hadoop-mapreduce-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Hemanth Yamijala (JIRA)" <j...@apache.org>
Subject [jira] Updated: (MAPREDUCE-1030) Reduce tasks are getting starved in capacity scheduler
Date Sun, 11 Oct 2009 16:10:31 GMT

     [ https://issues.apache.org/jira/browse/MAPREDUCE-1030?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel

Hemanth Yamijala updated MAPREDUCE-1030:

    Attachment: MAPREDUCE-1030-7.patch

I made a few changes as I was reviewing the last patch by Rahul, and am attaching a new patch
with those. 

The changes are mainly in the area of making checkMultipleTaskAssignment to be more strict
in checking the tasks expected and returned. In particular, I am now failing the test if a
task is scheduled when it should not be. This necessitated that all tasks returned by the
scheduler are checked, even for the overloaded method checkAssignment, and correspondingly
tests in TestCapacityScheduler needed to be changed so that it uses multiple task assignment
checks where appropriate. I remember changing all the user limit tests and testClusterBlockingForLackOfMemory
to accomodate this.

Other than this, I verified other changes are fine. So, I am pushing this patch through Hudson.

> Reduce tasks are getting starved in capacity scheduler
> ------------------------------------------------------
>                 Key: MAPREDUCE-1030
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-1030
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: contrib/capacity-sched
>    Affects Versions: 0.21.0
>            Reporter: rahul k singh
>            Assignee: rahul k singh
>            Priority: Blocker
>             Fix For: 0.21.0
>         Attachments: MAPREDUCE-1030-1.patch, MAPREDUCE-1030-2.patch.txt, MAPREDUCE-1030-3.patch,
MAPREDUCE-1030-4.patch, MAPREDUCE-1030-5.patch, MAPREDUCE-1030-6.patch, MAPREDUCE-1030-7.patch
> reduce tasks are getting starved in capacity scheduler. 

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message