After upgrading to v7.2 my plex docker does not start and gets the error below, all other dockers startup just fine and they all share the same storage drive. If I roll it back to v7.1.4 everything works fine. Attached is the diagnostics.

I have rolled back to v7.1.4 in the meantime but any help in resolving this would be appreciated.

Nov 4 12:45:33 NAS rc.docker: PlexMediaServer: 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 createContainer hook #4: exit status 1, stdout: , stderr: 2025/11/04 12:45:33 Error updating ldcache: error running pivot_root: pivot_root .: invalid argument

Nov 4 12:45:33 NAS rc.docker: exit status 1: unknown

Nov 4 12:45:33 NAS rc.docker: Error: failed to start containers: PlexMediaServer

nas-diagnostics-20251104-1248.zip

I think you're having the same issues I'm having. Is that one container using the Nvidia runtime? All of my Nvidia runtime containers fail to start with that same error. I made sure I have the latest Nvidia driver.

I removed the --runtime=nvidia from my Ollama's "Extra Parameters" and it seems to work just fine. It's even still using the GPU. I did leave the --gpus=all but, I don't know if that is needed.

yep I am using the Nvidia drivers for my Plex. I will give that a shot to see if it fixes the issue and still will do the hardware transcoding. thanks.

Btw, at least for Ollama specifically I did have to leave the --gpus=all otherwise it never saw the GPU. I think that is on a per container basis though because it seems like some others did not need it.

+1 experiencing this with containers using the Nvidia driver

@FlexibleToast thanks that fixed the problem and I successfully upgraded to v7.2 and everything is running fine ;)

Allegedly there is an issue with one of the latest nvidia drivers and unraid 7.2.0 that is causing issues with passed through GPUs to containers. I would check the Nvidia Driver plugin support page here on the forums.

Rolling back a driver or two seems to fix it.

Edited November 5, 2025Nov 5 by MowMdown

4 hours ago, FlexibleToast said:

I removed the --runtime=nvidia from my Ollama's "Extra Parameters" and it seems to work just fine. It's even still using the GPU. I did leave the --gpus=all but, I don't know if that is needed.

On 11/4/2025 at 5:38 PM, FlexibleToast said:

I removed the --runtime=nvidia from my Ollama's "Extra Parameters" and it seems to work just fine. It's even still using the GPU. I did leave the --gpus=all but, I don't know if that is needed.

Thank you. I removed --runtime=NVIDIA and my Emby container started working again.

I can second @MowMdown suggestion, rolling back the driver worked for me.

9 hours ago, dukesmellington said:

I can second @MowMdown suggestion, rolling back the driver worked for me.

The TL;DR Is if you're on an older nvidia GPU, pascal or older, you are limited to the 580 driver branch due to nvidia dropping support for older GPUs on 590+

The 590 driver branch only supports ada series GPUs and newer. (And their workstation counterparts of course)