Nick Khamis
2013-Apr-09 18:49 UTC
[asterisk-users] Asterisk Peaking and 91 Calls And not a Dime More!
Hello Everyone,
We are running some torcher tests on our * box using SIPP. The overall idea
of the test is to contact asterisk and play a g729 encoded recording. On
the asterisk side, we are initiating the echo app for the contacted
extension, simulating a two way conversation.
For some reason we cannot get past *91* calls on every test, with a lot of
resources left:
*top*
top - 14:28:45 up 1 day, 1:45, 2 users, load average: 1.09, 0.80, 0.59
Tasks: 56 total, 1 running, 55 sleeping, 0 stopped, 0 zombie
%Cpu(s): 7.6 us, 8.5 sy, 0.0 ni, 82.7 id, 0.0 wa, 0.0 hi, 1.2 si,
0.0 st
KiB Mem: 3825108 total, 164480 used, 3660628 free, 16324 buffers
KiB Swap: 2097148 total, 0 used, 2097148 free, 97404 cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
7229 root 20 0 70400 25m 5808 S 35.5 0.7 7:29.26 asterisk
*iftop*
Press H or ? for help 1.91Mb 3.81Mb 5.72Mb
7.63Mb 9.54Mb
?????????????????????????????????????????????????????????
test.example.com => 192.168.2.100
1.75Mb 1.75Mb 1.71Mb
<
1.70Mb 1.70Mb 1.66Mb
test.example.com => db.example.com
37.3Kb 37.3Kb 36.5Kb
<
10.1Kb 10.1Kb 9.87Kb
????????????????????????????????????????????????????????????
TX: cumm: 8.28MB peak: 1.79Mb rates: 1.79Mb
1.79Mb 1.74Mb
RX: 7.93MB 1.72Mb
1.71Mb 1.71Mb 1.67Mb
TOTAL: 16.2MB 3.51Mb
3.50Mb 3.50Mb 3.41Mb
The SIPP Results
------------------------------ Scenario Screen -------- [1-9]: Change
Screen --
Call-rate(length) Port Total-time Total-calls Remote-host
10.0(0 ms)/1.000s 5060 2089.21 s 20802 192.168.2.10:5060(UDP)
0 new calls during 0.000 s period 0 ms scheduler resolution
0 calls (limit 100) Peak was 91 calls, after 9 s
0 Running, 332 Paused, 0 Woken up
0 dead call msg (discarded) 0 out-of-call msg (discarded)
1 open sockets
Messages Retrans Timeout
Unexpected-Msg
INVITE ----------> 20802 0 0
100 <---------- 20802 0 0 0
180 <---------- 0 0 0 0
200 <---------- E-RTD1 20802 0 0 0
ACK ----------> 20802 0
[ NOP ]
Pause [ 8000ms] 20802 0
[ NOP ]
Pause [ 1000ms] 20802 0
BYE ----------> 20802 0 0
200 <---------- 20802 0 0 0
------------------------------ Test Terminated
--------------------------------
----------------------------- Statistics Screen ------- [1-9]: Change
Screen --
Start Time | 2013-04-09 14:08:07:797 1365530887.797642
Last Reset Time | 2013-04-09 14:42:57:025 1365532977.025339
Current Time | 2013-04-09 14:42:57:025 1365532977.025537
-------------------------+---------------------------+--------------------------
Counter Name | Periodic value | Cumulative value
-------------------------+---------------------------+--------------------------
Elapsed Time | 00:00:00:000 | 00:34:49:227
Call Rate | 0.000 cps | 9.957 cps
-------------------------+---------------------------+--------------------------
Incoming call created | 0 | 0
OutGoing call created | 0 | 20802
Total Call created | | 20802
Current Call | 0 |
-------------------------+---------------------------+--------------------------
Successful call | 0 | 20802
Failed call | 0 | 0
-------------------------+---------------------------+--------------------------
Response Time 1 | 00:00:00:000 | 00:00:00:003
Call Length | 00:00:00:000 | 00:00:09:010
------------------------------ Test Terminated
--------------------------------
Can we clear OS and * bottlenecks down into the different parts:
OS - Simple commands such as ulimit etc...
Asterisk - Startup directives that will increase whatever (i.e., allocated
memory, -p value)
before addressing hardware resources?
Your help is greatly appreciated,
Nick.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.digium.com/pipermail/asterisk-users/attachments/20130409/27171f88/attachment.htm>
Paul Belanger
2013-Apr-09 19:21 UTC
[asterisk-users] Asterisk Peaking and 91 Calls And not a Dime More!
On 13-04-09 02:49 PM, Nick Khamis wrote:> Hello Everyone, > > We are running some torcher tests on our * box using SIPP. The overall idea > of the test is to contact asterisk and play a g729 encoded recording. On > the asterisk side, we are initiating the echo app for the contacted > extension, simulating a two way conversation. > > For some reason we cannot get past *91* calls on every test, with a lot of > resources left: > > *top* > > top - 14:28:45 up 1 day, 1:45, 2 users, load average: 1.09, 0.80, 0.59 > Tasks: 56 total, 1 running, 55 sleeping, 0 stopped, 0 zombie > %Cpu(s): 7.6 us, 8.5 sy, 0.0 ni, 82.7 id, 0.0 wa, 0.0 hi, 1.2 si, > 0.0 st > KiB Mem: 3825108 total, 164480 used, 3660628 free, 16324 buffers > KiB Swap: 2097148 total, 0 used, 2097148 free, 97404 cached > > PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND > > 7229 root 20 0 70400 25m 5808 S 35.5 0.7 7:29.26 asterisk > > > > > *iftop* > > Press H or ? for help 1.91Mb 3.81Mb 5.72Mb > 7.63Mb 9.54Mb > ????????????????????????????????????????????????????????? > test.example.com => 192.168.2.100 > 1.75Mb 1.75Mb 1.71Mb > <> 1.70Mb 1.70Mb 1.66Mb > test.example.com => db.example.com > 37.3Kb 37.3Kb 36.5Kb > <> 10.1Kb 10.1Kb 9.87Kb > > ???????????????????????????????????????????????????????????? > TX: cumm: 8.28MB peak: 1.79Mb rates: 1.79Mb > 1.79Mb 1.74Mb > RX: 7.93MB 1.72Mb > 1.71Mb 1.71Mb 1.67Mb > TOTAL: 16.2MB 3.51Mb > 3.50Mb 3.50Mb 3.41Mb > > > > The SIPP Results > > > ------------------------------ Scenario Screen -------- [1-9]: Change > Screen -- > Call-rate(length) Port Total-time Total-calls Remote-host > 10.0(0 ms)/1.000s 5060 2089.21 s 20802 192.168.2.10:5060(UDP) > > 0 new calls during 0.000 s period 0 ms scheduler resolution > 0 calls (limit 100) Peak was 91 calls, after 9 s > 0 Running, 332 Paused, 0 Woken up > 0 dead call msg (discarded) 0 out-of-call msg (discarded) > > 1 open sockets > > Messages Retrans Timeout > Unexpected-Msg > INVITE ----------> 20802 0 0 > 100 <---------- 20802 0 0 0 > 180 <---------- 0 0 0 0 > 200 <---------- E-RTD1 20802 0 0 0 > > ACK ----------> 20802 0 > [ NOP ] > Pause [ 8000ms] 20802 0 > [ NOP ] > Pause [ 1000ms] 20802 0 > BYE ----------> 20802 0 0 > 200 <---------- 20802 0 0 0 > > ------------------------------ Test Terminated > -------------------------------- > > > ----------------------------- Statistics Screen ------- [1-9]: Change > Screen -- > Start Time | 2013-04-09 14:08:07:797 1365530887.797642 > > Last Reset Time | 2013-04-09 14:42:57:025 1365532977.025339 > > Current Time | 2013-04-09 14:42:57:025 1365532977.025537 > > -------------------------+---------------------------+-------------------------- > Counter Name | Periodic value | Cumulative value > -------------------------+---------------------------+-------------------------- > Elapsed Time | 00:00:00:000 | 00:34:49:227 > > Call Rate | 0.000 cps | 9.957 cps > > -------------------------+---------------------------+-------------------------- > Incoming call created | 0 | 0 > > OutGoing call created | 0 | 20802 > > Total Call created | | 20802 > > Current Call | 0 | > > -------------------------+---------------------------+-------------------------- > Successful call | 0 | 20802 > > Failed call | 0 | 0 > > -------------------------+---------------------------+-------------------------- > Response Time 1 | 00:00:00:000 | 00:00:00:003 > > Call Length | 00:00:00:000 | 00:00:09:010 > > ------------------------------ Test Terminated > -------------------------------- > > > > Can we clear OS and * bottlenecks down into the different parts: > > OS - Simple commands such as ulimit etc... > Asterisk - Startup directives that will increase whatever (i.e., allocated > memory, -p value) > > before addressing hardware resources? > > Your help is greatly appreciated, > > Nick. >You failed to say what happens when 92 channels are created. Show us your errors. -- Paul Belanger | PolyBeacon, Inc. Jabber: paul.belanger at polybeacon.com | IRC: pabelanger (Freenode) Github: https://github.com/pabelanger | Twitter: https://twitter.com/pabelanger