Docker NVIDIA Linux

This is a generic guide to installing NVIDIA support on a docker instance running on a linux machine.

Step 1. Drivers

Install the latest NVIDIA Drivers

Should be something along the lines of this

sudo apt install nvidia-driver-515

Once installed, reboot.

Step 2. NVIDIA Docker Runtime

Next you the NVIDIA runtime for docker

This is a long guide, go slowly and follow each step closely.

Step 3. Turn off cgroups

Edit the NVIDIA Container Runtime configuration file

sudo nano /etc/nvidia-container-runtime/config.toml

Uncomment or add the line to the [nvidia-container-cli] section

no-cgroups = false

Save the file

Goto to nvidia/cuda dockerhub and get the latest tag for the container. Note :latest has been deprecated and you will need to manually look.
For example, 11.8.0-devel-ubuntu22.04

Restart docker and test (with tag from previous step)

sudo systemctl restart docker
sudo docker run --rm --gpus all nvidia/cuda:11.8.0-devel-ubuntu22.04 nvidia-smi

Step 4. FileFlows

Next its time to get FileFlows working. This requires the "nvidia" runtime for docker from step 2.

sudo docker run -d \
--name FileFlows \
-p 19200:5000 \
-v /mnt/docker/fileflows/data:/app/Data \
-v /mnt/docker/fileflows/logging:/app/Logs \
-v /mnt/docker/fileflows/temp:/temp \
-v /mnt/media:/media \
--runtime nvidia \
--restart unless-stopped \
revenz/fileflows