Docker container stop responding

Describe the Bug

sometimes libreOffice stops responding
the web/admin interface does not open
url:9980 does not return ok
no new lines in the log
docker container is running
memory used 0.8 / 6 GB
when I restart the container, everything returns to normal
how to find the problem ??

Steps to Reproduce

without a specific scenario. it can happen without any particular logic

Expected Behavior

remain working

Actual Behavior

CODE don’t respond

Desktop

(Please complete the following information)

  • Collabora version:
    COOLWSD 24.04.1.4 / (git hash: f12ee1f)
    LOKit Collabora Office 24.04.1.4 (git hash: 91ae3cd)
  • OS and version:
    Ubuntu 22.04.4 LTS (GNU/Linux 5.15.0-106-generic x86_64)
    Docker version 26.0.0, build 2ae903e
  • Browser and version: [e.g. Chrome 114.0]

Additional Context

ping -t don’t show network failure

Thanks for all the versions + extra info. :slight_smile:

I also see you crossposted on the Github too:

Like thebearon said there, try out the latest version, then see if it’s still an issue.

Hmmm… just randomly? Were you working on a specific document? Was Collabora running for a very long time / idling in the background, and then you returned to it?

same with last version.
yes randomly, different documents.
not running for long time as i restart the container every day
not idling in the background.

i ended up installing CODE as linux service and it’s fine now for more than a week.

thanks

1 Like

I only experienced this in yesterday’s release 24.04.7.1 for the first time, the earlier versions worked for me as designed. However, it’s not a “sometimes it works, and sometimes it doesn’t”, but a reproducible permanent failure. Already mentioned it in the above (closed) GitHub issue.

Hello @nursoda can you share log file ? And also if possible the docker xml file which you are using so we can see what is the main cause here…

Thanks,
Darshan

I posted the docker log here. Is that what you need? If not, please tell me how: I can easily temporarily restart the container without --cap-add=SYS_ADMIN and pull whatever logs you need.