[0.41.1] after node crash csi-s3 fails umounting a not mounted directory

[pasztorl]

I0718 03:05:00.497445       1 utils.go:97] GRPC call: /csi.v1.Node/NodeUnpublishVolume
I0718 03:05:00.501810       1 nodeserver.go:152] s3: volume example-s3-csi/pvc-5f6fffe7-3407-4f0f-a156-cd45b22ca751 has been unmounted.
E0718 03:05:00.502595       1 utils.go:101] GRPC error: rpc error: code = Internal desc = Unmount failed: exit status 1
Unmounting arguments: /var/lib/kubelet/pods/4fb3f2dd-7c90-473e-b897-f39c14c3efaa/volumes/kubernetes.io~csi/pvc-b21e92f8-433c-44cb-91b6-87a4d2750f08/mount
Output: umount: can't unmount /var/lib/kubelet/pods/4fb3f2dd-7c90-473e-b897-f39c14c3efaa/volumes/kubernetes.io~csi/pvc-b21e92f8-433c-44cb-91b6-87a4d2750f08/mount: Invalid argument

i solved it by mounting tmpfs to the mountpoint, but checking needed, so if the directory not a mountpoint do not try to umount it.

[vherrlein]

PR152 would fix that issue.

Thanks for your trick by mounting with tmpfs.

[stillya]

any update?

[insoln]

+1, hitting this in prod — pod stuck in Terminating for 16 days after unexpected node reboot.