[
https://issues.apache.org/jira/browse/HDFS-4711?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13635619#comment-13635619
]
Colin Patrick McCabe commented on HDFS-4711:
--------------------------------------------
I don't think this is a good idea. Consider moving a large directory with
hundreds of thousands of files. Are you going to check each file to see if its
replication is equal to the current dfs.replication setting? And if it's not,
are you going to do setrep on it? This will also create a lot of problems for
current users, since they'll find that mv unexpectedly changes their
replication settings. Remember that dfs.replication is usually set to a value
due to the configuration XML files, even if the user didn't set it explicitly.
> Can not change replication factor of file during moving or deleting.
> --------------------------------------------------------------------
>
> Key: HDFS-4711
> URL: https://issues.apache.org/jira/browse/HDFS-4711
> Project: Hadoop HDFS
> Issue Type: Bug
> Affects Versions: 2.0.0-alpha
> Reporter: Vladimir Barinov
> Assignee: Colin Patrick McCabe
> Priority: Minor
>
> I don't know is it a feature or a bug.
> According to hdfs dfs -help we can use key -D to set specific options for
> action;
> When we copying or uploading file to hdfs we can override replication factor
> with -D dfs.replication=N. That's works well.
> But it doesn't work for moving or removing(to trash) file.
> Steps to reproduce:
> Uploading file
> hdfs dfs -put somefile /tmp/somefile
> Copying with changing replication:
> hdfs dfs -D dfs.replication=1 -mv /tmp/somefile /tmp/somefile2
> hadoop version:
> Hadoop 2.0.0-cdh4.1.2
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
