Webodm does not start with GPU

I am trying to start webodm using GPU.

The command I have used is sudo ./webodm.sh --gpu start

Action taken:

  1. I have installed nvidia-drivers, cuda drivers and rebooted the system.
  2. I have removed old docker containers and started fresh installation.

I am using Tesla T4.

Any suggestions on how to troubleshoot?

what does this show , when you run it in terminal:

docker run --rm --gpus all nvidia/cuda:10.0-base nvidia-smi

or this:

docker info |grep Runtimes

Did you install nvidia-container-runtime ?
sudo apt-get install nvidia-container-runtime

Then , you can append this (if it`s missing):

{
    "runtimes": {
        "nvidia": {
            "path": "/usr/bin/nvidia-container-runtime",
            "runtimeArgs": []
        }
    }
}

here:
/etc/docker/daemon.json

1 Like

I ran all the commands you said. Here’s the output:

nvidia-container-runtime is missing
Install it.
follow the link in previous post

1 Like

Yeah I got it…thanks a ton for the help…
I followed the official installation guide: Installation Guide — NVIDIA Cloud Native Technologies documentation

1 Like

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.