Ravi Nishant
08/14/2023, 6:17 AMRECONCILING to FAILED . However, this status change in not reflected in FlinkDeployment resource. Please see screenshot.
o.a.f.k.o.l.AuditUtils [m [32m[INFO ][samplejob/samplejob-jm-ue1a] >>> Event | Info | JOBSTATUSCHANGED | Job status changed from RECONCILING to FAILEDRavi Nishant
08/14/2023, 6:20 AMGyula Fóra
08/14/2023, 7:09 AMRavi Nishant
08/14/2023, 7:41 AMNullPointerException in operator logs. Only other error logged by the operator is
[m[33m2023-08-14 06:02:40,028[m [36mo.a.f.k.o.o.JobStatusObserver [m [1;31m[ERROR][hubblederivedeventsjob/hubblederivedeventsjob-jm-ue1a] Job 5c8133740a6e7c5c08492d05d56963b4 failed with error: org.apache.flink.client.program.ProgramInvocationException: An error occurred while invoking the program's main method: 'com.fasterxml.jackson.core.io.ContentReference com.fasterxml.jackson.dataformat.yaml.YAMLFactory._createContentReference(java.lang.Object)
Which is the error from the actual job.Ravi Nishant
08/14/2023, 8:12 AMNullPointer message with ReconciliationExceptionRavi Nishant
08/14/2023, 8:27 AMFailed ? The UI also shows as the job is in "Failed" state (as part of completed job list).Ravi Nishant
08/14/2023, 8:31 AMFlinkDeployment resource reflects correct state and not stuck in RECONCILING state, we can instracut our CI/CD to delete the resource and recreate with the new spec. As of now, this does not seem to be feasible.