[
https://issues.apache.org/jira/browse/OOZIE-102?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13101836#comment-13101836
]
Hadoop QA commented on OOZIE-102:
---------------------------------
mislam77 remarked:
We can identify the job as "catchup" by checking creationTime > nominalTime.
In that case, I prefer to use this :
Catchup Timeout = Creation Time + Concurrency * Timeout
Concurrency means how many max job could be in WAITING state or data checking
state.
Overall the reason is: if there are concurrency = 5 )say), there will be at
most 5 actions actively checking for their data. Now if they have chain/ripple
dependency, the last one has to wait until the other 4 are finished.
To accommodate this command scenario, I want to give a relax/loose timeout.
Comments?
> GH-67: input data check should have a timeout for catch-up mode too.
> ---------------------------------------------------------------------
>
> Key: OOZIE-102
> URL: https://issues.apache.org/jira/browse/OOZIE-102
> Project: Oozie
> Issue Type: Bug
> Reporter: Hadoop QA
>
> For normal,timeout when Nominal Time + timeout > current time
> For catchup,timeout when Created Time + timeout > current time
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
