[OpenSIPS-Users] Fwd: opensips is restarting after call is finished

Simon Gajski simon at softnet.si
Mon Dec 11 08:00:35 UTC 2023


Dear Bogdan,

yes, we do use CDR via AAA.
Here are used modparams:

#### ACCounting module
loadmodule "acc.so"
modparam("acc", "early_media", 0)
modparam("acc", "report_cancels", 0)
modparam("acc", "detect_direction", 1)
modparam("acc", "db_url", "mysql://opensips:opensipsrw@localhost/opensips")
#modparam("acc", "db_table_missed_calls", "acc")
modparam("acc", "service_type", 15)
modparam("acc", "leg_fields", "db: 
CALLER->Anum;CALLEE->Bnum;TYPE->leg_type;SRCIP->srcIP;DESTIP->destIP;DIRECTION->direction")

and we do  receive CDR in mysql acc table

gdb returns
Core was generated by `/usr/sbin/opensips -P /run/opensips/opensips.pid 
-f /etc/opensips/opensips.cfg'.
Program terminated with signal SIGSEGV, Segmentation fault.
#0  0x0000000000000000 in ?? ()
(gdb) f 1
#1  0x00007f334e06aad5 in acc_aaa_cdrs (dlg=<optimized out>, 
msg=<optimized out>, ctx=0x7f334fcc52c8) at 
/build/opensips-3.4.2/modules/acc/acc.c:965
965     /build/opensips-3.4.2/modules/acc/acc.c: No such file or directory.
(gdb) p proto
$1 = {init_prot = 0x0, create_aaa_message = 0x0, destroy_aaa_message = 
0x0, send_aaa_request = 0x0, dictionary_find = 0x0, avp_add = 0x0, 
avp_get = 0x0}


BR
Simon

Bogdan-Andrei Iancu je 08.12.2023 ob 16:38 napisal:
> It looks you are using (or trying to) the CDR support via AAA.
>
> Could you share the full modparam's for the acc module and the 
> radius/diameter support?
>
> Also, in using GDB, just do:
>   f 1
>   p proto
>
> Regards,
>
> Bogdan-Andrei Iancu
>
> OpenSIPS Founder and Developer
> https://www.opensips-solutions.com
> https://www.siphub.com
>
> On 08.12.2023 16:30, Neven Grzancic wrote:
>> Hi Bogdan,
>>
>> good hit on the apport service - thanks!
>>
>> here's the trace we get from one such crash:
>>
>> [New LWP 1130]
>> [Thread debugging using libthread_db enabled]
>> Using host libthread_db library 
>> "/lib/x86_64-linux-gnu/libthread_db.so.1".
>> Core was generated by `/usr/sbin/opensips -P 
>> /run/opensips/opensips.pid -f /etc/opensips/opensips.cfg'.
>> Program terminated with signal SIGSEGV, Segmentation fault.
>> #0  0x0000000000000000 in ?? ()
>> (gdb) bt full
>> #0  0x0000000000000000 in ?? ()
>> No symbol table info available.
>> #1  0x00007f334e06aad5 in acc_aaa_cdrs (dlg=<optimized out>, 
>> msg=<optimized out>, ctx=0x7f334fcc52c8)
>>     at /build/opensips-3.4.2/modules/acc/acc.c:965
>>         i = <optimized out>
>>         j = <optimized out>
>>         ret = 6
>>         res = -1
>>         nr_leg_vals = <optimized out>
>>         start_time = {tv_sec = 1702044564, tv_usec = 417047}
>>         core_s = {s = 0x7f33705e9878 "\006", len = 101}
>>         leg_s = {s = 0x0, len = <optimized out>}
>>         extra_s = {s = 0x0, len = <optimized out>}
>>         send = 0x0
>>         offset = <optimized out>
>>         av_type = 32563
>>         r_stat = <optimized out>
>>         locked = 0
>>         duration = 1338791072
>>         ms_duration = 0
>>         extra = <optimized out>
>>         error = <optimized out>
>>         __FUNCTION__ = "acc_aaa_cdrs"
>> #2  0x00007f334e078ade in acc_cdr_cb (t=0x0, type=5, ps=0x7ffdb12e09e0)
>>     at /build/opensips-3.4.2/modules/acc/acc_logic.c:986
>>         ctx = 0x7f334fcc52c8
>>         dlg = 0x7f334fcc54e0
>>         __FUNCTION__ = "acc_cdr_cb"
>> [...]
>>
>>
>> On 08.12.23 15:16, Bogdan-Andrei Iancu wrote:
>>> Simon, I do not need the corefile, usually it is huge - as per web 
>>> instructions (see the prev link), extract the backtrace (using gdb) 
>>> and share it with me.
>>>
>>> Regards,
>>>
>>> Bogdan-Andrei Iancu
>>>
>>> OpenSIPS Founder and Developer
>>> https://www.opensips-solutions.com
>>> https://www.siphub.com
>>>
>>> On 08.12.2023 16:13, Simon Gajski wrote:
>>>> Hi Bogdan
>>>>
>>>> you were right
>>>> we have the core file now
>>>>
>>>> Please let me know where to send it (gzipped is approx 2MB)
>>>>
>>>>
>>>> BR
>>>> Simon
>>>>
>>>> Bogdan-Andrei Iancu je 08.12.2023 ob 14:17 napisal:
>>>>> Hi Simon,
>>>>>
>>>>> Maybe you have some apport service running on the machine, moving 
>>>>> the core file.
>>>>>
>>>>> Regards,
>>>>>
>>>>> Bogdan-Andrei Iancu
>>>>>
>>>>> OpenSIPS Founder and Developer
>>>>> https://www.opensips-solutions.com
>>>>> https://www.siphub.com
>>>>>
>>>>> On 08.12.2023 13:35, Neven Grzancic wrote:
>>>>>> Dear Bogdan,
>>>>>>
>>>>>> we tried following 
>>>>>> https://opensips.org/Documentation/TroubleShooting-Crash and 
>>>>>> trying different things, and even though the syslog shows that 
>>>>>> the core file was dumped:
>>>>>>
>>>>>> Dec  8 12:25:13 sbc2 /usr/sbin/opensips[4363]: 
>>>>>> CRITICAL:core:sig_usr: segfault in process pid: 4363, id: 7
>>>>>> Dec  8 12:25:13 sbc2 kernel: [ 3394.269733] opensips[4363]: 
>>>>>> segfault at 0 ip 0000000000000000 sp 00007ffdc458f6a8 error 14 in 
>>>>>> opensips[55816fbd0000+1f000]
>>>>>> Dec  8 12:25:18 sbc2 mariadbd[922]: 2023-12-08 12:25:18 400 
>>>>>> [Warning] Aborted connection 400 to db: 'opensips' user: 
>>>>>> 'opensips' host: 'localhost' (Got an error reading communication 
>>>>>> packets)
>>>>>> Dec  8 12:25:18 sbc2 /usr/sbin/opensips[4356]: 
>>>>>> INFO:core:handle_sigs: child process 4363 exited by a signal 11
>>>>>> Dec  8 12:25:18 sbc2 /usr/sbin/opensips[4356]: 
>>>>>> INFO:core:handle_sigs: core was generated
>>>>>> Dec  8 12:25:18 sbc2 /usr/sbin/opensips[4356]: 
>>>>>> INFO:core:handle_sigs: terminating due to SIGCHLD
>>>>>> Dec  8 12:25:18 sbc2 /usr/sbin/opensips[4359]: INFO:core:sig_usr: 
>>>>>> signal 15 received
>>>>>> Dec  8 12:25:18 sbc2 /usr/sbin/opensips[4360]: INFO:core:sig_usr: 
>>>>>> signal 15 received
>>>>>>
>>>>>> the core does not get generated, or at least not in the directory 
>>>>>> provided via options at runtime (-w /tmp). The issue is always 
>>>>>> repeatable (the daemon crashes after every finished call), if 
>>>>>> there are more active calls at the moment of the crash, whose 
>>>>>> calls remain active and proceed normally until termination (which 
>>>>>> again causes a crash). We also tried enabling debug mode, but in 
>>>>>> that case the process does not start successfully and runs into a 
>>>>>> timeout. In any case, any additional help or suggestion on how to 
>>>>>> successfully generate the core dump would be much appreciated, as 
>>>>>> we seem to have ran into a wall here...
>>>>>>
>>>>>> On 08.12.23 12:26, Simon Gajski wrote:
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> -------- Posredovano sporočilo --------
>>>>>>> Zadeva:     Re: [OpenSIPS-Users] opensips is restarting after 
>>>>>>> call is finished
>>>>>>> Datum:     Thu, 7 Dec 2023 17:52:37 +0200
>>>>>>> Od:     Bogdan-Andrei Iancu <bogdan at opensips.org>
>>>>>>> Za:     Simon Gajski <simon at softnet.si>, OpenSIPS users mailling 
>>>>>>> list <users at lists.opensips.org>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> Hi Simon,
>>>>>>>
>>>>>>> Well, the log is very self explanatory:
>>>>>>> Dec 7 15:59:47 sbc2 /usr/sbin/opensips[28488]: 
>>>>>>> INFO:core:handle_sigs: child process 28495 exited by a signal 11
>>>>>>> Dec  7 15:59:47 sbc2 /usr/sbin/opensips[28488]: 
>>>>>>> INFO:core:handle_sigs: core was generated
>>>>>>> Dec  7 15:59:47 sbc2 /usr/sbin/opensips[28488]: 
>>>>>>> INFO:core:handle_sigs: terminating due to SIGCHLD
>>>>>>>
>>>>>>> Or shortly, you opensips just crashed :). So see 
>>>>>>> https://opensips.org/Documentation/TroubleShooting-Crash for how 
>>>>>>> to report further.
>>>>>>>
>>>>>>> Regards,
>>>>>>>
>>>>>>> Bogdan-Andrei Iancu
>>>>>>>
>>>>>>> OpenSIPS Founder and Developer
>>>>>>> https://www.opensips-solutions.com
>>>>>>> https://www.siphub.com
>>>>>>>
>>>>>>> On 07.12.2023 17:17, Simon Gajski via Users wrote:
>>>>>>>>
>>>>>>>> Hi
>>>>>>>>
>>>>>>>> we are running opensips 3.4.2
>>>>>>>> on Ubuntu 22.04.3 LTS
>>>>>>>> and also use RTPengine Version: 10.5.0.0+0~mr10.5.0.0 
>>>>>>>> git-master-74075f63
>>>>>>>>
>>>>>>>> Opensips acts as SBC with RTP engine enabled.
>>>>>>>>
>>>>>>>> Calls are working fine, however after each call is finished, 
>>>>>>>> bellow action happens.
>>>>>>>>
>>>>>>>> We had same problem with opensips 3.2 and Ubuntu 20. So we did 
>>>>>>>> upgrade to latest stable release, and it is the same.
>>>>>>>>
>>>>>>>> Is this opensips script configuration issue or a bug? And how 
>>>>>>>> could I fix it?
>>>>>>>>
>>>>>>>> [....]
>>>>>>>>
>>>>>>>> Thanks for all tips.
>>>>>>>>
>>>>>>>> BR
>>>>>>>> Simon
>>>>>>>>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> Users mailing list
>>>>>>>> Users at lists.opensips.org
>>>>>>>> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>>>>>>>
>>>>>>
>>>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensips.org/pipermail/users/attachments/20231211/7f9eb425/attachment.html>


More information about the Users mailing list