Supervisor configuration status reports an "Error" on the vCenter UI, "CoreDNS configuration failed on Master node with identifier ############ . Details: failed to retrieve current version from etcd"

Products

• VMware vSphere Kubernetes Service

Issue/Introduction

• Supervisor configuration status reports an "Error" with the  following error message on the vCenter UI:
  CoreDNS configuration failed on Master node with identifier ############ . Details: failed to retrieve current version from etcd
  
  
• The Control Plane node is in NotReady state. 
  
  
• Containers such as "etcd, kube-apiserver, kube-scheduler" are in exited state. 
  
  
• Verifying logs for exited etcd container:
  crictl logs <Container_ID_Exited_etcd>
etcdmain: tls: private key does not match public key

Environment

vSphere Kubernetes Service

Cause

The primary failure is due to a TLS certificate mismatch on the Control Plane node, preventing the etcd container from initializing.

Resolution

• Renew the certs on the impacted node using the following command:
  kubeadm certs renew all 
  
  
• To ensure that the renewed certificates are persistent, run the following command:
  /usr/lib/vmware-wcp/hypercrypt.py --reencrypt