Pods inaccessible when run on specific node

[gioargyr]

Summary

I have a Kubernetes cluster built using microk8s.

• All servers (6 in total) have Ubuntu 24.04
• The Kubernetes/microk8s version is v1.34.3

I recently added the 6th server, I'm calling it rp-server and from the outside everything seems to work fine: No failures, no errors, node has "Ready" status.
BUT, whenever a pod runs on this node, it's totally inaccessible. And I explain:

• When a Pod test is sheduled on rp-server, if I try to:
  • $ kubectl exec -it test -- <any_command>, kubectl is hanging and then I get an error.
  • Create a K8s service and expose the Pod. If I try to port-forward the service, kubectl hangs and I get an error
  • Create an Ingress for this Pod/service, I get again nothing.
    kubectl is pointing to a different node in this cluster.

However, if ssh in rp-server and start running microk8s kubectl commands on Pods that are scheduled on this server, kubectl works fine!

Can someone help me start from somewhere. This issue is difficult for me to even search about it.

What Should Happen Instead?

The server rp-server should work fine like all the other servers

Reproduction Steps

I do not think this error is reproducible. I ma trying to find someone or something to help me at least start debugging this issue.

Introspection Report

By running microk8s inspect, I see that the localnode.yaml is missing!!!
cp: cannot stat '/var/snap/microk8s/8585/var/kubernetes/backend/localnode.yaml': No such file or directory

Can you suggest a fix?

No, I cannot suggest a fix

Are you interested in contributing with a fix?

No, I am not able to contribute.