After 2 years of running flawlessly, the LSIO Docker stops working after ~30 seconds

[222]

Main: Jellyfin version: "10.8.11"
LSIO Docker - Unraid

Docker stops running after ~20-30 seconds. Before this worked flawlessly for 2+ years.

Using an Nvidia Shield as Client, but this is definitely not a client issue

I changed a few things recently but noting big or serious and I cant remmeber now. Should I do a clean install?

[xaque]

Pheres a reddit post woth solution to a similar issue as your.

https://www.reddit.com/r/jellyfin/commen...n_startup/

From the post:

If anyone else runs across this, all I did was purge the install with

sudo apt purge jellyfin --autoremove

And then reinstall and start over.

[bitmap]

If you run docker, have you tried stopping, removing, and building a new container? You haven't said whether it's vanilla docker or compose so either way it's...

Code:

docker stop <jellyfin> && docker rm -f <jellyfin> && docker run [options] <jellyfin>

Where <jellyfin> is the container name and you have to remember your run command. BEFORE YOU RUN THAT MAKE SURE YOU HAVE PERSISTENT VOLUMES. It looks like in your startup your volumes are mapped, so that shouldn't be an issue, but if your volumes are ephemeral, your data will be lost. For docker-compose:

Code:

docker compose stop <jellyfin> && docker compose rm -f <jellyfin> && docker compose up -d <jellfyin>

If you want to be really thorough, you can prune your volumes in between the container removal and bringing the container back up but you better be damn sure you have your data persisted. As for your question about a clean install...that shouldn't really ever be necessary with docker unless you screwed something up insanely bad, generally by messing with persistent files that you shouldn't be touching (e.g., databases, config files). If you weren't "making little changes" with those, it's worth troubleshooting a bit before you take the nuclear option, particularly if you have data on your instance you want to save.