hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mukund Madhugiri (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-3012) dfs -mv file to user home directory fails silently if the user home directory does not exist
Date Thu, 13 Mar 2008 21:02:24 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-3012?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12578489#action_12578489
] 

Mukund Madhugiri commented on HADOOP-3012:
------------------------------------------

This worked yesterday and the changes since then are:

HADOOP-2423.  Code optimization in FSNamesystem.mkdirs.
HADOOP-2908.  A document that describes the DFS Shell command.
HADOOP-2994. Code cleanup for DFSClient: remove redundant conversions from string to string.

HADOOP-2972. Fix for a NPE in FSDataset.invalidate.
HADOOP-2995. Fixes StreamBaseRecordReader's getProgress to return a floating point number.
HADOOP-2996. Fixes uses of StringBuffer in StreamUtils class.

I suspect the first one is the cause:
HADOOP-2423.  Code optimization in FSNamesystem.mkdirs.

> dfs -mv file to user home directory fails silently if the user home directory does not
exist
> --------------------------------------------------------------------------------------------
>
>                 Key: HADOOP-3012
>                 URL: https://issues.apache.org/jira/browse/HADOOP-3012
>             Project: Hadoop Core
>          Issue Type: Bug
>          Components: fs
>    Affects Versions: 0.17.0
>            Reporter: Mukund Madhugiri
>            Assignee: Robert Chansler
>            Priority: Blocker
>             Fix For: 0.17.0
>
>
> dfs -mv file to user home directory fails silently if the user home directory does not
exist. This was working before. It will move the file if the user home directory exists.
> Here is the sequence:
> 1. bin/hadoop dfs -lsr /
> /file1  <r 3>   0       2008-03-13 19:54        rw-r--r--       hadoopqa      
 supergroup
> 2. bin/hadoop dfs -mv /file1 file2
> 3. bin/hadoop dfs -lsr /
> /file1  <r 3>   0       2008-03-13 19:54        rw-r--r--       hadoopqa      
 supergroup
> 4. hadoop dfs -lsr /
> /file1  <r 3>   0       2008-03-13 19:54        rw-r--r--       hadoopqa      
 supergroup
> /user   <dir>           2008-03-13 20:07        rwxr-xr-x       hadoopqa      
 supergroup
> /user/hadoopqa  <dir>           2008-03-13 20:07        rwxr-xr-x       hadoopqa
       supergroup
> /user/hadoopqa/file0    <r 3>   0       2008-03-13 20:07        rw-r--r--     
 hadoopqa        supergroup
> 5. bin/hadoop dfs -mv /file1 file2
> 6. bin/hadoop dfs -lsr /
> /user   <dir>           2008-03-13 20:07        rwxr-xr-x       hadoopqa      
 supergroup
> /user/hadoopqa  <dir>           2008-03-13 20:08        rwxr-xr-x       hadoopqa
       supergroup
> /user/hadoopqa/file0    <r 3>   0       2008-03-13 20:07        rw-r--r--     
 hadoopqa        supergroup
> /user/hadoopqa/file2    <r 3>   0       2008-03-13 19:54        rw-r--r--     
 hadoopqa        supergroup
> In step #2, it fails to move the file.
> In step #5, it moves the file as /user/hadoopqa directory exists.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message