Plex Docker Image becomes unhealthy when a database optimisation is complete

Plex Media Server v1.24.5.5173-8dcc73a59 - Docker Docker Container x86_64 - build: linux-x86_64 Ubuntu Server 18.04

Linux version: 5.4.0-89-generic, language: en-US

It is time I asked for help for this issue as it has happened 5 times in the past 2 months. I have come to the conclusion that this issue is triggered when an optimisation is ran or completed. I have a script that creates/edits collections and runs at 10:00, 16:00, 19:00 and when it finishes it optimises the plex database. The past 5 times it has happened it always happened when the script was ran. I optimise that often as I use translated titles and for them to be searchable the db needs to be optimised (elan has told me this)

Anyhow, today it happened again and this time it happened during the third run at 19:00. The script finished around 19:01:48 (it will be easier to detect the issue). I think the container turned unhealthy around 19:10:00 cause that’s when the debug log ends and it starts again when I docker restarted the container.

I would like help if possible as I do not know what to do. For any plex employees/moderators reading please pm me so I can provide you with the logs. Do not want to post them as they have so many sensitive info and I dont think I can remove them all. I trust you. I happen to have today’s logs and fortunately I kept the logs it happened another time as well just in case.

Is the /config on a local file system or network share somewhere?

thanks for the response. it on a local file system.

That rules that out.

Most common case of DB corruption during vacuuming (the optimization) is because it can’t get an exclusive lock on the file.

One of the most common ways to not get a lock (guaranteed) is network file share.

The /config is on an EXT or XFS filesystem? (NTFS won’t support what’s needed)

May I see the Logs ZIP please ?

yeah ext4 filesystem

can you create a room so I can send you the logs in private? :grimacing:

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.