[
https://issues.apache.org/jira/browse/HADOOP-3858?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12618018#action_12618018
]
Owen O'Malley commented on HADOOP-3858:
---------------------------------------
We should also increase the number fetched at once to 500, like the TaskTracker.
> JobClient doesn't wait for all task completion events before exiting
> --------------------------------------------------------------------
>
> Key: HADOOP-3858
> URL: https://issues.apache.org/jira/browse/HADOOP-3858
> Project: Hadoop Core
> Issue Type: Bug
> Components: mapred
> Reporter: Owen O'Malley
> Assignee: Owen O'Malley
>
> The JobClient should pull all of the task completion events from the
> JobTracker before exiting the loop and returning to the application.
> Otherwise the failing task may be lost in the backlog since only 10 task
> events are fetched per a second.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
