[OpenSIPS-Users] OpenSIPS stops responding - timer lag + recv-queue buildup

Bogdan-Andrei Iancu bogdan at opensips.org
Mon Dec 29 09:35:01 UTC 2025


Hi,

without the output of the "trap" cmd there is not much I can do here.

Regards,

Bogdan-Andrei Iancu

OpenSIPS Founder and Developer
   https://www.opensips-solutions.com
   https://www.siphub.com

On 10.12.2025 07:40, inderjeet sharma wrote:
> Hi Bogdan
>
> Could you please help me with this.
>
>
> On Tue, 2 Dec 2025 at 7:19 pm, inderjeet sharma 
> <inder.itpro at gmail.com> wrote:
>
>     Hi Bogdan
>
>     OpenSIPS is running inside a Docker container, and there is no
>     apport service within the container. The apport service on the
>     host was running, but I’ve already disabled it. However, I still
>     don’t see any core dump files being generated. Is there anything
>     else I should check?
>     Additionally:
>
>     root at opensips-3-1:/# netstat -tupn | grep 8089 | wc -l
>     472
>
>     root at opensips-3-1:/# opensips-cli -x mi ul_dump | grep Socket | wc -l
>     510
>
>     root at opensips-3-1:/# opensips-cli -x mi list_tcp_conns | grep
>     "Remote" | wc -l
>     500
>
>     I have around 500 registered users on OpenSIPS, and most of them
>     are using WSS/TCP. I want to know how many tcp_workers and
>     udp_workers I should configure. Currently, I have udp_workers=20
>     and I haven’t set any value for tcp_workers in the configuration.
>     Also are there any other value I should check?
>
>
>     On Mon, 24 Nov 2025 at 12:50 pm, Bogdan-Andrei Iancu
>     <bogdan at opensips.org> wrote:
>
>         Hi,
>
>         Maybe your OS has `apport` service, which is a core dump file
>         manager ?
>
>         Regards,
>
>         Bogdan-Andrei Iancu
>
>         OpenSIPS Founder and Developer
>            https://www.opensips-solutions.com
>            https://www.siphub.com
>
>         On 22.11.2025 12:52, inderjeet sharma wrote:
>>         Thanks Bogdan,
>>
>>         After more analysis, I noticed that the recv-queue buildup
>>         occurs only after the OpenSIPS service crashes. I’ve looked
>>         in all expected directories, but the core dump file is not
>>         present.
>>
>>         Below are the logs showing the sequence of events during the
>>         crash:
>>         Nov 22 11:46:58 opensips-3-1 /usr/local/sbin/opensips[1595]:
>>         CRITICAL:core:sig_usr: segfault in process pid: 1595, id: 46
>>         Nov 22 11:46:58 opensips-3-1 /usr/local/sbin/opensips[1503]:
>>         INFO:core:handle_sigs: child process 1595 exited by a signal 11
>>         Nov 22 11:46:58 opensips-3-1 /usr/local/sbin/opensips[1503]:
>>         INFO:core:handle_sigs: core was generated
>>         Nov 22 11:46:58 opensips-3-1 /usr/local/sbin/opensips[1503]:
>>         INFO:core:handle_sigs: terminating due to SIGCHLD
>>         Nov 22 11:46:58 opensips-3-1 /usr/local/sbin/opensips[1550]:
>>         INFO:core:sig_usr: signal 15 received
>>         Nov 22 11:46:58 opensips-3-1 /usr/local/sbin/opensips[1552]:
>>         INFO:core:sig_usr: signal 15 received
>>         Nov 22 11:46:58 opensips-3-1 /usr/local/sbin/opensips[1551]:
>>         INFO:core:sig_usr: signal 15 received
>>         Nov 22 11:46:58 opensips-3-1 /usr/local/sbin/opensips[1503]:
>>         INFO:core:shutdown_opensips: process 1(1550) [MI FIFO]
>>         terminated, still waiting for 51 more
>>         Nov 22 11:46:58 opensips-3-1 /usr/local/sbin/opensips[1503]:
>>         INFO:core:shutdown_opensips: process 2(1551) [time_keeper]
>>         terminated, still waiting for
>>
>>         On Tue, Nov 11, 2025 at 4:13 PM Bogdan-Andrei Iancu
>>         <bogdan at opensips.org> wrote:
>>
>>             Hi,
>>
>>             Have you checked the internal load of opensips, by
>>             running the `ps` and `get_statistics load:` MI cmds ?
>>
>>             I suspect some blocking I/O operation is blocking the
>>             OpenSIPS timer processes, resulting in the reported
>>             warning logs.
>>
>>             Also, if possible (not sure in docker) you should run
>>             `opensips-cli -x trap` to get an instant x-ray of what
>>             the opensips processes are doing, maybe this will help
>>             spotting the problem.
>>
>>             Regards,
>>
>>             Bogdan-Andrei Iancu
>>
>>             OpenSIPS Founder and Developer
>>                https://www.opensips-solutions.com
>>                https://www.siphub.com
>>
>>             On 04.11.2025 16:10, inderjeet sharma wrote:
>>>             version:
>>>             opensips 3.1.11
>>>             (x86 64/linux)
>>>             ٧-
>>>             flags: STATS: On,
>>>             DISABLE_NAGLE, USE MCAST,
>>>             PKG_MALLOC, Q_MALLOC, F_MALLOC, HP_MALLOC, DBG_MALLOC,
>>>             FAST_LOCK ADAPTIVE_WAIT
>>>             ADAPTIVE WAIT LOOPS=1024, MAX_RECV_BUFFER_SIZE 262144,
>>>             MAX_LISTEN 16, MAX_URI_SIZE 1024,
>>>             BUF_SIZE 65535
>>>             poll method support: poll, epoll, sigio_rt, select. git
>>>             revision: c51486e3a
>>>             main.c compiled on 09:20:17 Nov 20 2024 with gcc 11
>>>
>>>             OpenSSL 3.0.2
>>>
>>>             PRETTY_NAME="Ubuntu 22.04.5 LTS"
>>>             NAME="Ubuntu"
>>>             VERSION_ID="22.04"
>>>             VERSION="22.04.5 LTS (Jammy Jellyfish)™
>>>             VERSION CODENAME=] ammy
>>>
>>>
>>>
>>>             On Tue, 4 Nov 2025 at 1:50 pm, inderjeet sharma
>>>             <inder.itpro at gmail.com> wrote:
>>>
>>>                 Hi all,
>>>
>>>                 I've been running OpenSIPS (v3.x) in a Docker
>>>                 container for the past 1-2 years as a WebSocket
>>>                 (WSS) registrar handling SIP registrations.
>>>                 It has been stable until recently — for the past few
>>>                 weeks, I'm seeing severe performance stalls.
>>>
>>>                 • OpenSIPS suddenly stops responding to SIP REGISTER
>>>                 requests.
>>>                 • ss -unap shows high Recv-Q values on the listening
>>>                 ports (UDP + WSS).
>>>                 • The log is flooded with warnings like:
>>>                 08:55:31 opensips-3-1 /usr/local/sbin/opensips[79638]:
>>>                 WARNING: core: utimer_ticker: utimer task
>>>                 <tm-utimer> already
>>>                 117150 ms ago (now
>>>                 336619940 ms), delaying execution
>>>                 Nov 3 08:55:31
>>>                 opensips-3-1/usr/local/sbin/opensips[79638]:
>>>                 WARNING: core: utimer_ticker: timer task ‹tm-utimer>
>>>                 already
>>>                 scheduled 117280 ms ago (now
>>>                 336620070 ms), delaying execution
>>>                 Nov
>>>                 308:55:32
>>>                 opensips-3-1/us/local/sbin/opensips[79638]: WARNING:
>>>                 core: utimer_ticker: utimer task <tm-utimer> already
>>>                 scheduled 117350 ms ago (now
>>>                 336620140 ms), delaying execution
>>>                 NOT
>>>                 3 08:55:32
>>>                 opensips-3-1/usr/local/sbin/opensips[79638]:
>>>                 WARNING: core: timer_ ticker: timer task <1b-pinger>
>>>                 already scheduled ms ago (now 336620140 ms),
>>>                 delaying execution
>>>                 08:55:32 opensips-3-1/us/local/sbin/opensips[79638]:
>>>                 WARNING: core: timer_ticker: timer task ‹nh-timer>
>>>                 already scheduled ms ago (now 336620140 ms),
>>>                 delaying execution
>>>                 308:55:32
>>>                 opensips-3-1/us/local/sbin/opensips[79638]: WARNING:
>>>                 core:timer_ticker: timer task <pg-chk-timer> already
>>>                 scheduled 54750 ms ago (now 336620140 ms), delaying
>>>                 execution
>>>                 NOV
>>>                 308:55:32
>>>                 opensips-3-1/usr/local/sbin/opensips[79638]:
>>>                 WARNING: core: timer_ticker: timer task ‹ul-timer>
>>>                 already scheduled
>>>                 79650 ms ago (now 336620140 ms), delaying execution
>>>                 Nov
>>>                 308:55:32 opensips-3-1/us/local/sbin/opensips
>>>                 [79638]: WARNING: core: timer_ticker: timer task
>>>                 <d1g-reinvite-pinger> already
>>>                 scheduled 54750 ms ago (now 336620140 ms), delaying
>>>                 execution
>>>                 Nov 308:55:32
>>>                 opensips-3-1/us/local/sbin/opensips[79638]: WARNING:
>>>                 core:timer_ticker: timer task ‹dig-options-pinger>
>>>                 already
>>>                 308:55:32
>>>                 opensips-3-1/us/local/sbin/opensips[79638]: WARNING:
>>>                 core: timer_ticker: timer task ‹dlg-timer> already
>>>                 scheduled
>>>                 115460 ms ago (now 336620140 ms), delaying execution
>>>                 Nov 308:55:32
>>>                 opensips-3-1/usr/local/sbin/opensips[79638]:
>>>                 WARNING: core: timer_ticker: timer task <EBR
>>>                 timeout> already scheduled 54750 ms ago (now
>>>                 336620140 ms), skipping execution
>>>                 308:55:32
>>>                 opensips-3-1/usr/local/sbin/opensips[79638]:
>>>                 WARNING: core: timer_ticker: timer task ‹tm-timer>
>>>                 already
>>>
>>>                 • Killing the process or restarting the container
>>>                 sometimes does not help immediately.
>>>                 • CPU and memory usage inside Docker appear normal.
>>>
>>>                 Thanks
>>>                 Inderjeet
>>>
>>>
>>>             _______________________________________________
>>>             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/20251229/71dd659d/attachment-0001.html>


More information about the Users mailing list