from MNByChoice@midwest.social to jellyfin@lemmy.ml on 27 Jun 14:25
https://midwest.social/post/30640499
Hoping this helps someone in the future… (Likely myself when I forget all of this and reinstall.)
Setup: “OS install” of Jellyfin on Debian with package 10.10.7+deb11. Videos hosted on a FreeBSD and mounted with NFSv3.
When scanning libraries to find new videos, either the scan pauses midway through indefinitely, or completes quickly. CPUs are idle, several Jellyfin processes in state “D”. No information logged in Jellyfin logs. One error in dmesg note about nfs lock files (one every few hours).
Can partly diagnose with these commands from …freebsd.org/…/freebsd-v14-1-release-p3-nfsv3-no-… and user “tanis”:
touch /data/test/myfile
# obtain nfs lock from FreeBSD client
lockf /data/test/myfile true && echo ok || echo nok
# obtain nfs lock from Linux client
flock /data/test/myfile true && echo ok || echo nok
If the flock hangs, then the fix is to disable NFS locking. Add “nolock” to the NFS mount options.
After remounting and starting the scan, the Jellyfin processes should produce a lot of CPU load and network traffic.
Discussion: Why did the scan work completely for one and only one folder? I suspect ffprobe was not invoked for the successful run, but my other folders have home movies and are not easily detected.
threaded - newest