hi team, We are in the process of upgrading from F...
# troubleshooting
b
hi team, We are in the process of upgrading from Flink 1.13.6 to Flink 1.19.0. We have observed that Flink 1.19.0 doesn’t read configmaps with checkpoint data created by Flink 1.13.6. Please advise. Job using Flink 1.13.6 runs. Logs:
Copy code
[23-May-2024 11:35:08.492 UTC] INFO org.apache.flink.runtime.leaderelection.DefaultLeaderElectionService [] - Starting DefaultLeaderElectionService with KubernetesLeaderElectionDriver{configMapName='flink-job-eed05e489d8fe5ed5a7c7dc37fe9783c-jobmanager-leader'}.

[23-May-2024 11:35:08.553 UTC] INFO org.apache.flink.runtime.jobmaster.JobMaster                 [] - Initializing job flink-job_X (eed05e489d8fe5ed5a7c7dc37fe9783c).

[23-May-2024 11:38:09.641 UTC] INFO org.apache.flink.runtime.checkpoint.CheckpointCoordinator    [] - Completed checkpoint 1 for job eed05e489d8fe5ed5a7c7dc37fe9783c (57643 bytes in 734 ms).
At the point when we do the upgrade process:
Copy code
[2024-05-23 11:44:13.773564+0000] INFO  [23-May-2024 11:44:13.772 UTC] INFO org.apache.flink.runtime.leaderretrieval.DefaultLeaderRetrievalService [] - Starting DefaultLeaderRetrievalService with KubernetesLeaderRetrievalDriver{configMapName='flink-job-cluster-config-map'}.

[2024-05-23 11:44:13.774773+0000] INFO  [23-May-2024 11:44:13.772 UTC] INFO org.apache.flink.kubernetes.kubeclient.resources.KubernetesConfigMapSharedInformer [] - Starting to watch for zonda/flink-job-cluster-config-map, watching id:8ab9fa08-cfa6-4af9-b66a-2f5d991eeb43

[2024-05-23 11:44:13.782629+0000] INFO  [23-May-2024 11:44:13.781 UTC] INFO org.apache.flink.client.cli.CliFrontend                      [] - Waiting for response...

[2024-05-23 11:44:13.783614+0000] INFO  Waiting for response...

[2024-05-23 11:44:25.984593+0000] INFO  [23-May-2024 11:44:25.983 UTC] INFO org.apache.flink.client.cli.CliFrontend                      [] - Successfully retrieved list of jobs

[2024-05-23 11:44:25.985647+0000] INFO  No running jobs.

[2024-05-23 11:44:25.986618+0000] INFO  No scheduled jobs.

[2024-05-23 11:44:25.989245+0000] INFO  [23-May-2024 11:44:25.988 UTC] INFO org.apache.flink.runtime.leaderretrieval.DefaultLeaderRetrievalService [] - Stopping DefaultLeaderRetrievalService.
Please let me know if I can provide more details.
a
There are a lot of breaking changes between 1.13 and 1.19. Have considered jumping not more than 2 versions at once ? e.g. 1.13 => 1.15 => 1.17 => 1.19. I know it's more work, but it's way more safe