[
https://issues.apache.org/jira/browse/HADOOP-3080?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12581921#action_12581921
]
Hudson commented on HADOOP-3080:
--------------------------------
Integrated in Hadoop-trunk #441 (See
[http://hudson.zones.apache.org/hudson/job/Hadoop-trunk/441/])
> Remove flush calls from JobHistory
> ----------------------------------
>
> Key: HADOOP-3080
> URL: https://issues.apache.org/jira/browse/HADOOP-3080
> Project: Hadoop Core
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.17.0
> Reporter: Devaraj Das
> Assignee: Amareshwari Sriramadasu
> Priority: Blocker
> Fix For: 0.17.0
>
> Attachments: patch-3080.txt
>
>
> The flush calls from JobHistory should be removed. This has performance
> impact when lots of events are being logged to the history files since the JT
> lock (e.g. lostTracker where the tracker had ran a lot of tasks, and we call
> failedTask in a loop, etc.)
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
