[jira] [Created] (FLINK-9324) SingleLogicalSlot returns completed release future before slot is properly returned

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

[jira] [Created] (FLINK-9324) SingleLogicalSlot returns completed release future before slot is properly returned

Shang Yuanchun (Jira)
Till Rohrmann created FLINK-9324:
------------------------------------

             Summary: SingleLogicalSlot returns completed release future before slot is properly returned
                 Key: FLINK-9324
                 URL: https://issues.apache.org/jira/browse/FLINK-9324
             Project: Flink
          Issue Type: Bug
          Components: Distributed Coordination
    Affects Versions: 1.5.0, 1.6.0
            Reporter: Till Rohrmann
            Assignee: Till Rohrmann
             Fix For: 1.5.0


The {{SingleLogicalSlot#releaseSlot}} method returns a future which is completed once the slot has been returned to the {{SlotOwner}}. Unfortunately, we don't wait for the {{SlotOwner's}} response to complete the future but complete it directly after the call has been made. This causes that the {{ExecutionGraph}} can get restarted in case of a recovery before all of its slots have been returned to the {{SlotPool}}. As a consequence, the allocation of the new tasks might require more than the max parallelism because of collisions with old tasks (in case of slot sharing).



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)