XComp commented on pull request #18869:
URL: https://github.com/apache/flink/pull/18869#issuecomment-1050766013


   I did another minor refactoring in order to be consistent with the changes 
of #18901 (see 
[comment](https://github.com/apache/flink/pull/18901/files#r814672839)): 
`addAndLock` is going to throw an `AlreadyExistException` in any case. I 
updated the commit message accordingly:
   ```
   About ZooKeeperStateHandleStore.addAndLock:
     There shouldn't be a case where there is a concurrent delete on an
     addAndLock call because of the following reasons:
     - called through DefaultJobGraphStore.putJobGraph: Happens when a new
       job is submitted, i.e. a new JobID is used.
     - called through 
CheckpointCoordinator.addCompletedCheckpointToStoreAndSubsumeOldest:
       Each leader creates a new checkpoint ID using the CheckpointIDCounter,
       i.e. another JM cleaning up wouldn't have the handle to the newly
       created checkpoint.
     - called through CheckpointCoordinator.restoreSavepoint:
       A Savepoint is never subject to deletion
     That's why we keep the AlreadyExistException as a safe-guard
   ```


-- 
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