[ 
https://issues.apache.org/jira/browse/OOZIE-1319?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13966113#comment-13966113
 ] 

Robert Kanter commented on OOZIE-1319:
--------------------------------------

I tried out the patch by modifying the aggregator example to be LAST_ONLY and 
removing the input data from HDFS.  Once both actions were WAITING, I uploaded 
the data back into HDFS.  I was expecting the first (\@1) action to go WAITING 
--> READY --> SKIPPED and the second (\@2) action to go WAITING --> READY --> 
RUNNING --> SUCCEEDED.  However, they both went WAITING --> READY --> RUNNING 
--> SUCCEEDED; though they did it in LIFO order (that is, \@2 then \@1).  So, 
it doesn't look like this is working properly.  Looking at the code, I think it 
looks like it should work.  [~bowenzhangusa], any ideas?

Also, so I don't forget, the patch should also update the documentation to 
mention the SKIPPED status here: 
https://oozie.apache.org/docs/4.0.1/CoordinatorFunctionalSpec.html#a6.1.3.2._Coordinator_Action_Status
and also where it talks about LAST_ONLY here: 
https://oozie.apache.org/docs/4.0.1/CoordinatorFunctionalSpec.html#a6.3._Synchronous_Coordinator_Application_Definition

> "LAST_ONLY" in execution control for coordinator job still runs all the 
> actions
> -------------------------------------------------------------------------------
>
>                 Key: OOZIE-1319
>                 URL: https://issues.apache.org/jira/browse/OOZIE-1319
>             Project: Oozie
>          Issue Type: Bug
>            Reporter: Bowen Zhang
>            Assignee: Bowen Zhang
>         Attachments: oozie-1319.patch
>
>
> In execute() of CoordJobGetReadyActionsJPAExecutor.java, once we retrieve the 
> top item from a "LIFO" query result, we do not discard or delete the 
> remaining items from the result list. As a result, the next time execute() is 
> invoked, we will be retrieving the next item in line. Consequently, LAST_ONLY 
> strategy will also execute all ready actions for a given coordinator job, 
> making it no different than LIFO.



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Reply via email to