[OpenSIPS-Devel] FIFO Bug / lb_reload
Alec Fett
al at boberdoo.com
Wed Aug 12 16:06:48 CEST 2009
I've figured out the problem is not related to the number of calls, but
actually occurs after using lb_status to change the status of a gateway, and
then later reloading the list with lb_reload. The following occurred after
issuing `lb_status 1 0` and then later trying to lb_reload
Thanks!
root at phonev2:~# opensipsctl fifo ps
Process:: ID=0 PID=14538 Type=attendant
Process:: ID=1 PID=14541 Type=SIP receiver udp:XXX.XXX.XXX.XXX:5060
Process:: ID=2 PID=14542 Type=SIP receiver udp:XXX.XXX.XXX.XXX:5060
Process:: ID=3 PID=14543 Type=time_keeper
Process:: ID=4 PID=14544 Type=timer
Process:: ID=5 PID=14545 Type=MI XMLRPC
Process:: ID=6 PID=14546 Type=MI FIFO
root at phonev2:~# ps auxw | grep opensips
root 14538 0.0 0.0 101184 5072 ? S 13:02 0:00
/usr/local/sbin/opensips -P /var/run/opensips.pid
root 14541 0.0 0.0 101184 2348 ? S 13:02 0:00
/usr/local/sbin/opensips -P /var/run/opensips.pid
root 14542 0.0 0.0 101184 2312 ? S 13:02 0:00
/usr/local/sbin/opensips -P /var/run/opensips.pid
root 14543 0.0 0.0 101184 1028 ? S 13:02 0:00
/usr/local/sbin/opensips -P /var/run/opensips.pid
root 14544 0.3 0.0 101184 1992 ? R 13:02 0:34
/usr/local/sbin/opensips -P /var/run/opensips.pid
root 14545 0.0 0.0 101184 1464 ? S 13:02 0:00
/usr/local/sbin/opensips -P /var/run/opensips.pid
root 14546 0.2 0.0 101192 1624 ? R 13:02 0:24
/usr/local/sbin/opensips -P /var/run/opensips.pid
root 16480 0.0 0.0 10948 2156 pts/0 S+ 15:55 0:00 /bin/bash
/usr/local/sbin/opensipsctl fifo lb_reload
root 16499 0.0 0.0 10948 1324 pts/0 S+ 15:55 0:00 /bin/bash
/usr/local/sbin/opensipsctl fifo lb_reload
root 16553 0.0 0.0 5160 776 pts/1 R+ 15:57 0:00 grep
opensips
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
14544 root 20 0 98.8m 1992 1200 S 36 0.0 0:38.26 opensips
14546 root 20 0 98.8m 1624 836 R 20 0.0 0:26.72 opensips
Bogdan-Andrei Iancu wrote:
>
> Hi Alec,
>
> I'm trying to reproduce the issue (generating traffic with sipp - 20K
> calls), but nothing so far...no blocking of the lb_reload.
>
> Please try to followings:
>
> 1) start opensips
>
> 2) do "opensipsctl fifo ps " to get a list with all the processes
>
> 3) when blocking, do a "ps -auxw | grep opensips" to see the remaining
> processes ; also do a top to see if any opensips proc is using too much
> CPU.
>
> Regards,
> Bogdan
>
> Alec Fett wrote:
>> I've confirmed, I can run lb_reload multiple times, right after starting
>> openSIPS, it is just after OpenSIPS has been running for a while that it
>> hangs.
>>
>> -Alec
>> Alec Fett wrote:
>>
>>> lb_reload called from fifo ('opensipsctl fifo lb_reload') hangs if you
>>> try running it after OpenSIPS has been running for a while.
>>>
>>> opensipsctl fifo lb_reload
>>> database engine 'MYSQL' loaded
>>> Control engine 'FIFO' loaded
>>> entering fifo_cmd lb_reload
>>>
>>> All subsequent FIFO commands hang in the same manner..
>>>
>>> opensipsctl fifo lb_list
>>> database engine 'MYSQL' loaded
>>> Control engine 'FIFO' loaded
>>> entering fifo_cmd lb_list
>>>
>>> I also have high CPU .8 load (with no calls) after a while if I don't
>>> restart OpenSIPS... I'm pretty sure this is related, because I haven't
>>> seen this before.
>>>
>>> Regards,
>>>
>>> Alec
>>>
>>> _______________________________________________
>>> Devel mailing list
>>> Devel at lists.opensips.org
>>> http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
>>>
>>>
>>
>>
>> _______________________________________________
>> Devel mailing list
>> Devel at lists.opensips.org
>> http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
>>
>>
>
>
> _______________________________________________
> Devel mailing list
> Devel at lists.opensips.org
> http://lists.opensips.org/cgi-bin/mailman/listinfo/devel
>
>
--
View this message in context: http://n2.nabble.com/FIFO-Bug---lb_reload-tp3310977p3430964.html
Sent from the OpenSIPS - Devel mailing list archive at Nabble.com.
More information about the Devel
mailing list