[Bug] [Zeta][IMAP] can't restore job when master node down

[liunaijie]

Search before asking

• [X] I had searched in the issues and found no similar issues.

What happened

The master node shutdown because memory exceed the limit. Another node starting the mastership claim process, and become the master node. When this node initCoordinatorService, it very slow, can't restart the cluster. Here are some new master log screenshots

Here is the old master node log screenshots

And i check the code, when a node became to master, it will call initCoordinatorService method.

in this method, it will

1. get IMAP
2. create JobHistoryService
3. re-submit running jobs

In the log, i can't see the log about begin restore job (%s) from master active switch and the 2 create JobHistoryService doesn't have any operations

So i think is get IMAP fail, slow.

SeaTunnel Version

2.3.4

SeaTunnel Config

na

Running Command

na

Error Exception

na

Zeta or Flink or Spark Version

No response

Java or Scala Version

No response

Screenshots

No response

Are you willing to submit PR?

• [ ] Yes I am willing to submit a PR!

Code of Conduct

• [X] I agree to follow this project's Code of Conduct

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity for 30 days. It will be closed in next 7 days if no further activity occurs.