[jira] [Commented] (APEXCORE-595) Master incorrectly updates committedWindowId when all partitions are terminated.

["Thomas Weise (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/APEXCORE-595?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15793243#comment-15793243 ] 

Thomas Weise commented on APEXCORE-595:
---------------------------------------

The logic to update the checkpoints and committed window is in updateRecoveryCheckpoints(). It only traverses operators that exist in the physical plan and then also skips when there are deploy requests pending. 

We probably never had to consider a case where a logical operator without physical operator (0 partitions) is in the plan and that as a root operator. Is that the scenario you are looking at?


> Master incorrectly updates committedWindowId when all partitions are terminated.
> --------------------------------------------------------------------------------
>
>                 Key: APEXCORE-595
>                 URL: https://issues.apache.org/jira/browse/APEXCORE-595
>             Project: Apache Apex Core
>          Issue Type: Bug
>            Reporter: Tushar Gosavi
>            Assignee: Tushar Gosavi
>
> When all partitions are terminated application master incorrectly updates the committedWindowId to Long.MAX_VALUE and send purge requet to StramChild causing exception in StramChild.
> The problematic code is in StreamingContainerManager.updateCheckpoints.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)