lyft / flinkk8soperator

Kubernetes operator that provides control plane for managing Apache Flink applications

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

FlinkApplication CRD Update not being applied

franciscolopezsancho opened this issue · comments

Hi there,

I have an application that after updating the CRD get's into state:

NAME                                      PHASE           CLUSTER HEALTH   JOB HEALTH   JOB RESTARTS   AGE
validation-app-cisco-validation-event     SubmittingJob   Red                                          23h
validation-app-cisco-validation-process   SubmittingJob   Red                                          23h

and never terminates any Pod and restart with the new Spec. In here states that the phase must be Running or Deployed Failed in order to allow the Flink Operator to redeploy. Is that the case?
If so I'd like to add that our application get's into that state only after the CRD get's updated. Seems to me it tries to checkpointing before terminate any Pods. Does this make sense to you?

Thank you in advance

commented

I also find the same issue as well. This happened only on the first app created. Because there's no older version to roll back, so it seems like it stuck on this state forever. It would be nice that it still goes into the DeployFailed after some timeout.