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 <[email protected]>
Za: Simon Gajski <[email protected]>, OpenSIPS users
mailling list <[email protected]>
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
[email protected]
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
_______________________________________________
Users mailing list
[email protected]
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
