Help with transcoding

[troothdotcom]

I recently upgraded to this  https://www.amazon.com/Beelink-Intel-N10...S94J5?th=1 and I set up Jellyfin through docker compose but now I'm not sure how to proceed to be able to use hardware acceleration. I tried chat-gpt but it just confused me more. Any and all help is greatly appreciated. Thanks!

[TheDreadPirate]

What OS are you running on? We have documentation for docker compose setup of Intel GPUs.

https://jellyfin.org/docs/general/admini...ial-docker

Pretty simple. Just add this to your compose.

Code:

devices:
      - /dev/dri/renderD128:/dev/dri/renderD128
    group_add:
      - "993" # Change this to match your "render" host group id and remove this comment

The group_add requires that you run a couple commands (documented in the link) to find the proper ID number on your system for the render group. Here is my full docker compose as an example.

Code:

services:
  jellyfin-testing:
    image: jellyfin/jellyfin:10.9.11
    container_name: jellyfin-testing
    user: 110:110
    group_add:
      - "993" # Intel render group
    networks:
      [censored bridge name]:
        ipv4_address: 172.16.100.2
    expose:
      - 8096
    volumes:
      - ./jellyfin-stable-data/config:/config
      - ./jellyfin-stable-data/cache:/cache
      - /media/library:/media/library:ro
      - /media/storage2:/media/storage2
      - /media/ramdisk:/media/ramdisk
    devices:
      - /dev/dri/renderD128:/dev/dri/renderD128
    environment:
      - TZ=America/New_York
    restart: 'unless-stopped'

[troothdotcom]

What OS are you running on?  We have documentation for docker compose setup of Intel GPUs.

https://jellyfin.org/docs/general/admini...ial-docker

Pretty simple.  Just add this to your compose.

Code:

    devices:
      - /dev/dri/renderD128:/dev/dri/renderD128
    group_add:
      - "993" # Change this to match your "render" host group id and remove this comment

The group_add requires that you run a couple commands (documented in the link) to find the proper ID number on your system for the render group.  Here is my full docker compose as an example.

Code:

services:
  jellyfin-testing:
    image: jellyfin/jellyfin:10.9.11
    container_name: jellyfin-testing
    user: 110:110
    group_add:
      - "993" # Intel render group
    networks:
      [censored bridge name]:
        ipv4_address: 172.16.100.2
    expose:
      - 8096
    volumes:
      - ./jellyfin-stable-data/config:/config
      - ./jellyfin-stable-data/cache:/cache
      - /media/library:/media/library:ro
      - /media/storage2:/media/storage2
      - /media/ramdisk:/media/ramdisk
    devices:
      - /dev/dri/renderD128:/dev/dri/renderD128
    environment:
      - TZ=America/New_York
    restart: 'unless-stopped'

Thank you this was very detailed. I am using Ubuntu server OS and docker compose and still fairly new to all this so I am not sure what i am doing wrong but the issue I am having right now is that I don't seem to have a /dev/dri directory.  Do I have to create my own or should it have been created when downloading the drivers?

[TheDreadPirate]

And I'm assuming you mean you don't have a /dev/dri directory on the host.

What version of Ubuntu? Intel's drivers are built into the Linux kernel, as long as your kernel is new enough.

Code:

cat /etc/os-release
uname -a

Share the output of those two commands.

[troothdotcom]

I got it! Thanks
i had to run sudo apt install --install-recommends linux-generic-hwe-22.04

What OS are you running on?  We have documentation for docker compose setup of Intel GPUs.

https://jellyfin.org/docs/general/admini...ial-docker

Pretty simple.  Just add this to your compose.

Code:

    devices:
      - /dev/dri/renderD128:/dev/dri/renderD128
    group_add:
      - "993" # Change this to match your "render" host group id and remove this comment

The group_add requires that you run a couple commands (documented in the link) to find the proper ID number on your system for the render group.  Here is my full docker compose as an example.

Code:

services:
  jellyfin-testing:
    image: jellyfin/jellyfin:10.9.11
    container_name: jellyfin-testing
    user: 110:110
    group_add:
      - "993" # Intel render group
    networks:
      [censored bridge name]:
        ipv4_address: 172.16.100.2
    expose:
      - 8096
    volumes:
      - ./jellyfin-stable-data/config:/config
      - ./jellyfin-stable-data/cache:/cache
      - /media/library:/media/library:ro
      - /media/storage2:/media/storage2
      - /media/ramdisk:/media/ramdisk
    devices:
      - /dev/dri/renderD128:/dev/dri/renderD128
    environment:
      - TZ=America/New_York
    restart: 'unless-stopped'

[TheDreadPirate]

Ah. So you were on kernel 5.15 on Ubuntu 22.04. That will do it.

[troothdotcom]

Thank you so much! Finally got Jellyfin working and transcoding how I wanted. You're the best. 
   

What OS are you running on?  We have documentation for docker compose setup of Intel GPUs.

https://jellyfin.org/docs/general/admini...ial-docker

Pretty simple.  Just add this to your compose.

Code:

    devices:
      - /dev/dri/renderD128:/dev/dri/renderD128
    group_add:
      - "993" # Change this to match your "render" host group id and remove this comment

The group_add requires that you run a couple commands (documented in the link) to find the proper ID number on your system for the render group.  Here is my full docker compose as an example.

Code:

services:
  jellyfin-testing:
    image: jellyfin/jellyfin:10.9.11
    container_name: jellyfin-testing
    user: 110:110
    group_add:
      - "993" # Intel render group
    networks:
      [censored bridge name]:
        ipv4_address: 172.16.100.2
    expose:
      - 8096
    volumes:
      - ./jellyfin-stable-data/config:/config
      - ./jellyfin-stable-data/cache:/cache
      - /media/library:/media/library:ro
      - /media/storage2:/media/storage2
      - /media/ramdisk:/media/ramdisk
    devices:
      - /dev/dri/renderD128:/dev/dri/renderD128
    environment:
      - TZ=America/New_York
    restart: 'unless-stopped'