hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Amar Kamat (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-3333) job failing because of reassigning same tasktracker to failing tasks
Date Wed, 07 May 2008 04:35:55 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-3333?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12594774#action_12594774
] 

Amar Kamat commented on HADOOP-3333:
------------------------------------

Why dont we do something like
!) If {{(num-unique-nodes-amongst-the-trackers / total-trackers-registered-with-jt) > K}}
then blacklist the node for that TIP i.e do as discussed above.
2) Else avoid blacklisting the host for that TIP (similar to the current blacklisting of trackers).
Where K = 0.25
This will overcome the corner case where the cluster is running on smaller number of nodes
and the TIP has failed on atleast one tracker on each node. Also the test can be kept as it
is.

> 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: Critical
>             Fix For: 0.18.0
>
>         Attachments: HADOOP-3333_0_20080503.patch, HADOOP-3333_1_20080505.patch, HADOOP-3333_2_20080506.patch
>
>
> 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