Supervisor Cluster update stalls due to concurrent administrative tasks VMware vSphere Kubernetes Service

Products

• VMware vSphere Kubernetes Service

Issue/Introduction

This article addresses Supervisor Cluster updates that stall in a "Configuring" or "Updating" state. It specifically outlines scenarios where concurrent administrative interventions trigger synchronization interruptions, leading to stalled upgrade tasks.

• Supervisor Cluster update is stuck in a "Configuring" or "Updating" state for an extended period.
• Recent Tasks in the vSphere Client show upgrade-related tasks that are not progressing or completing.
• Manual administrative tasks or interventions were performed concurrent to the upgrade process.

Environment

• VMware vSphere Kubernetes Service (vSphere with Tanzu), all supported versions

Cause

Manual interventions or administrative tasks performed during an automated upgrade process can interrupt synchronization between the Supervisor Cluster components, causing the upgrade task to stall.

Resolution

1. Complete all pending manual administrative tasks or interventions initiated on the cluster.
2. Verify the stability of the cluster components.
3. If upgrade tasks remain in a stale state after manual tasks are completed, clear the stale tasks by restarting the vsphere-ui and vpxd services:
   • Restart vSphere UI service: service-control --stop vsphere-ui followed by service-control --start vsphere-ui
   • Restart vpxd service: service-control --stop vmware-vpxd followed by service-control --start vmware-vpxd
4. Once services are restarted, retry the Supervisor Cluster upgrade process through the vSphere Client.

Additional Information

Clear stale tasks from the vCenter "Recent Tasks" panel