You are viewing a plain text version of this content. The canonical link for it is here.
Posted to common-dev@hadoop.apache.org by "Zheng Shao (JIRA)" <ji...@apache.org> on 2008/04/24 08:17:21 UTC
[jira] Created: (HADOOP-3303) Hadoop Streaming shoulf fail if user
process exits with non-zero exit code
Hadoop Streaming shoulf fail if user process exits with non-zero exit code
--------------------------------------------------------------------------
Key: HADOOP-3303
URL: https://issues.apache.org/jira/browse/HADOOP-3303
Project: Hadoop Core
Issue Type: Bug
Components: contrib/streaming
Reporter: Zheng Shao
Assignee: Zheng Shao
Hadoop streaming quietly ignores the user process exit code. This is extremely dangerous because the user won't notice the problem while the output data is likely to be corrupted or incomplete.
User process may have read all input data into input buffer, and then finds a problem in the input data, and exit with an error code and hopes that hadoop streaming can capture that.
The current hadoop streaming just put this exit code into a log file and ignores it.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Resolved: (HADOOP-3303) Hadoop Streaming shoulf fail if user
process exits with non-zero exit code
Posted by "Amareshwari Sriramadasu (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/HADOOP-3303?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Amareshwari Sriramadasu resolved HADOOP-3303.
---------------------------------------------
Resolution: Duplicate
This is duplicate of HADOOP-2057
> Hadoop Streaming shoulf fail if user process exits with non-zero exit code
> --------------------------------------------------------------------------
>
> Key: HADOOP-3303
> URL: https://issues.apache.org/jira/browse/HADOOP-3303
> Project: Hadoop Core
> Issue Type: Bug
> Components: contrib/streaming
> Reporter: Zheng Shao
> Assignee: Zheng Shao
>
> Hadoop streaming quietly ignores the user process exit code. This is extremely dangerous because the user won't notice the problem while the output data is likely to be corrupted or incomplete.
> User process may have read all input data into input buffer, and then finds a problem in the input data, and exit with an error code and hopes that hadoop streaming can capture that.
> The current hadoop streaming just put this exit code into a log file and ignores it.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.