Does anyone have a compose.yaml for an Nvidia GPU that works that they would like to share? Here’s my current file, it gives a white screen with “server error” on it: https://pastebin.com/AaV17cTz

I went through Jellyfin’s instructions on setting a GPU up, but the instructions weren’t clear (in my opinion) so who knows if it’s correct. I installed some Nvidia tools as a prerequisite and ‘nvidia-smi’ shows the card. I attached my Jellyfin settings from before it self-destructed according to Nvidia’s transcoding matrix (which also wasn’t descriptive enough in my opinion), do they look right for a 2080?

Update: after making this post, and changing nothing, it suddenly works

  • jia_tan@lemmy.blahaj.zone
    link
    fedilink
    English
    arrow-up
    4
    arrow-down
    1
    ·
    22 hours ago

    I have an intel igpu. It was hella painful to pass through the guy into a normal container and I never figured it out. I just ended up running the container with the —privileged flag. QuickSync hwaccel works fine now, I assume it would be the same for NVENC, since the flag basically just passes everything to the container.

    • ohshit604@sh.itjust.works
      link
      fedilink
      English
      arrow-up
      1
      ·
      edit-2
      2 hours ago

      Jellyfin isn’t the most secure piece of software out there, I would avoid giving it permissions it doesn’t need.


      Step 1) Check /dev/dri for the GPU

      user@debian:~/compose$ ls /dev/dri
      total 0
      drwxr-xr-x  3 root root        120 Jan 25 11:50 .
      drwxr-xr-x 18 root root       3360 Feb 11 03:03 ..
      drwxr-xr-x  2 root root        100 Jan 25 11:50 by-path
      crw-rw----  1 root video  226,   0 Jan 25 11:50 card0
      crw-rw----  1 root video  226,   1 Jan 25 16:39 card1
      crw-rw----  1 root render 226, 128 Jan 25 11:50 renderD128
      

      Documentation indicates renderDXXX typically refers to Intel GPU’s

      Make sure at least one renderD* device exists in /dev/dri. Otherwise upgrade your kernel or enable the iGPU in the BIOS.

      1. Edit your docker-compose.yaml and add this In your Jellyfin block
      devices:
       - /dev/dri/renderD128:/dev/dri/renderD128
      
      1. Start your container and enter it to verify the device is recognized.

      sudo docker compose up -d; sudo docker exec -it jellyfin bash

      Once inside ls /dev/dri to confirm the GPU is recognized inside the container, once you confirm it then you can exit the container.

      user@debian:~/compose$ sudo docker exec -it jellyfin bash
      I have no name!@jellyfin:/$ ls /dev/dri
      renderD128
      I have no name!@jellyfin:/$ exit
      exit
      user@debian:~/compose$
      
      1. On the Jellyfin dashboard go to the hardware acceleration page and follow the notes left by Jellyfin devs.

    • DonutsRMeh@lemmy.world
      link
      fedilink
      English
      arrow-up
      3
      ·
      20 hours ago

      Man, I have an intel iGPU myself, in a little Dell optiplex 7090 and it was a breeze on my Debian sever. Installed through apt and it’s running as a systemd service. No issues so far. Only one issue I had was when I played a 70GB 4k HDR movie that’s loaded with audio and subs and picture enhancements on my OLED TV. The server’s little fan was screaming and the movie kept pausing every 20 seconds. Other than that I have a ton of other movies and shows and I have no issues.

    • Ebby@lemmy.ssba.com
      link
      fedilink
      English
      arrow-up
      1
      ·
      21 hours ago

      I use Intel too and had a heck of a time getting things working with portainer. Turns out portainer only worked with Nvidia (at least the version I used when I set it up). If I spun up the container via terminal, it worked.

      I think a newer version may have added compatibility because I don’t remember jumping through hoops with Immich.