You are viewing a plain text version of this content. The canonical link for it is here.
Posted to hdfs-issues@hadoop.apache.org by "Faraz Ahmad (JIRA)" <ji...@apache.org> on 2010/12/05 19:23:11 UTC
[jira] Created: (HDFS-1528) First uploaded file replaces
/home/hadoop directory
First uploaded file replaces /home/hadoop directory
---------------------------------------------------
Key: HDFS-1528
URL: https://issues.apache.org/jira/browse/HDFS-1528
Project: Hadoop HDFS
Issue Type: Bug
Components: name-node
Affects Versions: 0.20.2
Environment: distributed file system
Reporter: Faraz Ahmad
This bug shows up in the following scenario:
1. There is no input data on HDFS (i.e., it has been formatted recently).
2. Upload any single file (not directory of files) e.g., build.xml.
3. The original directory /user/hadoop on hdfs will change itself from directory to a file and its contents will contain the contents of the uploaded file.
4. Now HDFS will have a single file /user/hadoop.
5. HDFS filesystem gets corrupted and we cannot upload any more files or directories.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (HDFS-1528) First uploaded file replaces
/home/hadoop directory
Posted by "Faraz Ahmad (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HDFS-1528?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12966996#action_12966996 ]
Faraz Ahmad commented on HDFS-1528:
-----------------------------------
A quick work around which keeps file system intact is :
1. Create an empty directory on hdfs.
2. Then upload the input file.
> First uploaded file replaces /home/hadoop directory
> ---------------------------------------------------
>
> Key: HDFS-1528
> URL: https://issues.apache.org/jira/browse/HDFS-1528
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: name-node
> Affects Versions: 0.20.2
> Environment: distributed file system
> Reporter: Faraz Ahmad
>
> This bug shows up in the following scenario:
> 1. There is no input data on HDFS (i.e., it has been formatted recently).
> 2. Upload any single file (not directory of files) e.g., build.xml.
> 3. The original directory /user/hadoop on hdfs will change itself from directory to a file and its contents will contain the contents of the uploaded file.
> 4. Now HDFS will have a single file /user/hadoop.
> 5. HDFS filesystem gets corrupted and we cannot upload any more files or directories.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.