Getting a Signal 9 Error with Docker on Arch Linux – What’s Going Wrong?

0
84
Asked By CuriousCat123 On

I'm new to Docker and currently running Arch Linux. I tried to test the installation using the command `docker run --rm --gpus=all nvidia/cuda:12.1.1-base-ubuntu22.04 nvidia-smi`, but it returned a signal 9 error. The exact error was:

`docker: Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running prestart hook #0: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy'`

Additionally, I saw this message: `nvidia-container-cli: ldcache error: process /sbin/ldconfig terminated with signal 9`. I've tried reinstalling the nvidia-dkms drivers and the nvidia-container-toolkit, but nothing seems to work. I'm using the Linux Zen Kernel version 6.16.0, and basic Docker containers work fine. Any thoughts on how to resolve this?

4 Answers

Answered By John On

Solution to the problem can be found here: https://github.com/NVIDIA/nvidia-container-toolkit/issues/1246#issuecomment-3194227600

sudo nvidia-ctk config --in-place --set nvidia-container-runtime.mode=cdi

Answered By John On

Hi ! Any chance you found a solution? I'm having the exact same problem on my server that uses the nvidia card to transcode for jellyfin. I'm verified that everything is installed ok but can't get passed the same errors. I've also tried multiple kernels, e.g., lts and zen, to no avail.

Answered By TechWhiz87 On

It looks like you’re running into some compatibility issues because Arch isn’t officially supported by Docker. Check out the Docker installation docs for supported platforms. Also, make sure the NVIDIA container runtime is correctly set up, as that seems to be crucial for GPU access. Before running your command, try checking the installation with `docker info` to see if the NVIDIA runtime shows up. Also, you might want to use the command `sudo docker run --rm --runtime=nvidia --gpus all ubuntu nvidia-smi` to test if it works that way. If you don’t specify the runtime, the container won’t be able to access your GPU, so that could be part of the problem!

Answered By LinuxLover99 On

Yeah, I agree with TechWhiz. Since Arch is a rolling release, some packages might not be stable with NVidia's container toolkit. First, verify that the NVIDIA driver and toolkit are properly installed. You might also want to consult NVIDIA's documentation for issues specific to your setup. If you're still stuck, consider switching to a more frequently supported distro for Docker.

Related Questions

LEAVE A REPLY

Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.