Pramod Ramarao comments

Results 5 comments of


                                            Pramod Ramarao

RFE: Export GPU serial number

Hi @mareklibra - why aren't UUIDs sufficient? They are unique for every GPU that is shipped.

RFE: Export GPU serial number

> get a list of all GPUs installed in a cluster and their additional info. Most of the components in the gpu-operator are daemonsets - so they would report devices...

failed to pull and unpack image "nvcr.io/nvidia/driver:470.82.01-sles15.3"

/cc @shivamerla @cdesiniotis as we've been having an email thread offline with Suse on this topic.

Error: failed to start container "nvidia-device-plugin-ctr": Error response from daemon: OCI runtime create failed: container_linux.go:349: starting container process caused "process_linux.go:449: container init caused \"process_linux.go:432: running prestart hook 0 caused \\\"error running hook: signal: segmentation fault (core dumped), stdout: , stderr: \\\"\"": unknown

hi @Hsuey - what is the Windows NVIDIA driver version that you're in the system? Did you download at least 465.12 from https://developer.nvidia.com/cuda/wsl and then follow the steps as described...

device plugin start with --device-list-strategy=volume-mounts, pod create error : nnvidia-container-cli: device error: unknown device id: /var/run/nvidia-container-devices\\\\n\\\"\"": unknown

FWIW, I made a write-up of reproducing this issue and verifying the fix that Kevin implemented here: `https://gist.github.com/dualvtable/43695ee812c3625b68a0c765679041dd` The fix was in-turn verified by GCP. I think we can close...