Docker Nvidia T400 decoding

[mattb]

I'm in the process of adding nvidia support to my jellyfin docker container and looking for some confirmation.

Believe I have everything setup, however when I run nvidia-smi, I'm not seeing the ...jellyfin-ffmpeg/ffmpeg process.  And this is making me unsure it's working.

I also see the gpu and memory being used (screenshot)

From what I understand, I should see this in the ffmpeg transcode logs (which I do):

Stream mapping:
  Stream #0:0 -> #0:0 (hevc (native) -> h264 (h264_nvenc))
  Stream #0:1 -> #0:1 (aac (native) -> ac3 (native))

Also see this in the log, which also indicates hwaccel:

/usr/lib/jellyfin-ffmpeg/ffmpeg -analyzeduration 200M -probesize 1G -f mov,mp4,m4a,3gp,3g2,mj2 -init_hw_device cuda=cu:0 -filter_hw_device cu -hwaccel cuda -hwaccel_output_format cuda -noautorotate -hwaccel_flags +unsafe_output -threads 1 -i file:"/tv/Band of Brothers/S01/Band of Brothers S01E02.mp4" -noautoscale -map 0:0 -map 0:1 -map -0:s -codec:v:0 h264_nvenc -force_key_frames "expr:gte(t,n_forced*5)" -vf "setparams=color_primaries=bt709:color_trc=bt709:colorspace=bt709,scale_cuda=format=yuv420p" -preset p1 -b:v 8326013 -maxrate 8326013 -bufsize 16652026 -map_metadata -1 -map_chapters -1 -threads 0 -codec:a:0 ac3 -ac 6 -ab 224004 -y "/cache/transcodes/e5c677daec8c3d6759d5a8e613c8f59f.ts"

https://pastebin.com/whyQGcqC

[emailbin11]

Hey! Did you enable hardware acceleration in the jellyfin dasboard -> playback -> transcoding to NVENC?

and does your docker file follow this guide: https://jellyfin.org/docs/general/post-i...ion/nvidia

specifically this portion:
deploy:
resources:
reservations:
devices:
- driver: nvidia
count: all
capabilities: [gpu]

additionally I believe it has to run as root (just in case you were trying to run a rootless container)


Also did you run nvidia-smi at the console or did you run nvidia-smi in your container like so: docker exec -it jellyfin nvidia-smi

[mattb]

Nvidia NVENC is enabled for Hardware Acceleration.
nvidia-smi is run from docker (see screenshot above).
running as my userid, which is what I understood is the process.

Code:

version: '3.5'
services:
  jellyfin:
    image: jellyfin/jellyfin:latest
    container_name: jellyfin
    environment:
      - PUID=1026
      - GUID=100
      - TZ=Mountain
    network_mode: host
    volumes:
      - /volume2/dockerSSD/jellyfin/config:/config:rw
      - /volume2/dockerSSD/jellyfin/cache:/cache:rw
      - /volume1/videos/tvshows:/tv:rw
      - /volume1/videos/movies:/movies:rw
    ports:
      - 8096:8096
    restart: unless-stopped
    runtime: nvidia
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities:
                - gpu