falcon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "sandeep samudrala (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (FALCON-1807) Late Rerun is rerunning as soon as the data is available rather than waiting for the delay.
Date Thu, 04 Feb 2016 10:07:40 GMT

    [ https://issues.apache.org/jira/browse/FALCON-1807?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15132076#comment-15132076
] 

sandeep samudrala commented on FALCON-1807:
-------------------------------------------

Also further debugging has shown that the messages are continuously getting enqueued and dequeued
from the delayed/activemq, resulting huge amount of messages transmitted.
All this happening because of the the way getDelay in RerunEvent is being calculated. It considers
nominal time as reference each time to consider for the time taken in processing.
But once the instance goes far away(future) from the nominal time the delay calculation always
goes to -ve(enque instantly).


> Late Rerun is rerunning as soon as the data is available rather than waiting for the
delay.
> -------------------------------------------------------------------------------------------
>
>                 Key: FALCON-1807
>                 URL: https://issues.apache.org/jira/browse/FALCON-1807
>             Project: Falcon
>          Issue Type: Bug
>          Components: rerun
>    Affects Versions: 0.9
>            Reporter: Pragya Mittal
>            Assignee: sandeep samudrala
>            Priority: Blocker
>
> Ideally late rerun, runs the instance if and when the data becomes available in the late
rerun zone. This is not happening currently.



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

Mime
View raw message