[FLINK-36557] Stale Autoscaler Context in Kubernetes Operator

• Type: Improvement
• Status: Open
• Priority: Minor
• Resolution: Unresolved
• Component/s: Autoscaler, Kubernetes Operator
• Assignee: Unassigned
• Reporter: Sai Sharath Dandi
• Created: 16/Oct/24 21:51
• Updated: 21/Oct/24 17:34

Description

The KubernetesJobAutoScalerContext is cached in the FlinkResourceContext and reused. If the JobAutoscalerContext is initialized before the job reaches Running state, it can cause the autoscaler to not trigger - link

 

We need to either refresh the AutoScalerContext similar to the standalone implementation or the autoscaler module itself needs to refresh the job status