2025-02-26, 09:55 PM
2025-03-02, 10:58 AM
(This post was last modified: 2025-03-02, 11:16 AM by drindor83. Edited 1 time in total.)
Have to report, the problem is ongoing, today at 09:35 it crashed again
![]() I had found this in my ubuntu log: Code: Mär 02 09:34:47 Jellyfin systemd[1]: jellyfin.service: A process of this unit has been killed by the OOM killer. It seems there is a memory leak ongoing , I found a post on github, from a user which has a similar problem with the jellyfin version after 10.9.x . My LXC has enough memory, 8GB, so thats not the probleme here, because it uses barely 3-4GB in use. In the jellfin log file I found that: Code: [2025-03-02 08:59:34.988 +00:00] [WRN] As of "03/02/2025 08:59:34 +00:00", the heartbeat has been running for "00:00:01.3583598" which is longer than "00:00:01". This could be caused by thread pool starvation.
2025-03-02, 04:48 PM
The current theory is some sort of database corruption causing Jellyfin's memory to balloon. But we haven't been able to reproduce the conditions ourselves so we don't know how to "fix" the issue....besides starting over. Not ideal, obviously.
|
|
|