Load avg is often skewed in many CPU env from the scheduler and RT threads used in FS.
Its better to trust the results of top -H `cat /usr/local/freeswitch/run/freeswitch.pid` and make audible calls to listen to the audio and make sure its working.

Also, we do not really touch Centos7 other than to make packages for it, its only community-supported and we don't know the details of how its kernel is configured etc.



On Thu, Dec 10, 2015 at 11:36 AM, Adam Kuśmirek <amkusmirek@gmail.com> wrote:
Hello 

We see a high load on server with two 10 core HT processors.
cat /proc/cpuinfo shows 40 processors.

1 minute load average is often grater than 70 when processing ~120 concurrent calls.

CPU utilization shows 97%idle and 0.1%wa
I checked also io stats and everything looks OK.

Operating system Centos 7
uname -a 
Linux pbx-fs-drc.ib 3.10.0-229.7.2.el7.x86_64 #1 SMP Tue Jun 23 22:06:11 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux

Below link to pastebin with gdb output

gdb /usr/local/freeswitch/bin/freeswitch /usr/local/freeswitch/log/freeswitch.gcore.2699
set pagination off
thread apply all bt


We are planning to run much more concurrent calls but do not understand these stats.
Why processes are waiting in queue and we do not see any bottleneck ?


Please help



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

Official FreeSWITCH Sites
http://www.freeswitch.org
http://confluence.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       ♬ @anthmfs  ♬ @FreeSWITCH  ♬

ClueCon Weekly Development Call 
☎ sip:888@conference.freeswitch.org  ☎ +19193869900