[
https://issues.apache.org/jira/browse/HADOOP-4787?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12654763#action_12654763
]
Devaraj Das commented on HADOOP-4787:
-------------------------------------
I am beginning to think that mapred.max.tracker.failures should have been a
cluster-level configuration, and, with HADOOP-4305 committed, it seems more
necessary. This is because with HADOOP-4305, depending on the value of the
config item and the number of task failures at the various trackers, a
tasktracker can be blacklisted across jobs sooner than later. This may be
difficult to manage in a real-world setup..
> TestTrackerBlacklistAcrossJobs fails randomly
> ---------------------------------------------
>
> Key: HADOOP-4787
> URL: https://issues.apache.org/jira/browse/HADOOP-4787
> Project: Hadoop Core
> Issue Type: Bug
> Components: mapred, test
> Reporter: Hemanth Yamijala
> Assignee: Amareshwari Sriramadasu
> Priority: Blocker
> Fix For: 0.20.0
>
> Attachments: patch-4787-1.txt, patch-4787.txt
>
>
> While fixing HADOOP-4786, I found that TestTrackerBlacklistAcrossJobs is
> failing randomly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
