Subject: Why did JM fail on K8s (see original thread below)


We are investigating that.   But is the above theory plausible ( flink
gurus ) even though this, as in forcing restartPolicy: Never pretty much
nullifies HA on JM is it is a Job cluster ( at leats on k8s )
As for the reason we are investigating that.

One thing we looking as the QOS (
https://kubernetes.io/docs/tasks/configure-pod-container/quality-service-pod/ )
,  which was Burstable, a limit was not set on the job. The below for
example would have a QOS of guaranteed. So one reason could be this.

resources:
  requests:
    cpu: "8.0"
    memory: "16Gi"
  limits:
    cpu: "8.0"
    memory: "16Gi"

On Sat, Jun 29, 2019 at 9:35 AM Timothy Victor <[EMAIL PROTECTED]> wrote: