[OpenSIPS-Users] Fwd: opensips is restarting after call is finished
Bogdan-Andrei Iancu
bogdan at opensips.org
Mon Dec 11 16:30:09 UTC 2023
Thanks to Liviu, we now have a fix for this :
https://github.com/OpenSIPS/opensips/commit/42f1eb7ae7dfa702602ca8aa232e9b9016b2d146
Regards,
Bogdan-Andrei Iancu
OpenSIPS Founder and Developer
https://www.opensips-solutions.com
https://www.siphub.com
On 11.12.2023 16:46, Bogdan-Andrei Iancu wrote:
> Hi Simon,
>
> Right, this explains it. Still, we need to do a fix here, to prevent
> such cases, where a backend is required for CDRs, but not configured.
>
> Regards,
> Bogdan-Andrei Iancu
>
> OpenSIPS Founder and Developer
> https://www.opensips-solutions.com
> https://www.siphub.com
> On 11.12.2023 12:53, Simon Gajski wrote:
>>
>> Dear Bogdan
>>
>> We use only accounting.
>> But there was left a line do_accounting("aaa"); in cfg script that I
>> somehow missed.
>>
>> After removal, it works as expected.
>>
>> Thank you for all help and my apologies not for detecting this earlier.
>>
>> BR
>> Simon
>>
>> Bogdan-Andrei Iancu je 11.12.2023 ob 10:49 napisal:
>>> Hi Simon,
>>>
>>> As per the trace posted by Neven, the crash is related to the AAA
>>> support (for ACC), not the DB one. So, are you using RADIUS or
>>> DIAMETER for the CDRs?
>>>
>>> Regards,
>>> Bogdan-Andrei Iancu
>>>
>>> OpenSIPS Founder and Developer
>>> https://www.opensips-solutions.com
>>> https://www.siphub.com
>>> On 11.12.2023 10:00, Simon Gajski wrote:
>>>>
>>>> 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
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>
>
>
> _______________________________________________
> 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/a87290b7/attachment-0001.html>
More information about the Users
mailing list