NVIDIA 572.16 (January 30, 2025) update seems to break open-webui:cuda Docker container #3524

Closed
opened 2025-11-11 15:33:17 -06:00 by GiteaMirror · 0 comments
Owner

Originally created by @chepurko on GitHub (Jan 31, 2025).

Windows 11 system with NVIDIA RTX 4070 Ti SUPER and WSL2

NVIDIA App nagged me to install the latest update yesterday (572.16) - afterwards my open-webui:cuda container refused to start after restarting Docker Desktop and also after a couple system reboots.

No logs from the container as it couldn't even boot. I did however get the following error from the Docker Desktop interface (from notifications):

(HTTP code 500) server error - 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 hook #0: error running hook: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy' nvidia-container-cli: initialization error: WSL environment detected but no adapters were found: unknown

and the shell did throw some long SIGSEGV error after running my docker run command but I did not have a chance to capture the errors from my shell. I am using this command to run Open WebUI:

docker run -d -p 3000:8080 --gpus all --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:cuda

After uninstalling the driver plus the other various programs installed by the NVIDIA driver package and many reboots, I installed the 566.36 (Dec. 10 2024) driver and the container boots and runs just as smoothly as ever.

I believe the 572.16 driver is the first one to support the Blackwell architecture. If anyone else has this issue hopefully you find this issue and share. Otherwise I dunno what it could be. I've been running my instance fine until that NVIDIA driver update.

Originally created by @chepurko on GitHub (Jan 31, 2025). Windows 11 system with NVIDIA RTX 4070 Ti SUPER and WSL2 NVIDIA App nagged me to install the latest update yesterday (572.16) - afterwards my open-webui:cuda container refused to start after restarting Docker Desktop and also after a couple system reboots. No logs from the container as it couldn't even boot. I did however get the following error from the Docker Desktop interface (from notifications): `(HTTP code 500) server error - 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 hook #0: error running hook: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy' nvidia-container-cli: initialization error: WSL environment detected but no adapters were found: unknown` and the shell did throw some long SIGSEGV error after running my `docker run` command but I did not have a chance to capture the errors from my shell. I am using this command to run Open WebUI: `docker run -d -p 3000:8080 --gpus all --add-host=host.docker.internal:host-gateway -v open-webui:/app/backend/data --name open-webui --restart always ghcr.io/open-webui/open-webui:cuda` After uninstalling the driver plus the other various programs installed by the NVIDIA driver package and many reboots, I installed the [566.36 (Dec. 10 2024) driver](https://www.nvidia.com/en-us/drivers/details/237785/) and the container boots and runs just as smoothly as ever. I believe the 572.16 driver is the first one to support the Blackwell architecture. If anyone else has this issue hopefully you find this issue and share. Otherwise I dunno what it could be. I've been running my instance fine until that NVIDIA driver update.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/open-webui#3524