StatefulSet Ready State Displays 0/0 While Pods Remain "Running"

Products

• VMware vSphere Kubernetes Service

Issue/Introduction

In a VKS environment discrepancy is seen between the Stateful Set controller and the actual Pod status.

• kubectl get statefulset -n <namespace> displays the READY column as 0/0.
  
  

  k get statefulset -n <namespace>

  NAME                      READY       AGE
<stateful set name>       0/0         ##d

• kubectl get pods confirms that Pods belonging to the StatefulSet are in a Running status.
• Automated cleanup of Pods does not occur following a scale-down event, leading to "orphaned" workloads.

Environment

vSphere Kuberenetes Service 

Cause

This condition is typically caused by a synchronization failure between the Kubernetes Control Plane and the workload state

Resolution

To resolve the inconsistency, the controller must be forced to re-evaluate the Pod objects using the steps below

• Manually delete the Pods that are currently running but not being recognized by the StatefulSet. 
  
  kubectl delete pod <pod-name>


• Force a full reconciliation by scaling the StatefulSet replicas to zero.
  
  kubectl scale statefulset <name> --replicas=0

 

Additional Information

Podが実行中であるにも関わらず、StatefulSetのReady State が 0/0 になる