<div dir="ltr">Please try to reproduce with  -np switch.<br><div style>Thanks.</div></div><div class="gmail_extra"><br><br><div class="gmail_quote">2013/3/8 bratner bratner <span dir="ltr">&lt;<a href="mailto:ratner2@gmail.com" target="_blank">ratner2@gmail.com</a>&gt;</span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Here is sipp output and additional numbers for a test I ran with -nosql param.<br><br>The test ran 180CPS for ~3500seconds and the rest with 210cps.<br>
<br>Trouble (as in higher system cpu% ) started to appear around 8591seconds into the test. <br>
As you can see below the problem started just before 9124sec into the test  210cps 5sec calls<br>should not give you a lot more then 1050 concurrent calls. <br><br>------------------------------ Scenario Screen -------- [1-9]: Change Screen --<br>

  Call-rate(length)   Port   Total-time  Total-calls  Remote-host<br>210.0(5000 ms)/1.000s   5061    9157.32 s      1834024  192.96.201.164:5060(UDP)<br><br>  0 new calls during 0.000 s period      0 ms scheduler resolution<br>

  0 calls (limit 2000)                   Peak was 2000 calls, after 9124 s<br>  0 Running, 4640 Paused, 0 Woken up<br>  20 dead call msg (discarded)           0 out-of-call msg (discarded)        <br>  1 open sockets                        <br>

<br>                                 Messages  Retrans   Timeout   Unexpected-Msg<br>      INVITE ----------&gt;         1834024   74        0                  <br>         100 &lt;----------         1834024   0         0         0        <br>

         180 &lt;----------         1834024   0         0         0        <br>         183 &lt;----------         0         0         0         0        <br>         500 &lt;----------         0         0         0         0        <br>

         502 &lt;----------         0         0         0         0        <br>         503 &lt;----------         0         0         0         0        <br>         408 &lt;----------         0         0         0         0        <br>

         480 &lt;----------         0         0         0         0        <br>         200 &lt;----------  E-RTD1 1834024   81        0         0        <br><br>         ACK ----------&gt;         1834024   81                           <br>

       Pause [   5000ms]         1834024                       0        <br>         BYE ----------&gt;         1834024   7646      0                  <br>         503 &lt;----------         0         0         0         0        <br>

         200 &lt;----------         1834024   0         0         0        <br><br>------------------------------ Test Terminated --------------------------------<br><br><br>----------------------------- Statistics Screen ------- [1-9]: Change Screen --<br>

  Start Time             | 2013-03-08    15:22:18:204    1362756138.204833            <br>  Last Reset Time        | 2013-03-08    17:54:55:535    1362765295.535214            <br>  Current Time           | 2013-03-08    17:54:55:535    1362765295.535437            <br>

-------------------------+---------------------------+--------------------------<br>  Counter Name           | Periodic value            | Cumulative value<br>-------------------------+---------------------------+--------------------------<br>

  Elapsed Time           | 00:00:00:000              | 02:32:37:330             <br>  Call Rate              |    0.000 cps              |  200.279 cps             <br>-------------------------+---------------------------+--------------------------<br>

  Incoming call created  |        0                  |        0                 <br>  OutGoing call created  |        0                  |  1834024                 <br>  Total Call created     |                           |  1834024                 <br>

  Current Call           |        0                  |                          <br>-------------------------+---------------------------+--------------------------<br>  Successful call        |        0                  |  1834024                 <br>

  Failed call            |        0                  |        0                 <br>-------------------------+---------------------------+--------------------------<br>  Response Time 1        | 00:00:00:000              | 00:00:00:149             <br>

  Call Length            | 00:00:00:000              | 00:00:05:158             <br>------------------------------ Test Terminated --------------------------------<br><br><br>After stopping the load FS still hogs 22.1% of memory. <br>
<div class="im">
  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND                                                                                                                                        <br></div>15995 freeswit  -2 -10 4677m 873m 5028 S    0 22.1 755:28.65 freeswitch                                                                                                                                     <br>

<br><br>The symptoms of the crash are the same, just now with higher CPS and takes more time (more calls ) before crashing. <br><br>I will appreciate any suggestion.<br><br>Regards, <br>Boris Ratner.<div><div class="h5">
<br><br><br><div class="gmail_quote">
On Fri, Mar 8, 2013 at 6:22 PM, bratner bratner <span dir="ltr">&lt;<a href="mailto:ratner2@gmail.com" target="_blank">ratner2@gmail.com</a>&gt;</span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">

<div dir="ltr">The original test was done on git master at the date mentioned. The sqlite core.db file was on /run/shm which is a tmpfs on unbuntu 12.04. <br>I will be recompiling from git master and test running with -nosql. <br>


<br>Testing my existing setup with -nosql seems more stable now running at 210CPS for some time (500k calls already passed) with ~35% idle cpu. <br>But the free mem is slowly going down. I will let it run untill the kernel will kill it to see how many calls it can handle. <br>


<br>During my tests i did not run FS with RT priority but according to htop some of the threads are scheduled as RT. <br>My setup is doing bypass-media , thus FS handling only call establishment and teardown on both legs. <br>


<br>cat /proc/&lt;FS pid&gt;/status <br><br>Name:   freeswitch<br>State:  S (sleeping)<br>Tgid:   15995<br>Pid:    15995<br>PPid:   1<br>TracerPid:      0<br>Uid:    999     999     999     999<br>Gid:    999     999     999     999<br>


FDSize: 64<br>Groups:<br>VmPeak:  5002808 kB<br>VmSize:  5002088 kB<br>VmLck:         0 kB<br>VmPin:         0 kB<br>VmHWM:    625900 kB<br>VmRSS:    624156 kB  &lt;-- this is going up <br>VmData:  4855788 kB<br>VmStk:       136 kB<br>


VmExe:        20 kB<br>VmLib:     18288 kB<br>VmPTE:      2352 kB<br>VmSwap:        0 kB<br>Threads:        1866<br>SigQ:   0/18446744073709551615<br>SigPnd: 0000000000000000<br>ShdPnd: 0000000000000000<br>SigBlk: 0000000000000000<br>


SigIgn: 0000000010003006<br>SigCgt: 0000000180014209<br>CapInh: 0000000000000000<br>CapPrm: 0000000000000000<br>CapEff: 0000000000000000<br>CapBnd: ffffffffffffffff<br>Cpus_allowed:   ffffff<br>Cpus_allowed_list:      0-23<br>


Mems_allowed:   00000000,00000003<br>Mems_allowed_list:      0-1<br>voluntary_ctxt_switches:        1803<br>nonvoluntary_ctxt_switches:     23<br><br><br>output of &#39;top -H&#39; at 180CPS<br><br><br>top - 15:27:00 up 2 days,  5:32,  5 users,  load average: 8.19, 91.07, 65.03<br>


Tasks: 2066 total,   3 running, 2063 sleeping,   0 stopped,   0 zombie<br>Cpu(s): 50.1%us,  3.9%sy,  0.0%ni, 45.9%id,  0.0%wa,  0.0%hi,  0.2%si,  0.0%st<br>Mem:   4038512k total,  2282260k used,  1756252k free,   114112k buffers<br>


Swap:        0k total,        0k used,        0k free,  1165868k cached<br><br>  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND                                                                                                                                        <br>


16000 freeswit  RT -10 4885m 594m 4964 R   69 15.1   3:10.26 freeswitch                                                                                                                                     <br>16009 freeswit  RT -10 4885m 594m 4964 S   33 15.1   1:26.20 freeswitch                                                                                                                                     <br>


16008 freeswit  RT -10 4885m 594m 4964 S   28 15.1   1:17.30 freeswitch                                                                                                                                     <br>16007 freeswit  RT -10 4885m 594m 4964 S    4 15.1   0:10.80 freeswitch                                                                                                                                     <br>


16004 freeswit  RT -10 4885m 594m 4964 S    2 15.1   0:06.63 freeswitch                                                                                                                                     <br>19171 root      20   0 18988 2948  944 R    2  0.1   0:00.64 top                                                                                                                                            <br>


18735 freeswit  -2 -10 4885m 594m 4964 S    1 15.1   0:00.29 freeswitch                                                                                                                                     <br>16003 freeswit  -2 -10 4885m 594m 4964 S    1 15.1   0:01.61 freeswitch                                                                                                                                     <br>


16690 freeswit  -2 -10 4885m 594m 4964 S    1 15.1   0:00.42 freeswitch                                                                                                                                     <br>16730 freeswit  -2 -10 4885m 594m 4964 S    1 15.1   0:00.42 freeswitch                                                                                                                                     <br>


16750 freeswit  -2 -10 4885m 594m 4964 S    1 15.1   0:00.45 freeswitch                                                                                                                                     <br>16764 freeswit  -2 -10 4885m 594m 4964 S    1 15.1   0:00.44 freeswitch                                                                                                                                     <br>


&lt;more of the above&gt;<br>....<br>....<br><br><br>Thanks to all of you , <br>Boris Ratner. <br><div><div><br><div class="gmail_quote">On Fri, Mar 8, 2013 at 4:22 AM, Dmitry Lysenko <span dir="ltr">&lt;<a href="mailto:dvl36.ripe.nick@gmail.com" target="_blank">dvl36.ripe.nick@gmail.com</a>&gt;</span> wrote:<br>


<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"><div>I can&#39;t reproduce such cps load on my ARMv5TE system. )</div><div>bratner, please give us &#39;top -H&#39;. I guess freeswitch running at realtime priority. </div>


</div><div><div><div class="gmail_extra"><br>
<br><div class="gmail_quote">2013/3/8 Ken Rice <span dir="ltr">&lt;<a href="mailto:krice@freeswitch.org" target="_blank">krice@freeswitch.org</a>&gt;</span><br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">







<div>
<font face="Monaco, Courier New"><span style="font-size:11pt">Sqlite is probably getting hammered... Trust me... Mount the fs db dir as tmpfs or use the –nosql flag when starting freeswitch<br>
<br>
I routinely run dialer traffic at much higher CPS then that<div><br>
<br>
<br>
On 3/7/13 7:58 PM, &quot;Dmitry Lysenko&quot; &lt;<a href="http://dvl36.ripe.nick@gmail.com" target="_blank">dvl36.ripe.nick@gmail.com</a>&gt; wrote:<br>
<br>
</div></span></font><blockquote><div><font face="Monaco, Courier New"><span style="font-size:11pt">bi, bo and wa field is low, so it seems that is not disk subsystem</span><span style="font-size:11.5pt">.<br>
</span><span style="font-size:11pt"><br>
<br>
2013/3/8 Ken Rice &lt;<a href="http://krice@freeswitch.org" target="_blank">krice@freeswitch.org</a>&gt;<br>
</span></font></div><blockquote><font face="Monaco, Courier New"><span style="font-size:11pt"><div>You are probably hammering the disk subsystem... Keep in mind that FS uses multiple sqlite databases by default... Mount the fs db dir as tmpfs and try again<br>




<br>
<br>
<br></div><div>
On 3/7/13 7:35 PM, &quot;Dmitry Lysenko&quot; &lt;<a href="http://dvl36.ripe.nick@gmail.com" target="_blank">dvl36.ripe.nick@gmail.com</a> &lt;<a href="http://dvl36.ripe.nick@gmail.com" target="_blank">http://dvl36.ripe.nick@gmail.com</a>&gt; &gt; wrote:<br>




<br>
</div></span></font><div><blockquote><font face="Monaco, Courier New"><span style="font-size:11pt">Hm... But what about huge interrupt and context switching  number?<br>
<br>
</span></font></blockquote></div></blockquote><font face="Monaco, Courier New"><span style="font-size:11pt"><br>
<hr align="CENTER" size="3" width="95%"></span></font><div><font><font face="Consolas, Courier New, Courier"><span style="font-size:10pt">_________________________________________________________________________<br>

Professional FreeSWITCH Consulting Services:<br>
<a href="http://consulting@freeswitch.org" target="_blank">consulting@freeswitch.org</a><br>
<a href="http://www.freeswitchsolutions.com" target="_blank">http://www.freeswitchsolutions.com</a><br>
<br>
FreeSWITCH-powered IP PBX: The CudaTel Communication Server<br>
<a href="http://www.cudatel.com" target="_blank">http://www.cudatel.com</a><br>
<br>
Official FreeSWITCH Sites<br>
<a href="http://www.freeswitch.org" target="_blank">http://www.freeswitch.org</a><br>
<a href="http://wiki.freeswitch.org" target="_blank">http://wiki.freeswitch.org</a><br>
<a href="http://www.cluecon.com" target="_blank">http://www.cluecon.com</a><br>
<br>
FreeSWITCH-users mailing list<br>
<a href="http://FreeSWITCH-users@lists.freeswitch.org" target="_blank">FreeSWITCH-users@lists.freeswitch.org</a><br>
<a href="http://lists.freeswitch.org/mailman/listinfo/freeswitch-users" target="_blank">http://lists.freeswitch.org/mailman/listinfo/freeswitch-users</a><br>
UNSUBSCRIBE:<a href="http://lists.freeswitch.org/mailman/options/freeswitch-users" target="_blank">http://lists.freeswitch.org/mailman/options/freeswitch-users</a><br>
<a href="http://www.freeswitch.org" target="_blank">http://www.freeswitch.org</a><br>
</span></font></font></div></blockquote><div><font><font face="Consolas, Courier New, Courier"><span style="font-size:10pt"><br>
</span></font></font><font face="Monaco, Courier New"><span style="font-size:11pt">-- <br>
Ken<br>
<font color="#0000FF"><u><a href="http://www.FreeSWITCH.org" target="_blank">http://www.FreeSWITCH.org</a><br>
<a href="http://www.ClueCon.com" target="_blank">http://www.ClueCon.com</a><br>
<a href="http://www.OSTAG.org" target="_blank">http://www.OSTAG.org</a><br>
</u></font><a href="http://irc.freenode.net" target="_blank">irc.freenode.net</a> #freeswitch<br>
</span></font>
</div></div>


<br>_________________________________________________________________________<br>
Professional FreeSWITCH Consulting Services:<br>
<a href="mailto:consulting@freeswitch.org" target="_blank">consulting@freeswitch.org</a><br>
<a href="http://www.freeswitchsolutions.com" target="_blank">http://www.freeswitchsolutions.com</a><br>
<br>
FreeSWITCH-powered IP PBX: The CudaTel Communication Server<br>
<a href="http://www.cudatel.com" target="_blank">http://www.cudatel.com</a><br>
<br>
Official FreeSWITCH Sites<br>
<a href="http://www.freeswitch.org" target="_blank">http://www.freeswitch.org</a><br>
<a href="http://wiki.freeswitch.org" target="_blank">http://wiki.freeswitch.org</a><br>
<a href="http://www.cluecon.com" target="_blank">http://www.cluecon.com</a><br>
<br>
FreeSWITCH-users mailing list<br>
<a href="mailto:FreeSWITCH-users@lists.freeswitch.org" target="_blank">FreeSWITCH-users@lists.freeswitch.org</a><br>
<a href="http://lists.freeswitch.org/mailman/listinfo/freeswitch-users" target="_blank">http://lists.freeswitch.org/mailman/listinfo/freeswitch-users</a><br>
UNSUBSCRIBE:<a href="http://lists.freeswitch.org/mailman/options/freeswitch-users" target="_blank">http://lists.freeswitch.org/mailman/options/freeswitch-users</a><br>
<a href="http://www.freeswitch.org" target="_blank">http://www.freeswitch.org</a><br>
<br></blockquote></div><br></div>
</div></div><br>_________________________________________________________________________<br>
Professional FreeSWITCH Consulting Services:<br>
<a href="mailto:consulting@freeswitch.org" target="_blank">consulting@freeswitch.org</a><br>
<a href="http://www.freeswitchsolutions.com" target="_blank">http://www.freeswitchsolutions.com</a><br>
<br>
FreeSWITCH-powered IP PBX: The CudaTel Communication Server<br>
<a href="http://www.cudatel.com" target="_blank">http://www.cudatel.com</a><br>
<br>
Official FreeSWITCH Sites<br>
<a href="http://www.freeswitch.org" target="_blank">http://www.freeswitch.org</a><br>
<a href="http://wiki.freeswitch.org" target="_blank">http://wiki.freeswitch.org</a><br>
<a href="http://www.cluecon.com" target="_blank">http://www.cluecon.com</a><br>
<br>
FreeSWITCH-users mailing list<br>
<a href="mailto:FreeSWITCH-users@lists.freeswitch.org" target="_blank">FreeSWITCH-users@lists.freeswitch.org</a><br>
<a href="http://lists.freeswitch.org/mailman/listinfo/freeswitch-users" target="_blank">http://lists.freeswitch.org/mailman/listinfo/freeswitch-users</a><br>
UNSUBSCRIBE:<a href="http://lists.freeswitch.org/mailman/options/freeswitch-users" target="_blank">http://lists.freeswitch.org/mailman/options/freeswitch-users</a><br>
<a href="http://www.freeswitch.org" target="_blank">http://www.freeswitch.org</a><br>
<br></blockquote></div><br></div></div></div>
</blockquote></div><br></div></div></div>
<br>_________________________________________________________________________<br>
Professional FreeSWITCH Consulting Services:<br>
<a href="mailto:consulting@freeswitch.org">consulting@freeswitch.org</a><br>
<a href="http://www.freeswitchsolutions.com" target="_blank">http://www.freeswitchsolutions.com</a><br>
<br>
FreeSWITCH-powered IP PBX: The CudaTel Communication Server<br>
<a href="http://www.cudatel.com" target="_blank">http://www.cudatel.com</a><br>
<br>
Official FreeSWITCH Sites<br>
<a href="http://www.freeswitch.org" target="_blank">http://www.freeswitch.org</a><br>
<a href="http://wiki.freeswitch.org" target="_blank">http://wiki.freeswitch.org</a><br>
<a href="http://www.cluecon.com" target="_blank">http://www.cluecon.com</a><br>
<br>
FreeSWITCH-users mailing list<br>
<a href="mailto:FreeSWITCH-users@lists.freeswitch.org">FreeSWITCH-users@lists.freeswitch.org</a><br>
<a href="http://lists.freeswitch.org/mailman/listinfo/freeswitch-users" target="_blank">http://lists.freeswitch.org/mailman/listinfo/freeswitch-users</a><br>
UNSUBSCRIBE:<a href="http://lists.freeswitch.org/mailman/options/freeswitch-users" target="_blank">http://lists.freeswitch.org/mailman/options/freeswitch-users</a><br>
<a href="http://www.freeswitch.org" target="_blank">http://www.freeswitch.org</a><br>
<br></blockquote></div><br></div>