You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@oozie.apache.org by "Hadoop QA (JIRA)" <ji...@apache.org> on 2011/09/08 06:47:09 UTC
[jira] [Created] (OOZIE-186) GH-208: Initialization of FS instances
are wrong in many places and fail with Kerberos
GH-208: Initialization of FS instances are wrong in many places and fail with Kerberos
--------------------------------------------------------------------------------------
Key: OOZIE-186
URL: https://issues.apache.org/jira/browse/OOZIE-186
Project: Oozie
Issue Type: Bug
Reporter: Hadoop QA
Empty configurations are used/propagated and this causes Oozie to fail when Kerberos is ON.
Testcases fail depending on the order they are executed. If a FS is already in the cache with the right values, things are OK, otherwise, they fail.
This creates a difficult to trace problem.
It seems related to to Configuration initialization (mostly for testcases) where the K principal rules are not present.
Coordinator commands have several issues on this regard.
Addressing the ones that I'm finding now in this patch with minimal changes.
However, commands logic must be done properly as part of refactorying/cleanup
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Closed] (OOZIE-186) GH-208: Initialization of FS instances
are wrong in many places and fail with Kerberos
Posted by "Roman Shaposhnik (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/OOZIE-186?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Roman Shaposhnik closed OOZIE-186.
----------------------------------
Resolution: Fixed
> GH-208: Initialization of FS instances are wrong in many places and fail with Kerberos
> --------------------------------------------------------------------------------------
>
> Key: OOZIE-186
> URL: https://issues.apache.org/jira/browse/OOZIE-186
> Project: Oozie
> Issue Type: Bug
> Reporter: Hadoop QA
>
> Empty configurations are used/propagated and this causes Oozie to fail when Kerberos is ON.
> Testcases fail depending on the order they are executed. If a FS is already in the cache with the right values, things are OK, otherwise, they fail.
> This creates a difficult to trace problem.
> It seems related to to Configuration initialization (mostly for testcases) where the K principal rules are not present.
> Coordinator commands have several issues on this regard.
> Addressing the ones that I'm finding now in this patch with minimal changes.
> However, commands logic must be done properly as part of refactorying/cleanup
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (OOZIE-186) GH-208: Initialization of FS
instances are wrong in many places and fail with Kerberos
Posted by "Hadoop QA (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/OOZIE-186?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13099871#comment-13099871 ]
Hadoop QA commented on OOZIE-186:
---------------------------------
tucu00 remarked:
Closed by e5b4e90335b40ef46a18321f6fb3b7c79b736675 Initialization of FS instances are wrong in many places and fail with Kerberos
Empty configurations are used/propagated and this causes Oozie to fail when Kerberos is ON.
Testcases fail depending on the order they are executed. If a FS is already in the cache
with the right values, things are OK, otherwise, they fail.
This creates a difficult to trace problem.
It seems related to to Configuration initialization (mostly for testcases) where the K principal
rules are not present.
Coordinator commands have several issues on this regard.
Addressing the ones that I'm finding now in this patch with minimal changes.
However, commands logic must be done properly as part of refactorying/cleanup
> GH-208: Initialization of FS instances are wrong in many places and fail with Kerberos
> --------------------------------------------------------------------------------------
>
> Key: OOZIE-186
> URL: https://issues.apache.org/jira/browse/OOZIE-186
> Project: Oozie
> Issue Type: Bug
> Reporter: Hadoop QA
>
> Empty configurations are used/propagated and this causes Oozie to fail when Kerberos is ON.
> Testcases fail depending on the order they are executed. If a FS is already in the cache with the right values, things are OK, otherwise, they fail.
> This creates a difficult to trace problem.
> It seems related to to Configuration initialization (mostly for testcases) where the K principal rules are not present.
> Coordinator commands have several issues on this regard.
> Addressing the ones that I'm finding now in this patch with minimal changes.
> However, commands logic must be done properly as part of refactorying/cleanup
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira