Jellyfin Docker Compose File

Operating System ZimaOS
1

Hey All,

New here. First of all, let me say, I am loving ZimaOS. It solves a lot of issues and definitely makes things easier. Everything you need for a relatively simple home server setup is there. It’s saved me alot of time installing separate software to get certain things done. So great Job!

I am having issues after editing the jellyfin docker compose file, I have definitely done research to find the answer, and no luck.

When I edit the docker compose.yml file, after restarting it, i get these error’s and I know it’s probably something simple, but can’t find it.. I am only editing the paths for storage etc, and I am not changing anything else.

How do I restart the docker comose file, and have it restart the proper services and have it use the default settings of the original docker-compose.yml?

I am talking about these error’s, and also the “devices” that do not start automatically after editing the file, even knowing I’m not changing any of that..

    devices:
        - /dev/dri:/dev/dri
        - /dev/vcsm:/dev/vcsm
        - /dev/vchiq:/dev/vchiq
        - /dev/video10:/dev/video10
        - /dev/video11:/dev/video11
        - /dev/video12:/dev/video12

WARN[0000] The “PGID” variable is not set. Defaulting to a blank string.
WARN[0000] The “PUID” variable is not set. Defaulting to a blank string.
WARN[0000] The “TZ” variable is not set. Defaulting to a blank string.
WARN[0000] The “AppID” variable is not set. Defaulting to a blank string.

Any help would be appreciated!

Thanks all

2

Great to hear you’re loving ZimaOS, it really does make home-server setups so much easier.

About your Jellyfin issue, those warnings usually appear when ZimaOS can’t inject the standard variables (PUID, PGID, TZ, AppID) because the compose file was edited manually.

The safest way to reset everything is to remove the Jellyfin app from the dashboard and reinstall it from the App Store, that restores the original compose file and default settings.

If you want to keep your custom storage paths, you can add them directly through App Settings → Volumes instead of editing the YAML by hand. ZimaOS will then handle the rest automatically.

After reinstalling, just restart the app from the dashboard, no need to run docker commands. That will also reload the devices like /dev/dri and /dev/video* properly.

If you prefer to fix it manually, just make sure the missing variables (PUID, PGID, TZ, AppID) are defined before restarting the service.

3

Thanks for the reply!

I finally have it all figured out. I was having a little difficulty because I am working with Proxmox installed over Debian 13. So when I would “qm set 1010 --scsi2” It was basically attaching to both the Node, and to the VM. After a few reboots, it would error and go into Maintenance Mode. I would then have to edit the /etc/fstab to comment out the MNT command. Once I done that, then everything came back online.

Also, Zima would mount that drive as “sdc” and I was looking for “sda” etc haha

Anyway, I was dumb and assumed I had to edit the docker-compose the old school manual way, once I done it through the Zima App Settings, figured out the correct paths of where I wanted things stored on my secondary drive, it all worked perfectly! So thanks!

1 other question… The time on the front page of Zima doesn’t stay at the set time settings, I found the below fix, which works for a bit, then it goes right back to being a different time than what I have set.

Enable NTP Synchronization:
This command enables ZimaOS to get the correct time from the internet.
Open the terminal on your ZimaBoard.
Type the following command and press Enter:

Code

sudo timedatectl set-ntp true

Set the Correct Time Zone:
You’ll need to know the correct time zone for your location.
To see a list of available time zones, you can use:

Code

timedatectl list-timezones

Once you find your time zone, enter it into the command:

Code

sudo timedatectl set-timezone "America/Detroit"

(Replace “America/Detroit” with your specific time zone)

Verify the Settings:
Run the timedatectl command to check if the time and time zone are now correct.
Reboot the ZimaBoard:
Restart your ZimaBoard for the changes to take full effect.

Any ideas?

4

Hey @MjTech, great to hear you have everything running smoothly. Sounds like you really went through a proper setup. When ZimaOS is running inside Proxmox on top of Debian, it can definitely get tricky with drive mappings and fstab, so nice job sorting that out.

About the time issue, NTP (Network Time Protocol) is what keeps your system clock in sync with accurate internet time servers. Without NTP, the time can slowly drift out of sync, especially if the system runs for long periods.

Try the following steps to keep your time stable:

  1. Check if NTP is active by running timedatectl status.
  2. If it shows “System clock synchronized: no”, restart the time service using systemctl restart systemd-timesyncd.
  3. Then enable it again with timedatectl set-ntp true.
  4. Reboot and confirm the timezone with timedatectl.

That should make the time stick correctly even after restarts.

I am also running ZimaOS inside Proxmox, and I noticed the VM clock can drift if the Proxmox host time is not synced properly. Make sure NTP is enabled both on the Proxmox host and inside your ZimaOS VM. Once both are synchronized, the clock will stay perfectly accurate.

Glad to see it all coming together for you.