Github user tzulitai commented on the issue:
https://github.com/apache/flink/pull/3982
Verified that Flink YARN sessions + HA works nicely now.
1. Started non-detached YARN session
2. Submit job
3. Kill JobManager
4. YARN session correctly messages disassociation with first JobManager
5. Failure of cluster status retrieval is correctly ignored, and session
does not abort.
6. New JobManager correctly registered, TMs correctly re-connect.
7. Job remains intact.
Logs are sane and nice (minus my comment on the "retrying" working).
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
