Running deepstream 6.2 DGPU container using containerd only

Hi,

I am trying to run a deepstream container by importing the image into ctr (containerd) and running it using this :

ctr -n k3s.io run -t --runtime=nvidia myregistry/test:1.0 test_img

but I get an error saying : ctr: failed to start shim: failed to resolve runtime path: invalid runtime name nvidia, correct runtime name should be either format like io.containerd.runc.v1 or a full path to the binary: unknown

But I have added the correct lines in /etc/containerd/config.toml

Any reason why this could be happening ?

Attaching my config.toml here

config.toml.txt (7.2 KB)

This seems to be a k3s problem. I am not familiar with it. You can go to the k3s community to discuss the issue.
I tested the following command and the container can be started normally.

sudo k3s ctr -n k3s.io run  --rm -t --runc-binary=/usr/bin/nvidia-container-runtime  nvcr.io/nvidia/deepstream:7.1-triton-multiarch test_img

I did not modify any configuration files

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.