kube-system pods are 100's of days old and need to be restarted.
search cancel

kube-system pods are 100's of days old and need to be restarted.

book

Article ID: 440961

calendar_today

Updated On:

Products

VCF Operations/Automation (formerly VMware Aria Suite)

Issue/Introduction

The kube-system pods have been running for an extended period (e.g., 370 days) and must be restarted.

Environment

Aria Automation 8.18.1

Cause

Extended uptime of kube-system pods can lead to stale states. Refreshing the pods ensures the Kubernetes environment remains stable during appliance maintenance or reboots.

 

Resolution

  • Take snapshots of Aria Automation appliance(s).

  • Run the following command to delete the kube-system pods: kubectl -n kube-system delete pods --all

  • Monitor the pods until they have successfully restarted.

  • Shut down the prelude pods to prepare for the appliance reboot: /opt/scripts/deploy.sh --shutdown

  • Once the prelude pods are down, restart the cluster in vCenter.

  • After the appliances are back online, bring the prelude pods back up: /opt/scripts/deploy.sh

  • Verify the operational state of the environment by accessing the UI.

Powered by Wolken Software