library.db.bak using huge amounts of space

[Jensemann]

I recently noticed that something was stealing a lot of space on my unRaid server, when I start looking into it the Appdata folder for Jellyfin was using over 1TB! It looks like the library.db is taking a backup really frequently and are eventually using huge amounts of space. How do I change the backup frequency to prevent this from happening


--- /mnt/user/appdata/jellyfin/data/data ----------------------------------------------------------------------------------------------------------
                                    /..
  232.8 MiB [#####################]  library.db
  229.6 MiB [#################### ]  library.db.bak1
  229.1 MiB [#################### ]  library.db.bak4761
  229.1 MiB [#################### ]  library.db.bak4760
  229.1 MiB [#################### ]  library.db.bak4759
  229.1 MiB [#################### ]  library.db.bak4758
  229.1 MiB [#################### ]  library.db.bak4757
 
Total disk usage:  1.0 TiB  Apparent size:  1.0 TiB  Items: 4863

[TheDreadPirate]

Jellyfin should only be keeping a single backup.  Its almost like the permissions on your directory aren't allowing Jellyfin to delete the old backups.

   

[Jensemann]

hmm odd... I had some trouble with the previous version so that may have caused the backup routine to go bananas. I looked a bit more into it now and it looks like the files has been created the last weeks where the problems with the last version started. Hopfully it was just a bug and it is solved now on the newest version.

Just in case, do you know if it is possible to edit the frequency of the backup?

[TheDreadPirate]

Honestly, I never noticed that it even made a backup. Mine is 10 days old, which is when I upgraded from 10.8.13 to 10.9.0. So I don't think it is a recurring backup job and only happens during upgrades.

[Jensemann]

That makes sense since the fault in the previous version made the container crash shortly after starting. That must have triggered a backup and since I did not shutdown the container because I was waiting for it to auto update to the newest version with a fix. The container made a backup every time it tried to restart and the process just looped.

[RAvExel]

I have Jellyfin (most recent) installed (no containers) on Ubuntu 22.04.4 LTS
I was getting errors in Nextcloud and found out my /var (100Gb) was full.
It turns out that as soon as I start Jellyfin, it produces on avarage 2 library.db.dak### files of 245M per minute!
I could run a cron job to delete them every couple of minutes but there should be a solution to this behavior.
Anyone?

[TheDreadPirate]

Is your jellyfin in a crash loop or something? Can you post your full jellyfin log to pastebin?