Jellyfind stopped recognizing my library after Linux update

[Daniel_King]

For testing (and out of convenience) I installed Ubuntu server in a VM, there I set up your service file, then I installed Docker + Portainer and I wanted to set up Jellyfin.

Everything worked more or less smooth, but when I try to deploy the following stack:

Code:

services:
  jellyfin:
    image: jellyfin/jellyfin:latest
    container_name: jellyfin
    user: 1000:1000
    network_mode: 'bridge'
    ports:
      - 8096:8096/tcp
      - 1900:1900/udp
    volumes:
      - /var/jellyfin/config:/config
      - /var/jellyfin/cache:/cache
      - /media/movies:/movies
      - /media/shows:/shows
    restart: 'unless-stopped'

I receive this error:
Failed to deploy a stack: compose up operation failed: Error response from daemon: error while creating mount source path '/media/movies': mkdir /media/movies: file exists



The above service is mounting my Shared Drive from Google under "/media/movies".
This works, and I can see my files.

ls -l shows me the following for the mounted movies folder:

Code:

uid=1000(test) gid=1000(test) groups=1000(test),4(adm),24(cdrom),27(sudo),30(dip),46(plugdev),101(lxd)

I also tried this image: linuxserver/jellyfin:latest and beside deploying a stack I also tried it as Docker YML or in Portainer via the GUI.
Always the same error.

Any idea what am I doing wrong?

[TheDreadPirate]

The service file I provided specifies UID 1005 and GID 1007. I'm assuming you changed those to 1000:1000.

Can you share the output of "sudo cat /proc/mounts"? Specifically the line(s) for your rclone mount.

[Daniel_King]

Yes, I changed the UID and GID to 1000.


output of "sudo cat /proc/mounts":

shows: /media/shows fuse.rclone rw,nosuid,nodev,relatime,user_id=1000,group_id=1000 0 0
movies: /media/movies fuse.rclone rw,nosuid,nodev,relatime,user_id=1000,group_id=1000 0 0

[TheDreadPirate]

I'm a bit ignorant about how rclone works and the very specifics of your setup. But why have two rclone mounts? Doesn't it make more sense to have one rclone mount for all your media? Then in Docker you can break it down further if you need different permissions. If you don't need different permissions for each folder in your rclone, just pass in the entire rclone mount into Jellyfin and just select the subfolder when setting up the library?

/media/rclone

Then in that have a movies and shows folder, etc., and just pass in /media/rclone in the docker config.

Another thought occurred, where is the rclone storage mounted? On the physcal host? Or in the VM?

[Daniel_King]

Well, it might look like that the two rclone mounts don't make sense right now.
But in the long term, there should be more rclone mounts, and each rclone mount corresponds to a shared drive (Google), and these shared drives are managed separately.

But even if I would just mount one shared drive, it makes no difference.

The rclone storage is mounted in the VM. I'm running my Linux VM on a Windows PC. Actually, beside running VirtualBox, nothing happens on the Windows side.

[Daniel_King]

For testing purpose, I mounted my folder directly via rclone without the service, and it worked:

sudo rclone mount movies /media/movies2 \
--config /home/test/.config/rclone/rclone.conf \
--allow-other \
--default-permissions \
--dir-cache-time 5000h \
--drive-pacer-min-sleep 10ms \
--drive-pacer-burst 200 \
--poll-interval 30s \
--cache-dir=/opt/rclone/cache \
--drive-chunk-size 32M \
--vfs-cache-mode full \
--vfs-cache-max-size 350G \
--vfs-cache-max-age 15m \
--vfs-cache-poll-interval 5m \
--vfs-read-ahead 2G \
--vfs-read-chunk-size 32M \
--vfs-read-chunk-size-limit off \
--vfs-write-back 10m \
--bwlimit 8650kff \
--transfers 3 \
--bwlimit-file 100M \
> /dev/null 2>&1 &


1) Any idea why it works like this but not with the service?
2) When setting "--vfs-cache-max-size 350G \", does this mean I need to have at least 350G of free storage on my system?