Applicable versions

All Longhorn versions.

Symptoms

After node reboot, a recurring job pod remains stuck in the Pending state, preventing scheduled tasks from running.

Potential Cause

This issue may occur when a node is not ready at the time of a CronJob execution. The Kubernetes scheduler may attempt to place the job on a node that is still initializing after a reboot, leaving the pod in a Pending state.

Workaround

Manually delete the Pending pod. The CronJob associated with the recurring job will create a new pod at the next scheduled execution time.

Related Information

• Longhorn issue: #7956