QNAP TVS-682 Docker Container with NVidia P2000

Server Version#:
Player Version#:
I have (as the title suggests), a qnap tvs-682 with QTS 4.4.1 and very recently, an nVidia P2000. I have been using Plex in a docker container for more than a year now and much prefer it. However, after getting a P2000, I cannot, for the life of me, get HW transcoding on the P2000. I use a docker-compose.yaml to get an instance of Plex running and was hoping someone could share the compose file or step me through how to get it working. Note that I have had HW transcoding working on it in the past using the Intel GPU inside the CPU and I suspect it is still accessing it as the GPU load show 0% and CPU 80-50% when watching a 10 bit H265 movie on Chrome Browser.
Here is my current attempt at the compose file:

plex:
  net: host
  image: "plexinc/pms-docker:plexpass"
  restart: always
  privileged: true
  container_name: "plex-1"
  labels:
    - "com.qnap.qcs.gpu=true"
    - "com.qnap.qcs.network.mode=host"
  volumes:
    - "./config:/config"
    - "./transcode:/transcode"
    - "/share:/share"
    - "/share/CACHEDEV1_DATA/.qpkg/NVIDIA_GPU_DRV/usr:/usr/local/nvidia"
  environment:
    - TZ=Australia/Melbourne
    - PLEX_CLAIM="claim-*******"
    - PLEX_UID=0
    - PLEX_GID=0
    - PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
    - LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/lib64
    - LIBRARY_PATH=/usr/local/cuda/lib64/stubs
    - CUDA_PKG_VERSION=8-0=8.0.61-1
    - CUDA_VERSION=8.0.61
    - CUDNN_VERSION=6.0.21
    - NVIDIA_DRIVER_CAPABILITIES=compute,utility
    - NVIDIA_REQUIRE_CUDA=cuda>=8.0
    - NVIDIA_VISIBLE_DEVICES=all
  devices:
    - "/dev/dri:/dev/dri"
    - "/dev/nvidia0:/dev/nvidia0"
    - "/dev/nvidiactl:/dev/nvidiactl"
    - "/dev/nvidia-uvm:/dev/nvidia-uvm"```

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.