k8s-device-plugin icon indicating copy to clipboard operation
k8s-device-plugin copied to clipboard

Xid 68 marked as use app error, different than official NVIDIA Xid doc

Open gyuho opened this issue 5 months ago • 0 comments

https://github.com/NVIDIA/k8s-device-plugin/blob/f566a821e75e93d190c38178b50f8049cce1c006/internal/rm/health.go#L65-L71

but says different in https://docs.nvidia.com/deploy/pdf/XID_Errors.pdf

Image

Can we document whether it is safe to mark xid 68 as user app error?

gyuho avatar Sep 06 '24 11:09 gyuho