You are viewing a plain text version of this content. The canonical link for it is here.
Posted to mapreduce-issues@hadoop.apache.org by "Dick King (JIRA)" <ji...@apache.org> on 2009/07/17 23:43:14 UTC
[jira] Created: (MAPREDUCE-770)
org.apache.hadoop.tools.TestCopyFiles may leave junk files when an
assertion fails
org.apache.hadoop.tools.TestCopyFiles may leave junk files when an assertion fails
----------------------------------------------------------------------------------
Key: MAPREDUCE-770
URL: https://issues.apache.org/jira/browse/MAPREDUCE-770
Project: Hadoop Map/Reduce
Issue Type: Bug
Reporter: Dick King
In most of the testXxxYyyZzz methods, the code runs:
preliminaries
ToolRunner( ..., new String[] { local and DFS filenames and more filenames });
assertMaybe("this result stank", conditions);
assertMaybe("this other result stank", conditions);
deldir(deletee's name);
deldir(second deletee's name);
The assertMaybe's throw AssertionFailedError . That's what they DO. Shouldn't this stuff be protected with a try ... finally construct?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (MAPREDUCE-770)
org.apache.hadoop.tools.TestCopyFiles may leave junk files when an
assertion fails
Posted by "Tsz Wo (Nicholas), SZE (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MAPREDUCE-770?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12732768#action_12732768 ]
Tsz Wo (Nicholas), SZE commented on MAPREDUCE-770:
--------------------------------------------------
> ... Some of the junk files are left on HDFS .
I may be wrong. I think the junk files are left in MiniDFSCluster. In such case, ant clean should remove everything.
> org.apache.hadoop.tools.TestCopyFiles may leave junk files when an assertion fails
> ----------------------------------------------------------------------------------
>
> Key: MAPREDUCE-770
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-770
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: test
> Reporter: Dick King
> Priority: Minor
>
> In most of the testXxxYyyZzz methods, the code runs:
> preliminaries
> ToolRunner( ..., new String[] { local and DFS filenames and more filenames });
> assertMaybe("this result stank", conditions);
> assertMaybe("this other result stank", conditions);
> deldir(deletee's name);
> deldir(second deletee's name);
> The assertMaybe's throw AssertionFailedError . That's what they DO. Shouldn't this stuff be protected with a try ... finally construct?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Updated: (MAPREDUCE-770)
org.apache.hadoop.tools.TestCopyFiles may leave junk files when an
assertion fails
Posted by "Tsz Wo (Nicholas), SZE (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MAPREDUCE-770?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Tsz Wo (Nicholas), SZE updated MAPREDUCE-770:
---------------------------------------------
Component/s: test
Priority: Minor (was: Major)
> That's what they DO. Shouldn't this stuff be protected with a try ... finally construct?
Yes, this is a good improvement over the tests.
Change the priority to minor since this helps when the unit test fails with an AssertionFailedError but the unit test is not supposed to fail. In the case that the unit test fails, we could do "ant clean".
> org.apache.hadoop.tools.TestCopyFiles may leave junk files when an assertion fails
> ----------------------------------------------------------------------------------
>
> Key: MAPREDUCE-770
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-770
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: test
> Reporter: Dick King
> Priority: Minor
>
> In most of the testXxxYyyZzz methods, the code runs:
> preliminaries
> ToolRunner( ..., new String[] { local and DFS filenames and more filenames });
> assertMaybe("this result stank", conditions);
> assertMaybe("this other result stank", conditions);
> deldir(deletee's name);
> deldir(second deletee's name);
> The assertMaybe's throw AssertionFailedError . That's what they DO. Shouldn't this stuff be protected with a try ... finally construct?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (MAPREDUCE-770)
org.apache.hadoop.tools.TestCopyFiles may leave junk files when an
assertion fails
Posted by "Dick King (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MAPREDUCE-770?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12732771#action_12732771 ]
Dick King commented on MAPREDUCE-770:
-------------------------------------
https://issues.apache.org/jira/browse/MAPREDUCE-770?focusedCommentId=12732768&page=com.atlassian.jira.plugin.system.issuetabpanels%3Acomment-tabpanel#action_12732768 could well be correct.
I don't claim to have enough knowledge to refute that claim, and it's plausible.
> org.apache.hadoop.tools.TestCopyFiles may leave junk files when an assertion fails
> ----------------------------------------------------------------------------------
>
> Key: MAPREDUCE-770
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-770
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: test
> Reporter: Dick King
> Priority: Minor
>
> In most of the testXxxYyyZzz methods, the code runs:
> preliminaries
> ToolRunner( ..., new String[] { local and DFS filenames and more filenames });
> assertMaybe("this result stank", conditions);
> assertMaybe("this other result stank", conditions);
> deldir(deletee's name);
> deldir(second deletee's name);
> The assertMaybe's throw AssertionFailedError . That's what they DO. Shouldn't this stuff be protected with a try ... finally construct?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
[jira] Commented: (MAPREDUCE-770)
org.apache.hadoop.tools.TestCopyFiles may leave junk files when an
assertion fails
Posted by "Dick King (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/MAPREDUCE-770?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12732764#action_12732764 ]
Dick King commented on MAPREDUCE-770:
-------------------------------------
ant clean does not solve the problem. Some of the junk files are left on HDFS .
> org.apache.hadoop.tools.TestCopyFiles may leave junk files when an assertion fails
> ----------------------------------------------------------------------------------
>
> Key: MAPREDUCE-770
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-770
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Components: test
> Reporter: Dick King
> Priority: Minor
>
> In most of the testXxxYyyZzz methods, the code runs:
> preliminaries
> ToolRunner( ..., new String[] { local and DFS filenames and more filenames });
> assertMaybe("this result stank", conditions);
> assertMaybe("this other result stank", conditions);
> deldir(deletee's name);
> deldir(second deletee's name);
> The assertMaybe's throw AssertionFailedError . That's what they DO. Shouldn't this stuff be protected with a try ... finally construct?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.