core dump was pretty unusable (stack corrupted) so I had some difficult to report to jira a useful info.

On Tue, Mar 27, 2012 at 7:14 PM, Christopher Rienzo <cmrienzo@gmail.com>wrote:

...

Please report crashes in jira, otherwise they'll never get fixed.

On Tue, Mar 27, 2012 at 12:34 PM, Stephen Wilde <wstephen80@gmail.com>wrote:

...

I have tried to use mod_posix_timer but with this timer I had some FS crashes (segmentation fault) with core dump so I'm moved back to soft timers.

Stephen

On Tue, Mar 27, 2012 at 6:28 PM, Christopher Rienzo <cmrienzo@gmail.com>wrote:

...

You can also try mod_posix_timer if you have it built. My CentOS 5.4 server performs better under load using it with heavy-timer vs using the default timer.

"timer_test 20 50 posix"

On Tue, Mar 27, 2012 at 12:07 PM, Stephen Wilde <wstephen80@gmail.com>wrote:

...

Ok, I'll try disabling the nanosleep.

In server.2 the current load of the box is 35% and after FreeSwitch (at 750% of cpu), in top there is Sangoma Transcoding server (4% of cpu).

The result of "timer_test 20 50" is:

http://pastebin.freeswitch.org/18756

Here the result of "timer_test 10 50" :

http://pastebin.freeswitch.org/18757

It's normal that only the timer_test on 20 is bad where 10ms is good and also is good 40,60,120?:

Stephen

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE: http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

Ok, thank you, I'll do a try with latest release. It remains my doubt on the result of timer_test with 10ms that is ok where the result with 20ms is not ok: there is an explanation? Stephen On Tue, Mar 27, 2012 at 10:52 PM, <cmrienzo@gmail.com> wrote:

No big deal. I fixed a few issues a while back, so maybe your problem is resolved. Check the revision history to see if the changes happened after your crash.

On Mar 27, 2012, at 14:24, Stephen Wilde <wstephen80@gmail.com> wrote:

The core dump was pretty unusable (stack corrupted) so I had some difficult to report to jira a useful info.

On Tue, Mar 27, 2012 at 7:14 PM, Christopher Rienzo <cmrienzo@gmail.com>wrote:

...

Please report crashes in jira, otherwise they'll never get fixed.

On Tue, Mar 27, 2012 at 12:34 PM, Stephen Wilde <wstephen80@gmail.com>wrote:

...

I have tried to use mod_posix_timer but with this timer I had some FS crashes (segmentation fault) with core dump so I'm moved back to soft timers.

Stephen

On Tue, Mar 27, 2012 at 6:28 PM, Christopher Rienzo <cmrienzo@gmail.com>wrote:

...

You can also try mod_posix_timer if you have it built. My CentOS 5.4 server performs better under load using it with heavy-timer vs using the default timer.

"timer_test 20 50 posix"

On Tue, Mar 27, 2012 at 12:07 PM, Stephen Wilde <wstephen80@gmail.com>wrote:

...

Ok, I'll try disabling the nanosleep.

In server.2 the current load of the box is 35% and after FreeSwitch (at 750% of cpu), in top there is Sangoma Transcoding server (4% of cpu).

The result of "timer_test 20 50" is:

http://pastebin.freeswitch.org/18756

Here the result of "timer_test 10 50" :

http://pastebin.freeswitch.org/18757

It's normal that only the timer_test on 20 is bad where 10ms is good and also is good 40,60,120?:

Stephen

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE: http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

To apply the changes I have to do a FS restart on the production server that I'll do during night. On Tue, Mar 27, 2012 at 11:03 PM, Anthony Minessale < anthony.minessale@gmail.com> wrote:

I provided you with a list, did you cross them all off?

On Tue, Mar 27, 2012 at 4:00 PM, Stephen Wilde <wstephen80@gmail.com> wrote:

...

Ok, thank you, I'll do a try with latest release. It remains my doubt on the result of timer_test with 10ms that is ok where the result with 20ms is not ok: there is an explanation?

Stephen

On Tue, Mar 27, 2012 at 10:52 PM, <cmrienzo@gmail.com> wrote:

...

No big deal. I fixed a few issues a while back, so maybe your problem

is

...

resolved. Check the revision history to see if the changes happened after your crash.

On Mar 27, 2012, at 14:24, Stephen Wilde <wstephen80@gmail.com> wrote:

The core dump was pretty unusable (stack corrupted) so I had some difficult to report to jira a useful info.

On Tue, Mar 27, 2012 at 7:14 PM, Christopher Rienzo <cmrienzo@gmail.com

...

wrote:

...

Please report crashes in jira, otherwise they'll never get fixed.

On Tue, Mar 27, 2012 at 12:34 PM, Stephen Wilde <wstephen80@gmail.com> wrote:

...

I have tried to use mod_posix_timer but with this timer I had some FS crashes (segmentation fault) with core dump so I'm moved back to soft timers.

Stephen

On Tue, Mar 27, 2012 at 6:28 PM, Christopher Rienzo <

cmrienzo@gmail.com>

...

...

wrote:

...

You can also try mod_posix_timer if you have it built. My CentOS 5.4 server performs better under load using it with heavy-timer vs using

the

...

default timer.

"timer_test 20 50 posix"

On Tue, Mar 27, 2012 at 12:07 PM, Stephen Wilde < wstephen80@gmail.com> wrote: > > Ok, I'll try disabling the nanosleep. > > In server.2 the current load of the box is 35% and after FreeSwitch > (at 750% of cpu), in top there is Sangoma Transcoding server (4% of cpu). > > > The result of "timer_test 20 50" is: > > http://pastebin.freeswitch.org/18756 > > > Here the result of "timer_test 10 50" : > > http://pastebin.freeswitch.org/18757 > > It's normal that only the timer_test on 20 is bad where 10ms is good > and also is good 40,60,120?: > > Stephen > >

---

...

...

...

...

...

Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users

UNSUBSCRIBE: http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

---

...

...

...

...

Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE: http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

---

...

...

...

Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE: http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

---

...

...

Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE: http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

---

...

...

Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE: http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

-- Anthony Minessale II

FreeSWITCH http://www.freeswitch.org/ ClueCon http://www.cluecon.com/ Twitter: http://twitter.com/FreeSWITCH_wire

AIM: anthm MSN:anthony_minessale@hotmail.com GTALK/JABBER/PAYPAL:anthony.minessale@gmail.com IRC: irc.freenode.net #freeswitch

FreeSWITCH Developer Conference sip:888@conference.freeswitch.org googletalk:conf+888@conference.freeswitch.org pstn:+19193869900

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

Very satisfied with mod_posix_timer (with this I have solved the issue of soft timer) but the problem of crash is present also with latest version. I have open a jira: http://jira.freeswitch.org/browse/FS-4064 Stephen On Tue, Mar 27, 2012 at 10:52 PM, <cmrienzo@gmail.com> wrote:

No big deal. I fixed a few issues a while back, so maybe your problem is resolved. Check the revision history to see if the changes happened after your crash.

On Mar 27, 2012, at 14:24, Stephen Wilde <wstephen80@gmail.com> wrote:

The core dump was pretty unusable (stack corrupted) so I had some difficult to report to jira a useful info.

On Tue, Mar 27, 2012 at 7:14 PM, Christopher Rienzo <cmrienzo@gmail.com>wrote:

...

Please report crashes in jira, otherwise they'll never get fixed.

On Tue, Mar 27, 2012 at 12:34 PM, Stephen Wilde <wstephen80@gmail.com>wrote:

...

I have tried to use mod_posix_timer but with this timer I had some FS crashes (segmentation fault) with core dump so I'm moved back to soft timers.

Stephen

On Tue, Mar 27, 2012 at 6:28 PM, Christopher Rienzo <cmrienzo@gmail.com>wrote:

...

You can also try mod_posix_timer if you have it built. My CentOS 5.4 server performs better under load using it with heavy-timer vs using the default timer.

"timer_test 20 50 posix"

On Tue, Mar 27, 2012 at 12:07 PM, Stephen Wilde <wstephen80@gmail.com>wrote:

...

Ok, I'll try disabling the nanosleep.

In server.2 the current load of the box is 35% and after FreeSwitch (at 750% of cpu), in top there is Sangoma Transcoding server (4% of cpu).

The result of "timer_test 20 50" is:

http://pastebin.freeswitch.org/18756

Here the result of "timer_test 10 50" :

http://pastebin.freeswitch.org/18757

It's normal that only the timer_test on 20 is bad where 10ms is good and also is good 40,60,120?:

Stephen

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE: http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

I am having the same issue as Stephen Wilde regarding the strange "timer_test" results. Was it determined that the mod_posix timer is the best solution? Likewise with Stephen, any timer except 20ms works great. -- View this message in context: http://freeswitch-users.2379917.n2.nabble.com/Strange-timer-test-result-tp73... Sent from the freeswitch-users mailing list archive at Nabble.com.

My experience in FS timers: 1. voice quality is related to timer_test result so if you have irregular result, you'll have bad voice quality (and low ACD) 2. mod_posix_timer is my solution (I haven't try timerfd because I'm using CentOS 5.8). 3. there are two version of mod_posix_timer, the "old" one (the one before the fix of git 'FS-4064 mod_posix_timer workaround of CentOS 5 glibc problem' of 30 april) and the 'new' one (present in latest git) 4. the 'old' mod_posix_timer is much more efficient that 'new' one (with 'new' mod_posix_timer is half way between 'old' mod_posix_timer and soft timer). 5. loading the 'old' mod_posix_timer there are random FS crashes, also if you don't use the posix timer in your profile 6. the 'new' mod_posix_timer solves the crashes of FS So my conclusion is that the 'old' mod_posix_timer is the best one but sometimes FS crashes but the 'new' mod_posix_timer is not good as the 'old' one but it solves the crash issue. In my production system I'm using the 'old' one in servers with high load and the 'new' one in serves with less load, according with timer_test results. Memory leak: I have done some test mixing different versions of FS and mod_posix_timer and in any case, if I use the latest Freeswitch git the used memory grows up continously, also if I use 'old' mod_posix_timer so, my conclusion is that the memory leakage is related to Freeswitch. In my production server, latest FS git + old mod_posix_timer consumes all server memory (12MB) in 2 days. For this reason, in all my installation I'm using the Freeswitch git of 26th march with new or old mod_posix_timer (depending on timer_test result). Stephen On Wed, Jun 27, 2012 at 12:00 AM, marksly <mark@callassistants.com> wrote:

I am having the same issue as Stephen Wilde regarding the strange "timer_test" results. Was it determined that the mod_posix timer is the best solution? Likewise with Stephen, any timer except 20ms works great.

-- View this message in context: http://freeswitch-users.2379917.n2.nabble.com/Strange-timer-test-result-tp73... Sent from the freeswitch-users mailing list archive at Nabble.com.

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

Join Us At ClueCon - Aug 7-9, 2012

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

Thank you Chirs, do you have any reference to how to patch the glibc for this issue? Stephen On Wed, Jun 27, 2012 at 1:46 PM, Christopher Rienzo <cmrienzo@gmail.com>wrote:

The crash in the old posix timer is actually caused by a glibc timer_create() race condition when using SIGEV_THREAD. So, you could should be able to use the old timer if you patch glibc. I figured that was not a good solution for end users, so I switched to SIGEV_SIGNAL which did degrade performance.

Chris

On Wed, Jun 27, 2012 at 4:44 AM, Stephen Wilde <wstephen80@gmail.com>wrote:

...

My experience in FS timers:

1. voice quality is related to timer_test result so if you have irregular result, you'll have bad voice quality (and low ACD)

2. mod_posix_timer is my solution (I haven't try timerfd because I'm using CentOS 5.8).

3. there are two version of mod_posix_timer, the "old" one (the one before the fix of git 'FS-4064 mod_posix_timer workaround of CentOS 5 glibc problem' of 30 april) and the 'new' one (present in latest git)

4. the 'old' mod_posix_timer is much more efficient that 'new' one (with 'new' mod_posix_timer is half way between 'old' mod_posix_timer and soft timer).

5. loading the 'old' mod_posix_timer there are random FS crashes, also if you don't use the posix timer in your profile

6. the 'new' mod_posix_timer solves the crashes of FS

So my conclusion is that the 'old' mod_posix_timer is the best one but sometimes FS crashes but the 'new' mod_posix_timer is not good as the 'old' one but it solves the crash issue.

In my production system I'm using the 'old' one in servers with high load and the 'new' one in serves with less load, according with timer_test results.

Memory leak: I have done some test mixing different versions of FS and mod_posix_timer and in any case, if I use the latest Freeswitch git the used memory grows up continously, also if I use 'old' mod_posix_timer so, my conclusion is that the memory leakage is related to Freeswitch. In my production server, latest FS git + old mod_posix_timer consumes all server memory (12MB) in 2 days. For this reason, in all my installation I'm using the Freeswitch git of 26th march with new or old mod_posix_timer (depending on timer_test result).

Stephen

On Wed, Jun 27, 2012 at 12:00 AM, marksly <mark@callassistants.com>wrote:

...

I am having the same issue as Stephen Wilde regarding the strange "timer_test" results. Was it determined that the mod_posix timer is the best solution? Likewise with Stephen, any timer except 20ms works great.

-- View this message in context: http://freeswitch-users.2379917.n2.nabble.com/Strange-timer-test-result-tp73... Sent from the freeswitch-users mailing list archive at Nabble.com.

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

Join Us At ClueCon - Aug 7-9, 2012

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

Join Us At ClueCon - Aug 7-9, 2012

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

Join Us At ClueCon - Aug 7-9, 2012

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

To give everyone an update: I enabled the mod_posix_timer last night and things appear to be running better today (fewer calls that have a sporadic choppy audio). The server load is lower, but it is not where I would like to see it. The load is around 10 on average. I think that this might be due to the fact that I have "enable-clock-nanosleep" set to true. I am going to try disabling this and see if the load decreases. The timer_test results are more consistent and regular; a lot closer to the 20ms that you would expect to see. Server Information: 2x AMD Opteron 6128 (8 cores) 8GB RAM CentOS release 5.7 Kernel: 2.6.18-274.7.1.el5 FreeSWITCH version: 1.2.0-rc2+git~20120625T203158Z~7dc9a9cacc+unclean~20120626T041419Z Feel free to provide other suggestions if you see fit. -- View this message in context: http://freeswitch-users.2379917.n2.nabble.com/Strange-timer-test-result-tp73... Sent from the freeswitch-users mailing list archive at Nabble.com.

Get a nice new many core intel box with a nice motherboard and give freeswitch a nice home. On Thu, Jun 28, 2012 at 5:21 PM, Stephen Wilde <wstephen80@gmail.com> wrote:

I have bad experience with combination Opteron + RTP (choppy audio). I think this is due to the size of the cache, for example for your 6128 is 128KB per core and, for example, for a Xeon E5649 is 12Mb per core. I own a 4 x Opteron 6176 (4 x 12 core) and after many tries to use it as Freeswitch server, now I'm using it as database server...

Stephen

On Thu, Jun 28, 2012 at 11:02 PM, marksly <mark@callassistants.com> wrote:

...

To give everyone an update: I enabled the mod_posix_timer last night and things appear to be running better today (fewer calls that have a sporadic choppy audio). The server load is lower, but it is not where I would like to see it. The load is around 10 on average. I think that this might be due to the fact that I have "enable-clock-nanosleep" set to true. I am going to try disabling this and see if the load decreases.

The timer_test results are more consistent and regular; a lot closer to the 20ms that you would expect to see.

Server Information: 2x AMD Opteron 6128 (8 cores) 8GB RAM CentOS release 5.7 Kernel: 2.6.18-274.7.1.el5 FreeSWITCH version: 1.2.0-rc2+git~20120625T203158Z~7dc9a9cacc+unclean~20120626T041419Z

Feel free to provide other suggestions if you see fit.

-- View this message in context: http://freeswitch-users.2379917.n2.nabble.com/Strange-timer-test-result-tp73... Sent from the freeswitch-users mailing list archive at Nabble.com.

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

Join Us At ClueCon - Aug 7-9, 2012

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

Join Us At ClueCon - Aug 7-9, 2012

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

-- Anthony Minessale II FreeSWITCH http://www.freeswitch.org/ ClueCon http://www.cluecon.com/ Twitter: http://twitter.com/FreeSWITCH_wire AIM: anthm MSN:anthony_minessale@hotmail.com GTALK/JABBER/PAYPAL:anthony.minessale@gmail.com IRC: irc.freenode.net #freeswitch FreeSWITCH Developer Conference sip:888@conference.freeswitch.org googletalk:conf+888@conference.freeswitch.org pstn:+19193869900

You need to help us find that leak. If you go back and look at the jira, I have asked for you to do some things. Thanks. On Wed, Jun 27, 2012 at 3:44 AM, Stephen Wilde <wstephen80@gmail.com> wrote:

My experience in FS timers:

1. voice quality is related to timer_test result so if you have irregular result, you'll have bad voice quality (and low ACD)

2. mod_posix_timer is my solution (I haven't try timerfd because I'm using CentOS 5.8).

3. there are two version of mod_posix_timer, the "old" one (the one before the fix of git 'FS-4064 mod_posix_timer workaround of CentOS 5 glibc problem'  of 30 april) and the 'new' one (present in latest git)

4. the 'old' mod_posix_timer is much more efficient that 'new' one (with 'new' mod_posix_timer is half way between 'old' mod_posix_timer and soft timer).

5. loading the 'old' mod_posix_timer there are random FS crashes, also if you don't use the posix timer in your profile

6. the 'new' mod_posix_timer solves the crashes of FS

So my conclusion is that the 'old' mod_posix_timer is the best one but sometimes FS crashes but the 'new' mod_posix_timer is not good as the 'old' one but it solves the crash issue.

In my production system I'm using the 'old' one in servers with high load and the 'new' one in serves with less load, according with timer_test results.

Memory leak: I have done some test mixing different versions of FS and mod_posix_timer and in any case, if I use the latest Freeswitch git the used memory grows up continously, also if I use 'old' mod_posix_timer so, my conclusion is that the memory leakage is related to Freeswitch. In my production server, latest FS git + old mod_posix_timer consumes all server memory (12MB) in 2 days. For this reason, in all my installation I'm using the Freeswitch git of 26th march with new or old mod_posix_timer (depending on timer_test result).

Stephen

On Wed, Jun 27, 2012 at 12:00 AM, marksly <mark@callassistants.com> wrote:

...

I am having the same issue as Stephen Wilde regarding the strange "timer_test" results. Was it determined that the mod_posix timer is the best solution? Likewise with Stephen, any timer except 20ms works great.

-- View this message in context: http://freeswitch-users.2379917.n2.nabble.com/Strange-timer-test-result-tp73... Sent from the freeswitch-users mailing list archive at Nabble.com.

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

Join Us At ClueCon - Aug 7-9, 2012

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

Join Us At ClueCon - Aug 7-9, 2012

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org

-- Anthony Minessale II FreeSWITCH http://www.freeswitch.org/ ClueCon http://www.cluecon.com/ Twitter: http://twitter.com/FreeSWITCH_wire AIM: anthm MSN:anthony_minessale@hotmail.com GTALK/JABBER/PAYPAL:anthony.minessale@gmail.com IRC: irc.freenode.net #freeswitch FreeSWITCH Developer Conference sip:888@conference.freeswitch.org googletalk:conf+888@conference.freeswitch.org pstn:+19193869900

Thank you Patrick, I'm sure that it's something hardware related because with same load, different hardware show different behaviour but I don't think it's a fault but something related to the architecture. For example to me it seems that HP server are more regular then Fujitsu or Dell at the same load. On Wed, Mar 28, 2012 at 1:59 PM, Patrick Lists < freeswitch-list@puzzled.xs4all.nl> wrote:

On 03/27/2012 05:37 PM, Stephen Wilde wrote:

...

Thank you for your reply. The motherboard cannot be the issue because 2 of these servers (server.2 and server.5) are exactly the same hardware (HP DL380 G6).

In the past I have seen similar hardware where one worked and the other did not. Yet all diagnostics said both boxes were fine. After a lot of digging the hardware in one of the boxes was faulty. Iirc it was PCIe related. So if you don't solve your problem, even with Anthony's recommendations, and you have ruled out any software issue(s) then perhaps it could be a hardware issue.

Are all firmware versions similar in both boxes? Last I looked for a DL360 G6 there were firmware fixes for a bunch of nasty problems.

Regards, Patrick

_________________________________________________________________________ Professional FreeSWITCH Consulting Services: consulting@freeswitch.org http://www.freeswitchsolutions.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server http://www.cudatel.com

Official FreeSWITCH Sites http://www.freeswitch.org http://wiki.freeswitch.org http://www.cluecon.com

FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org