kubeadm join is being run on a worker node that is already joined to a cluster

[rkahl]

Describe the Bug

kubeadm join is being run on a worker node that is already joined to a cluster.

Expected Behavior

The join phase should be skipped in that case.

Steps to Reproduce

Environment

• Puppet v8.10
• puppetlabs-kubernetes v8.1.0
• Kubernetes v1.32.3

Additional Context

Problem is that Kubernetes may forbid the worker nodes to list all nodes of the cluster; it may only list itself.
The unless command of the join should look like "kubectl get node ${node_name} | grep ${node_name}".

[kattal-forsk]

Hello,
I am experiencing the exact same issue. Do you know if a bugfix is planned?
Thank you

[asusk7m550]

A fix for this bug will be in #696