Zombie processes in libretime-liquidsoap container

I’ve moved the Libretime install that we have been using for several years now to a docker environment and am so far really happy with the environment, and the fine-tuning abilities through the docker-compose.yml file.
I’ve moved the the legacy HTML code to ./html the icecast.xml to ./icecast so that these are easier to maintain due to customizations like the SSL pem file reference in icecast.xml.
the postgress data and audio files are of course also volumes mounted into the respective containers.

However, the libretime-liquidsoap container is spawning of about 600 zombie processes in 24 H. this is a problem, has someone seen this too, and if so, how was it resolved??

Is there a way to start the container with --init or something like that, so that these processes can be killed. as they do take up resources on the host system.

Thanks for your thoughts and input,

Pst. Chris.

I think this has been fixed recently in the main branch.

Feel free to apply the changes to you own setup.


Thanks Joola, I’ll update the image and see where that leaves us, as for now I run a container restart for liquidsoap every day around midnight limiting the number of defunct processes.


Follow-Up the issue is still there, the defunct processes are still popping up.
for now I’ll keep on relying on the crontab fix.