+- Jellyfin Forum (https://forum.jellyfin.org)
+-- Forum: Support (https://forum.jellyfin.org/f-support)
+--- Forum: Troubleshooting (https://forum.jellyfin.org/f-troubleshooting)
+--- Thread: Excessive RAM consumption (/t-excessive-ram-consumption)
RE: Excessive RAM consumption - TheDreadPirate - 2025-03-17
The running theory is that there is some database issue that is causing the bloat. But it is hard to specifically point to a root cause if we're not able to reliably reproduce the issue.
RE: Excessive RAM consumption - gizmomelb - 2025-03-17
I'm running a docker version of Jellyfin (10.10.6) on my Synology (DSM 7.x) and on boot it uses about 3.2GB of RAM, after a few days it is up to 6.4GB of RAM. Happy to look for log files / other app versions if needed.
RE: Excessive RAM consumption - thymon - 2025-03-17
I rebooted at 4:00 AM, and it's already using 6.4GB of RAM by 10:00 AM.
Yesterday I tested :
Code:
docker inspect --format='{{.State.Pid}}' Jellyfin
2513672Code:
smem -t
PID User Command Swap USS PSS RSS
1098295 root bash 0 1672 2408 3800
1100781 root /usr/bin/python3 /usr/bin/s 0 15956 16871 18484
1098152 root /usr/lib/jellyfin-ffmpeg/ff 0 27792 28506 29928
1 root /jellyfin/jellyfin 0 11782316 11782767 11783748
-------------------------------------------------------------------------------
4 1 0 11827736 11830552 11835960RE: Excessive RAM consumption - TheDreadPirate - 2025-03-17
You can try enabling debug logs then sharing a full day's worth. You will likely need to zip the log(s) before attaching them to a forum post. Again, still very hard to figure out without a reliable way to reproduce on our own systems.
RE: Excessive RAM consumption - thymon - 2025-03-17
Okay, I'll do that.
I also tried deleting the Docker container and recreating it (while keeping my appdata, of course), but that didn't change anything.
For me, the solution would be to migrate to another repository like "lscr.io/linuxserver/jellyfin".
However, I would need guidance on the steps to follow because simply copying the appdata didn't work—the new linuxserver database didn't like it. 😆😜
Alternatively, I could wait for version 10.10.10 , which should introduce a backup feature, making the migration to another repository much easier.
RE: Excessive RAM consumption - gizmomelb - 2025-03-18
10.10.10? there's a roadmap of upcoming features/work?
RE: Excessive RAM consumption - Efficient_Good_5784 - 2025-03-18
(2025-03-18, 01:36 AM)gizmomelb Wrote: 10.10.10? there's a roadmap of upcoming features/work?You can check this out:
https://github.com/orgs/jellyfin/projects/55
RE: Excessive RAM consumption - thymon - 2025-03-18
(2025-03-18, 01:36 AM)gizmomelb Wrote: 10.10.10? there's a roadmap of upcoming features/work?
https://jellyfin.org/posts/jellyfin-release-10.10.0/
There is actually a procedure for backups.
Quote:The Future
Long-term, we have plans to provide an official backup and restore plugin bundled with Jellyfin, however this requires the completion of our EFCore rewrite, which is currently slated for our next major release 10.11.0. Once that becomes available, this document will be updated to reflect the process using that plugin.
10.11.0! I got the release wrong. 😆
RE: Excessive RAM consumption - TheDreadPirate - 2025-03-18
(2025-03-17, 05:05 PM)thymon Wrote: Okay, I'll do that.
I also tried deleting the Docker container and recreating it (while keeping my appdata, of course), but that didn't change anything.
For me, the solution would be to migrate to another repository like "lscr.io/linuxserver/jellyfin".
However, I would need guidance on the steps to follow because simply copying the appdata didn't work—the new linuxserver database didn't like it. 😆😜
Alternatively, I could wait for version 10.10.10 , which should introduce a backup feature, making the migration to another repository much easier.
Are you using the official image now? If so, the linuxserver image uses a slightly different config directory layout to the point that the official and linuxserver images are not interchangeable.
RE: Excessive RAM consumption - thymon - 2025-03-18
Yes, I’m using the official repository. And I understood that the two repositories are not interchangeable when I tried to test.