hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Devaraj Das (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-3333) job failing because of reassigning same tasktracker to failing tasks
Date Fri, 02 May 2008 17:46:55 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-3333?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12593825#action_12593825
] 

Devaraj Das commented on HADOOP-3333:
-------------------------------------

bq. 1. We might have to rework the logic to work around task FAILURES; currently the JT only
schedules around nodes where the task FAILED. However a lost tasktracker leads to tasks being
marked KILLED.

IMO we should leave this logic unchanged. The re-execution on this lost TT, if it fails, will
make the JT schedule tasks around that.

bq. 2. We also have to track hostnames rather than 'trackernames', trackername includes the
host:port... (#2)

This makes sense (as long as we don't depend host:port esp. in the unit tests).

> job failing because of reassigning same tasktracker to failing tasks
> --------------------------------------------------------------------
>
>                 Key: HADOOP-3333
>                 URL: https://issues.apache.org/jira/browse/HADOOP-3333
>             Project: Hadoop Core
>          Issue Type: Bug
>          Components: mapred
>    Affects Versions: 0.16.3
>            Reporter: Christian Kunz
>            Assignee: Arun C Murthy
>            Priority: Blocker
>
> We have a long running a job in a 2nd atttempt. Previous job was failing and current
jobs risks to fail as well, because  reduce tasks failing on marginal TaskTrackers are assigned
repeatedly to the same TaskTrackers (probably because it is the only available slot), eventually
running out of attempts.
> Reduce tasks should be assigned to the same TaskTrackers at most twice, or TaskTrackers
need to get some better smarts to find  failing hardware.
> BTW, mapred.reduce.max.attempts=12, which is high, but does not help in this case.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message