[OpenSIPS-Users] Segfault during load test

Bogdan-Andrei Iancu bogdan at voice-system.ro
Wed Apr 14 13:15:03 CEST 2010


Thanks to Andrei, the bug is fixed on trunk and 1.6 version

Regards,
Bogdan

Bogdan-Andrei Iancu wrote:
> Hi Takeshi,
>
> It seams there is a missing error handling in case of "no more mem".
>
> This will be fixed asap.
>
> Thanks for the report and help,
> Bogdan
>
> mayamatakeshi wrote:
>   
>> Hi Bogdan,
>>
>> On Mon, Apr 12, 2010 at 10:27 PM, Bogdan-Andrei Iancu 
>> <bogdan at voice-system.ro <mailto:bogdan at voice-system.ro>> wrote:
>>
>>     Hi Takeshi,
>>
>>     Do you still have the core file ?  If yes, could you print  "dest" and
>>     "*dest" ?
>>
>>
>>
>> (gdb) print dest
>> $1 = (void **) 0x0
>>  
>>
>>
>>     Do you get any previous errors (from the same process) ? like mem
>>     errors?
>>
>>
>> Now that you mentioned, I went to the log file and saw lots of this:
>>
>> Apr  3 05:25:19 centos53_02005 /usr/local/sbin/opensips[29732]: 
>> ERROR:tm:relay_reply: no more share memory
>> Apr  3 05:25:19 centos53_02005 /usr/local/sbin/opensips[29732]: 
>> ERROR:tm:_reply_light: failed to allocate shmem buffer
>>
>> (sorry, I should have looked into this as first thing)
>>  
>>
>>
>>     Also, can you reproduce this ?
>>
>>
>> I was load testing r.  6605 with 800 Registrations per second (32000 
>> users) and didn't saw this ever happen before (I ran this same load 
>> session several times: one of them reached 1 billion registration 
>> requests). And actually, in previous tests I was using opensips with 
>> just 256 MB of shared memory and not with "-m 1024" as above.
>> With r. 6758 so far nothing wrong happened.
>> I will update to 1.6 head and will report if problem happens again, 
>> but maybe in the above case, my SIPp instance went crazy in the other 
>> end and generated too much traffic.
>>
>> Thanks and regards,
>> takeshi
>>
>>
>>     mayamatakeshi wrote:
>>     > Hello,
>>     > I was load testing opensips-1.6 r. 6605
>>     > and I got a seg fault.
>>     > I will try updating to most recent revision of 1.6, but I thought
>>     > someone would like to take a look at this.
>>     >
>>     > regards,
>>     > takeshi
>>     >
>>     >
>>     > Core was generated by `/usr/local/sbin/opensips -m 1024'.
>>     > Program terminated with signal 11, Segmentation fault.
>>     > [New process 29728]
>>     > #0  slot_add (_s=0x2b361ce6ac10, _r=0x2b363326ca20) at hslot.c:151
>>     > 151             *dest = _r;
>>     > (gdb) bt
>>     > #0  slot_add (_s=0x2b361ce6ac10, _r=0x2b363326ca20) at hslot.c:151
>>     > #1  0x00002b3619b9ac65 in mem_insert_urecord (_d=0x2b361ce69560,
>>     > _aor=<value optimized out>, _r=0x7fff919cb740)
>>     >     at udomain.c:648
>>     > #2  0x00002b3619b9ad42 in insert_urecord (_d=0x2b361ce69560,
>>     > _aor=0x7fff919cb958, _r=0x7fff919cb740) at udomain.c:781
>>     > #3  0x00002b3619fb5769 in insert_contacts (_m=0x79b600, _c=0x79c468,
>>     > _d=0x2b361ce69560, _a=0x7fff919cb958,
>>     >     _sctx=0x7fff919cb950) at save.c:398
>>     > #4  0x00002b3619fb6592 in save (_m=0x79b600, _d=0x2b361ce69560 "",
>>     > _f=<value optimized out>, _s=0x0) at save.c:680
>>     > #5  0x000000000040e9e9 in do_action (a=0x78def8, msg=0x79b600) at
>>     > action.c:967
>>     > #6  0x00000000004114bc in run_action_list (a=<value optimized out>,
>>     > msg=0x79b600) at action.c:139
>>     > #7  0x0000000000466b24 in eval_expr (e=0x78dfd0, msg=0x79b600,
>>     > val=0x0) at route.c:1240
>>     > #8  0x000000000046659d in eval_expr (e=0x78e020, msg=0x79b600,
>>     > val=0x0) at route.c:1553
>>     > #9  0x0000000000466545 in eval_expr (e=0x78e070, msg=0x79b600,
>>     > val=0x0) at route.c:1558
>>     > #10 0x000000000040d9cd in do_action (a=0x78e2c8, msg=0x79b600) at
>>     > action.c:689
>>     > #11 0x00000000004114bc in run_action_list (a=<value optimized out>,
>>     > msg=0x79b600) at action.c:139
>>     > #12 0x000000000040fcc3 in do_action (a=0x78e888, msg=0x79b600) at
>>     > action.c:706
>>     > #13 0x00000000004114bc in run_action_list (a=<value optimized out>,
>>     > msg=0x79b600) at action.c:139
>>     > #14 0x0000000000411819 in run_top_route (a=0x78ad18,
>>     msg=0x79b600) at
>>     > action.c:119
>>     > #15 0x0000000000455ff5 in receive_msg (
>>     >     buf=0x754f20 "REGISTER sip:domain13564.com
>>     <http://domain13564.com>
>>     > <http://domain13564.com> SIP/2.0\r\nVia: SIP/2.0/UDP
>>     > 192.168.2.6:6666;rport;branch=z9hG4bK-4540-1443964-5\r\nFrom:
>>     > \"13564\" <sip:13564 at domain13564.com
>>     <mailto:sip%3A13564 at domain13564.com>
>>     > <mailto:sip%3A13564 at domain13564.com
>>     <mailto:sip%253A13564 at domain13564.com>>>;tag=4540SIPpTag001443964\r\nTo:
>>     > \"13564\" <sip:13564@"..., len=575,
>>     >     rcv_info=0x7fff919ccc10) at receive.c:162
>>     > #16 0x000000000049a304 in udp_rcv_loop () at udp_server.c:492
>>     > #17 0x0000000000429bbd in main (argc=3, argv=<value optimized
>>     out>) at
>>     > main.c:818
>>     > (gdb)
>>     >
>>     >
>>     ------------------------------------------------------------------------
>>     >
>>     > _______________________________________________
>>     > Users mailing list
>>     > Users at lists.opensips.org <mailto:Users at lists.opensips.org>
>>     > http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>>     >
>>
>>
>>     --
>>     Bogdan-Andrei Iancu
>>     www.voice-system.ro <http://www.voice-system.ro>
>>
>>
>>     _______________________________________________
>>     Users mailing list
>>     Users at lists.opensips.org <mailto:Users at lists.opensips.org>
>>     http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Users mailing list
>> Users at lists.opensips.org
>> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>>   
>>     
>
>
>   


-- 
Bogdan-Andrei Iancu
www.voice-system.ro




More information about the Users mailing list