[
https://issues.apache.org/jira/browse/HADOOP-1798?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12523550
]
Arun C Murthy commented on HADOOP-1798:
---------------------------------------
+1
Long term, I'd like to fix updateTaskTrackerStatus to handle this better...
thoughts?
> The counts of currently running maps and reduces isn't maintained correctly
> when task trackers fail
> ---------------------------------------------------------------------------------------------------
>
> Key: HADOOP-1798
> URL: https://issues.apache.org/jira/browse/HADOOP-1798
> Project: Hadoop
> Issue Type: Bug
> Components: mapred
> Affects Versions: 0.13.0
> Reporter: Owen O'Malley
> Assignee: Owen O'Malley
> Fix For: 0.14.1
>
> Attachments: 1798.patch
>
>
> The code in JobTracker.processHeartbeat removes the task tracker status from
> trackerStatus without updating the counts. Eventually the cluster will stop
> launching tasks.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
