hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Junping Du (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-3746) Invalid counter tag in HDFS balancer which lead to infinite loop
Date Wed, 01 Aug 2012 03:56:35 GMT

    [ https://issues.apache.org/jira/browse/HDFS-3746?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13426310#comment-13426310
] 

Junping Du commented on HDFS-3746:
----------------------------------

Meng, that's good finding. I think we should have a HashMap to keep connector to balancer
instance then have singleton for each connect. Please provide patch and upload. Before that,
I will reopen this issue and will assign to you later. 
                
> Invalid counter tag in HDFS balancer which lead to infinite loop
> ----------------------------------------------------------------
>
>                 Key: HDFS-3746
>                 URL: https://issues.apache.org/jira/browse/HDFS-3746
>             Project: Hadoop HDFS
>          Issue Type: Bug
>          Components: balancer
>    Affects Versions: 2.1.0-alpha, 3.0.0
>            Reporter: meng gong
>            Assignee: Junping Du
>              Labels: patch
>             Fix For: 2.1.0-alpha, 3.0.0
>
>
> Everytime banlancer try to move a block cross the rack. For every NameNodeConnector it
will instance a new banlancer. The tag notChangedIterations is reseted to 0. Then it won't
reach 5 and exit the thread for not moved in 5 consecutive iterations. This lead to a infinite
loop

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message