[Freeswitch-users] Freeswitch Performance Testing

[Michael S Collins]

If you remind me where in the configs to change the max sessions per  
sec I will update the performance testing wiki page.
-MC

Sent from my iPhone

On Jul 20, 2008, at 8:28 AM, Michael Jerris <mike at jerris.com> wrote:

> You wil get that error if you hit a system thread limit or run out of
> memory.  We generally recommend using a 64 bit cpu and os and the
> settings detailed:
>
> http://wiki.freeswitch.org/wiki/Performance_testing_and_configurations
>
> Also, not that the default config has sessions per second limited at  
> 30.
>
> Mike
>
>
> On Jul 20, 2008, at 10:57 AM, Juan Backson wrote:
>
>> Hi,
>>
>> We are running some stress tests  again  freeswitch  using sipp.
>>
>> At 100 concurrent call, Freeswitch starts to throw out some strange
>> error messages:
>>
>>
>>
>> freeswitch at lab> 2008-07-21 06:27:51 [CRIT] switch_core_session.c:826
>> switch_core_session$
>> 2008-07-21 06:27:51 [CRIT] sofia.c:3177 sofia_handle_sip_i_invite()
>> LUKE: I'm hit, but not $
>> 2008-07-21 06:27:51 [CRIT] sofia.c:3178 sofia_handle_sip_i_invite()
>> LUKE'S VOICE: Artoo, se$
>> Green laserfire moves past the beeping little robot as his head
>> turns.  After a few beeps a$
>> Artoo reduces the max sessions to 328 thus, saving the switch from
>> certain doom.
>>
>>
>>
>>
>> ----------------------------- Statistics Screen ------- [1-9]:
>> Change Screen --
>>  Start Time             | 2008-07-21    06:41:00:154
>> 1216593660.154593
>>  Last Reset Time        | 2008-07-21    06:49:25:136
>> 1216594165.136188
>>  Current Time           | 2008-07-21    06:49:26:139
>> 1216594166.139989
>> -------------------------+---------------------------
>> +--------------------------
>>  Counter Name           | Periodic value            | Cumulative
>> value
>> -------------------------+---------------------------
>> +--------------------------
>>  Elapsed Time           | 00:00:01:003              | 00:08:25:985
>>  Call Rate              |    9.970 cps              |    9.998 cps
>> -------------------------+---------------------------
>> +--------------------------
>>  Incoming call created  |        0                  |        0
>>  OutGoing call created  |       10                  |     5059
>>  Total Call created     |                           |     5059
>>  Current Call           |      100                  |
>> -------------------------+---------------------------
>> +--------------------------
>>  Successful call        |       10                  |     4387
>>  Failed call            |        0                  |      572
>> -------------------------+---------------------------
>> +--------------------------
>>  Response Time 1        | 00:00:00:018              | 00:00:08:545
>>  Call Length            | 00:00:10:021              | 00:00:22:125
>> ------ [+|-|*|/]: Adjust rate ---- [q]: Soft exit ---- [p]: Pause
>> traffic -----
>>
>> Last Error: Aborting call on unexpected message for Call-Id
>> '3143-6125 at 1...
>>
>>
>> When increasing the number of concurrent calls to several hundreds,
>> Freeswitch throws the following error:
>> [CRIT] switch_time.c:394 softtimer_runtime() Over Session Rate of 30!
>>
>> The test is run on Dual Core Intel P4 3GHz with 1G of RAM.
>>
>> Is this kind of performance normal for such kind of machine or there
>> are some OS configs required to enable more concurrent calls?  What
>> kind of number should I be expecting to get?
>>
>>
>>
>> JB
>>
>>
>>
>>
>> _______________________________________________
>> Freeswitch-users mailing list
>> Freeswitch-users at lists.freeswitch.org
>> http://lists.freeswitch.org/mailman/listinfo/freeswitch-users
>> UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users
>> http://www.freeswitch.org
>
>
> _______________________________________________
> Freeswitch-users mailing list
> Freeswitch-users at lists.freeswitch.org
> http://lists.freeswitch.org/mailman/listinfo/freeswitch-users
> UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users
> http://www.freeswitch.org