[
https://issues.apache.org/jira/browse/HADOOP-1488?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12504359
]
Devaraj Das commented on HADOOP-1488:
-------------------------------------
I think the patch for HADOOP-1462 addresses this concern.
> Remove all use of auto-progress threads in map/reduce
> -----------------------------------------------------
>
> Key: HADOOP-1488
> URL: https://issues.apache.org/jira/browse/HADOOP-1488
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Reporter: Owen O'Malley
>
> During the last rework of the shuffle, a lot of the stages of the shuffle had
> auto-progress threads added, leading to system lockups when the shuffle
> stalls. We need to add Progressables to the sort and fetching interfaces so
> that if any task gets stuck, it will eventually be killed by the framework.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
