CryoThrust commented on PR #12149:
URL: https://github.com/apache/seatunnel/pull/12149#issuecomment-6967300326

   Addressed the remaining points in the existing three checkpoint files:
   
   - The scheduler and cleanup Javadocs now describe RESET as the revival 
boundary for both pipeline restore and master failover. Non-reset cleanup 
leaves the pool stopped until `restoreCoordinator()` runs. The constructor 
description also makes lazy worker creation explicit.
   - Restored the explanation for the inner `catch (Throwable)`: any failed 
scheduled trigger must fail the coordinator, otherwise the checkpoint chain 
could stop silently. The outer rejection handler remains specific to scheduling 
against a stopped pool.
   - The recovered streaming coordinator's cancellation check now awaits both 
`isShutdown()` and `isTerminated()` before either member is shut down. This 
checks that all workers of that specific scheduler have exited, without relying 
on ambiguous thread names.
   
   On JDK 11, all 19 coordinator unit cases and the selected streaming 
master-failover IT pass without skips; scoped Spotless, Checkstyle and 
skip-tests packaging also pass. The title now uses `[Fix][Zeta]`. The earlier 
successful fork matrix applies to `036319a35`; current-head CI is separate.
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to