Closed terrelsa13 closed 11 months ago
Nevermind, found an answer here that seems to work.
In the frigatelxc, I set no-cgroups=true
in the /etc/nvidia-container-runtime/config.toml
. Then ran nvidia-ctk cdi generate --output=/etc/cdi/nvidia.yaml
.
Describe the problem you are having
Anyone able to help get nvidia HWAccel setup in a LXC? I have followed the NVIDIA GPUs section, but cannot get it to work for me.
Yes I know running Docker in a Proxmox LXC is not recommended. But my significant other also recommends I do not setup any more machines running 24hrs. The power bill is getting ridiculous.
Both of these video cards are being used in LXCs with Emby and Jellyfin.
Proxmox Info:
LXC Info:
NVIDIA-Linux-x86_64-535.129.03.run installed on Proxmox.
NVIDIA-Linux-x86_64-535.129.03.run installed on frigatelxc.
NVIDIA-Linux-x86_64-535.129.03.run installed on Frigate Docker container.
nvidia-container-toolkit installed on frigatelxc.
nvidia-container-toolkit installed on Frigate Docker container.
Docker daemon.json on frigatelxc.
Video card devices as seen from Proxmox.
Video card devices as seen from frigatelxc.
Video card devices as seen from Frigate Docker.
Configuration of frigatelxc.
Version
0.12.1-367d724
Frigate config file
docker-compose file or Docker CLI command
Relevant log output
FFprobe output from your camera
Operating system
Proxmox
Install method
Docker Compose
Network connection
Wired
Camera make and model
Amcrest IP8M-T2599EW, Amcrest IP8M-2496EB-28MM
Any other information that may be helpful
With both of the sections below commented out, the Frigate Docker container starts perfectly fine in Portainer.
With either or both of the sections un-commented...
I get this error in Portainer.
Failure failed to deploy a stack: Container frigate Recreate Container frigate Recreated Container frigate Starting 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 hook #0: error running hook: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy' nvidia-container-cli: mount error: failed to add device rules: unable to find any existing device filters attached to the cgroup: bpf_prog_query(BPF_CGROUP_DEVICE) failed: operation not permitted: unknown
I have also tried using Docker CLI...
And get a similar error.
Error 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 hook #0: error running hook: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy' nvidia-container-cli: mount error: failed to add device rules: unable to find any existing device filters attached to the cgroup: bpf_prog_query(BPF_CGROUP_DEVICE) failed: operation not permitted: unknown.
Cannot get past starting the Frigate Docker container with nvidia gpu options enabled.