[ 
https://issues.apache.org/jira/browse/HDFS-13356?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16448932#comment-16448932
 ] 

Konstantin Shvachko commented on HDFS-13356:
--------------------------------------------

+1 looks good.

> Balancer:Set default value of minBlockSize to 10mb 
> ---------------------------------------------------
>
>                 Key: HDFS-13356
>                 URL: https://issues.apache.org/jira/browse/HDFS-13356
>             Project: Hadoop HDFS
>          Issue Type: Bug
>          Components: balancer & mover
>    Affects Versions: 2.7.5
>            Reporter: Bharat Viswanadham
>            Assignee: Bharat Viswanadham
>            Priority: Major
>              Labels: balancer, upgrades
>         Attachments: HDFS-13356.00.patch, HDFS-13356.01.patch, 
> HDFS-13356.02.patch
>
>
>  It seems we can run into a problem while a rolling upgrade with this.
> The Balancer is upgraded after NameNodes, so once NN is upgraded it will 
> expect {{minBlockSize}} parameter in {{getBlocks()}}. The Balancer cannot 
> send it yet, so NN will use the default, which you set to 0. So NN will start 
> unexpectedly sending small blocks to the Balancer. So we should
>  # either change the default in protobuf to 10 MB
>  # or treat {{minBlockSize == 0}} in {{NameNodeRpcServer}} as a signal to use 
> the configuration variable 
> {{DFSConfigKeys.DFS_BALANCER_GETBLOCKS_MIN_BLOCK_SIZE_KEY}}.
> If you agree, we should create a follow up jira. I wanted to backport this 
> down the chain of branches, but this upgrade scenario is stopping me.
> [~shv] commented this in  HDFS-13222 jira.
> https://issues.apache.org/jira/browse/HDFS-13222?focusedCommentId=16414855&page=com.atlassian.jira.plugin.system.issuetabpanels%3Acomment-tabpanel#comment-16414855



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org

Reply via email to