[ 
https://issues.apache.org/jira/browse/NIFI-6285?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16839723#comment-16839723
 ] 

ASF subversion and git services commented on NIFI-6285:
-------------------------------------------------------

Commit f08c2ee43f39584b7d4e76d61d7d4aaa889068fd in nifi's branch 
refs/heads/master from Mark Payne
[ https://gitbox.apache.org/repos/asf?p=nifi.git;h=f08c2ee ]

NIFI-6285: Addressed issue that resulted in swapped data not being swapped back 
in if load balancing strategy changed while data was swapped out; added 
integration tests for swapping. In testing, also encountered an issue with data 
being swapped out while swap files were being recovered causing the queue size 
to be wrong and causing errors about not being able to swap data in, because it 
attempted to swap the data in twice.

This closes #3473.

Signed-off-by: Bryan Bende <[email protected]>


> Data fails to swap in if Load Balancing Strategy is changed while the data is 
> swapped out
> -----------------------------------------------------------------------------------------
>
>                 Key: NIFI-6285
>                 URL: https://issues.apache.org/jira/browse/NIFI-6285
>             Project: Apache NiFi
>          Issue Type: Bug
>          Components: Core Framework
>            Reporter: Mark Payne
>            Assignee: Mark Payne
>            Priority: Blocker
>             Fix For: 1.10.0
>
>          Time Spent: 0.5h
>  Remaining Estimate: 0h
>
> If data is swapped out in a connection, and the connection's Load Balance 
> Strategy is then changed, the data fails to swap back in. I see the following 
> log messages:
> 2019-05-09 15:59:54,633 WARN [main] o.a.n.controller.FileSystemSwapManager 
> Encountered unknown Swap File 
> ./flowfile_repository/swap/1557431864223-9e237e10-016a-1000-06cb-e202396d758c-02cf72f0-1c0b-4771-8ef7-a708ed1d9de5.rebalance.swap;
>  will ignore this Swap File. This file should be cleaned up manually
> 2019-05-09 15:59:54,633 WARN [main] o.a.n.controller.FileSystemSwapManager 
> Encountered unknown Swap File 
> ./flowfile_repository/swap/1557431862791-9e237e10-016a-1000-06cb-e202396d758c-1c0c0a53-97d3-4703-b8de-6c7d6295e23a.rebalance.swap;
>  will ignore this Swap File. This file should be cleaned up manually
> 2019-05-09 15:59:54,633 WARN [main] o.a.n.controller.FileSystemSwapManager 
> Encountered unknown Swap File 
> ./flowfile_repository/swap/1557431866657-9e237e10-016a-1000-06cb-e202396d758c-a6573e52-1758-475f-a0a9-eee211bb7b26.rebalance.swap;
>  will ignore this Swap File. This file should be cleaned up manually
>  
> To recreate, create a cluster and queue up around 50,000 FlowFiles on each 
> node. Change connection's Load Balance Strategy from "Do Not Load Balance" to 
> "Round Robin." This results in the data that is swapped out not getting 
> swapped back in.



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

Reply via email to