hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Yongjun Zhang (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-6133) Make Balancer support exclude specified path
Date Thu, 24 Apr 2014 21:01:18 GMT

    [ https://issues.apache.org/jira/browse/HDFS-6133?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13980282#comment-13980282

Yongjun Zhang commented on HDFS-6133:

Thanks [~zhaoyunjiong] for addressing my earlier comments.  [~vinayrpet], thanks for your
additional comments, they very good ones to me.

Yunjiong, when you address Vinayakumar's comments, would you please consider the following:

- Balancer.java, the variable excludeDir really meant "excludePaths", which contains a list
of paths separated by comma, each of the paths could be either a dir and a file. So suggest
to replace "excludeDir" with  "excludePaths".

- Line 422 of NamenodeRpcServer.java exceeded 80 chars.

- NamenodeProtocal.java, I'm suggesting to improve the following comments a bit more 
  *          exclude blocks belongs to files with specified path prefix, use
  *          comma to separate if have multiple paths
   *          exclude blocks belonging to files with path prefix in excludePaths, which
   *          is a single path or a list of paths separated by comma. 

- I happen to see a typo in test code:
  r = Balancer.run(namenodes, Balancer.Parameters.DEFALUT, conf);
The DEFALUT meant DEFAULT, would you please correct it as well?

I'm giving my +1 once you address Vinayakumar's comments and mine above. 

Thanks again for the good work.

> Make Balancer support exclude specified path
> --------------------------------------------
>                 Key: HDFS-6133
>                 URL: https://issues.apache.org/jira/browse/HDFS-6133
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: balancer, namenode
>            Reporter: zhaoyunjiong
>            Assignee: zhaoyunjiong
>         Attachments: HDFS-6133.patch
> Currently, run Balancer will destroying Regionserver's data locality.
> If getBlocks could exclude blocks belongs to files which have specific path prefix, like
"/hbase", then we can run Balancer without destroying Regionserver's data locality.

This message was sent by Atlassian JIRA

View raw message