[FLINK-35108] Deployment recovery is triggered on terminal jobs after jm shutdown ttl

• Type: Bug
• Status: Closed
• Priority: Critical
• Resolution: Fixed
• Affects Version/s: kubernetes-operator-1.7.0, kubernetes-operator-1.8.0
• Fix Version/s: kubernetes-operator-1.9.0
• Component/s: Kubernetes Operator
• Labels: pull-request-available
• Assignee: Gyula Fora
• Reporter: Gyula Fora
• Created: 15/Apr/24 08:37
• Updated: 15/Apr/24 12:21
• Resolved: 15/Apr/24 12:21

Description

The deployment recovery mechanism is incorrectly triggered for terminal jobs once the JM deployment is deleted after the TTL period.

This causes jobs to be resubmitted. This affects only batch jobs.

The workaround is to set
kubernetes.operator.jm-deployment-recovery.enabled: false

 for batch jobs.

Issue Links

• links to: GitHub Pull Request #815