[
https://issues.apache.org/jira/browse/HADOOP-13595?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15989566#comment-15989566
]
Anu Engineer edited comment on HADOOP-13595 at 4/28/17 11:15 PM:
-----------------------------------------------------------------
bq. A non-privileged user can run diskbalancer and mover? That's... surprising.
Also, mover has the
Sorry, I was not very clear. For diskbalancer you can read the data layout from
the datanode and create plan, you don't have to be an admin for that. But to
submit that plan for execution to the datanode, you need to be an admin.
So the command we are talking about is the client side tool. Actual data
movement is done inside a datanode thread.
was (Author: anu):
bq. A non-privileged user can run diskbalancer and mover? That's... surprising.
Also, mover has the
Sorry, I was not very clear. For diskbalancer you can read the data layout from
the datanode and create plan, you don't have to be an admin for that. But to
submit that plan for execution to the datanode, you need to be an admin.
> Rework hadoop_usage to be broken up by clients/daemons/etc.
> -----------------------------------------------------------
>
> Key: HADOOP-13595
> URL: https://issues.apache.org/jira/browse/HADOOP-13595
> Project: Hadoop Common
> Issue Type: Improvement
> Components: scripts
> Affects Versions: 3.0.0-alpha2
> Reporter: Allen Wittenauer
> Assignee: Allen Wittenauer
> Attachments: HADOOP-13595.00.patch, HADOOP-13595.01.patch
>
>
> Part of the feedback from HADOOP-13341 was that it wasn't obvious what was a
> client and what was a daemon. Reworking the hadoop_usage output so that it
> is obvious helps fix this issue.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
