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

Simon Gajski simon at softnet.si
Mon Dec 11 10:53:10 UTC 2023 

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
>>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>
>>>>
>>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensips.org/pipermail/users/attachments/20231211/59d33861/attachment-0001.html>

More information about the Users mailing list