[
https://issues.apache.org/jira/browse/MAPREDUCE-372?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12785897#action_12785897
]
Sharad Agarwal commented on MAPREDUCE-372:
------------------------------------------
Looked at the ChainBlockingQueue part of the code. Some comments:
1. Can we avoid the casting in Chain#stopAllThreads? One way could be to
override interrupt() in MapRunner and ReduceRunner. Also interruptAllThreads
would be a better name IMO.
2. I think instead of interrupting the runners and then calling interrupt on
both readers and writers, it would be simpler we can directly interrupt all the
blocking queues.
> Change org.apache.hadoop.mapred.lib.ChainMapper/Reducer to use new api.
> -----------------------------------------------------------------------
>
> Key: MAPREDUCE-372
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-372
> Project: Hadoop Map/Reduce
> Issue Type: Sub-task
> Reporter: Amareshwari Sriramadasu
> Assignee: Amareshwari Sriramadasu
> Fix For: 0.21.0
>
> Attachments: mapred-372.patch, mapred-372.patch, mapred-372.patch,
> patch-372-1.txt, patch-372-2.txt, patch-372.txt
>
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
