Kubernetes 1.27 Deprecated CSIStorageCapacity: Storage.k8s.io/v1beta1: use v1.

[MarkCupitt]

What steps did you take and what happened:
Deployed zfs-localpv under Kubernetes 1.27

Pods would not schedule because PVs would not provision with error 0/1 nodes are available: 1 node(s) did not have enough free storage. preemption: 0/1 nodes are available: 1 Preemption is not helpful for scheduling..

Openebs-zfs-controller had log entry

E0424 04:24:46.342908 1 reflector.go:138] k8s.io/client-go/informers/factory.go:134: Failed to watch *v1beta1.CSIStorageCapacity: failed to list *v1beta1.CSIStorageCapacity: the server could not find the requested resource

1.27 Deprecations

CSIStorageCapacity: Storage.k8s.io/v1beta1: use v1.

https://sysdig.com/blog/kubernetes-1-27-whats-new/

[MarkCupitt]

is this connected? #299

[MarkCupitt]

OpenEBS Version 2.1.0

[vharsh]

/assign @vharsh

[vharsh]

This issue is coming because an older version of CSI provisioner is being used(3.0.0). The fix was pushed in https://github.com/kubernetes-csi/external-provisioner/pull/710. At a first glance it doesn't look to be coupled with a Kubernetes version which only supports v1-CSIStorageCapacity and it manages it internally and should not impact it's running & functionality on older Kubernetes clusters. However this needs to be validated once in the CI.

The fix for this would require bumping up the zfsController.provisioner.tag from v3.0.0 to v3.2.0. I'd take a stock of what other changes are being introduced and make relevant changes to the CI and then raise a fix for this.

[vharsh]

got fixed in #461