State of cluster when provisioning shows `Updating`

[sowmyav27]

Rancher Server Setup

• Rancher version: on 2.6-head commit id: 7db4fe4
• Installation option (Docker install/Helm Chart): Helm
  • If Helm Chart, Kubernetes Cluster and version (RKE1, RKE2, k3s, EKS, etc): RKE1

Information about the Cluster

• Kubernetes version: v1.23.6 - local cluster
• Kubernetes version: v1.22.9-rc4+rke2r1 - Downstream cluster - RKE2 Do/Linode node driver

User Information

• What is the role of the user logged in? (Admin/Cluster Owner/Cluster Member/Project Owner/Project Member/Custom) Restricted Admin (Any)

Describe the bug State of cluster when provisioning shows Updating

To Reproduce

• Deploy a DO or a Linode RKE2 node driver cluster.
• Initially the state of the cluster is Provisioning

• Then the state of the cluster changes to Updating

Expected Result There is no update that has been triggered on the cluster, so ideally Provisioning status should be available on the cluster.

[snasovich]

This is currently considered working as expected as per https://github.com/rancher/rancher/issues/36993. However, an argument can be made that from a user’s point of view, it may be unexpected that the cluster goes to Updating state in this scenario as it was not Active yet and no changes were made to it. This will need to be discussed from requirements perspective.

[jakefhyde]

This should be no longer blocked as it was fixed as part of https://github.com/rancher/rancher/pull/40083 and https://github.com/rancher/rancher/pull/40201 for 2.7 and 2.6 respectively.

[snasovich]

@jakefhyde , this doesn't seem to be the case with all-role single-node RKE2 node-provisioned cluster as it still goes into "Updating" state at least for some time after "Provisioning" and before "Active". Removing from "To Test" for now.