You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "Lars Hofhansl (JIRA)" <ji...@apache.org> on 2012/11/15 02:06:12 UTC

[jira] [Created] (HBASE-7165) TestSplitLogManager.testUnassignedTimeout is flaky

Lars Hofhansl created HBASE-7165:
------------------------------------

             Summary: TestSplitLogManager.testUnassignedTimeout is flaky
                 Key: HBASE-7165
                 URL: https://issues.apache.org/jira/browse/HBASE-7165
             Project: HBase
          Issue Type: Bug
            Reporter: Lars Hofhansl
             Fix For: 0.96.0, 0.94.4


This is the most frequently failing test in the recent 0.94 jenkins builds.
Looking at the code it seem the timeouts are a bit too aggressive (there's only a 500ms window for the test to detect that a ZK was removed). I think we should just double the timeouts.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

[jira] [Updated] (HBASE-7165) TestSplitLogManager.testUnassignedTimeout is flaky

Posted by "Lars Hofhansl (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HBASE-7165?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Lars Hofhansl updated HBASE-7165:
---------------------------------

    Attachment: 7165.txt

Here's a patch that doubles the TO.
Can also quadruple.
                
> TestSplitLogManager.testUnassignedTimeout is flaky
> --------------------------------------------------
>
>                 Key: HBASE-7165
>                 URL: https://issues.apache.org/jira/browse/HBASE-7165
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Lars Hofhansl
>             Fix For: 0.96.0, 0.94.4
>
>         Attachments: 7165.txt
>
>
> This is the most frequently failing test in the recent 0.94 jenkins builds.
> Looking at the code it seem the timeouts are a bit too aggressive (there's only a 500ms window for the test to detect that a ZK was removed). I think we should just double the timeouts.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

[jira] [Commented] (HBASE-7165) TestSplitLogManager.testUnassignedTimeout is flaky

Posted by "Hudson (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/HBASE-7165?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13497950#comment-13497950 ] 

Hudson commented on HBASE-7165:
-------------------------------

Integrated in HBase-TRUNK-on-Hadoop-2.0.0 #262 (See [https://builds.apache.org/job/HBase-TRUNK-on-Hadoop-2.0.0/262/])
    HBASE-7165 TestSplitLogManager.testUnassignedTimeout is flaky (Revision 1409649)

     Result = FAILURE
larsh : 
Files : 
* /hbase/trunk/hbase-server/src/test/java/org/apache/hadoop/hbase/master/TestSplitLogManager.java

                
> TestSplitLogManager.testUnassignedTimeout is flaky
> --------------------------------------------------
>
>                 Key: HBASE-7165
>                 URL: https://issues.apache.org/jira/browse/HBASE-7165
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Lars Hofhansl
>            Assignee: Lars Hofhansl
>             Fix For: 0.96.0, 0.94.4
>
>         Attachments: 7165.txt
>
>
> This is the most frequently failing test in the recent 0.94 jenkins builds.
> Looking at the code it seem the timeouts are a bit too aggressive (there's only a 500ms window for the test to detect that a ZK was removed). I think we should just double the timeouts.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

[jira] [Commented] (HBASE-7165) TestSplitLogManager.testUnassignedTimeout is flaky

Posted by "Hudson (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/HBASE-7165?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13497836#comment-13497836 ] 

Hudson commented on HBASE-7165:
-------------------------------

Integrated in HBase-0.94 #594 (See [https://builds.apache.org/job/HBase-0.94/594/])
    HBASE-7165 TestSplitLogManager.testUnassignedTimeout is flaky (Revision 1409650)

     Result = FAILURE
larsh : 
Files : 
* /hbase/branches/0.94/src/test/java/org/apache/hadoop/hbase/master/TestSplitLogManager.java

                
> TestSplitLogManager.testUnassignedTimeout is flaky
> --------------------------------------------------
>
>                 Key: HBASE-7165
>                 URL: https://issues.apache.org/jira/browse/HBASE-7165
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Lars Hofhansl
>            Assignee: Lars Hofhansl
>             Fix For: 0.96.0, 0.94.4
>
>         Attachments: 7165.txt
>
>
> This is the most frequently failing test in the recent 0.94 jenkins builds.
> Looking at the code it seem the timeouts are a bit too aggressive (there's only a 500ms window for the test to detect that a ZK was removed). I think we should just double the timeouts.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

[jira] [Commented] (HBASE-7165) TestSplitLogManager.testUnassignedTimeout is flaky

Posted by "stack (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/HBASE-7165?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13497738#comment-13497738 ] 

stack commented on HBASE-7165:
------------------------------

Sounds good (or quadruple)
                
> TestSplitLogManager.testUnassignedTimeout is flaky
> --------------------------------------------------
>
>                 Key: HBASE-7165
>                 URL: https://issues.apache.org/jira/browse/HBASE-7165
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Lars Hofhansl
>             Fix For: 0.96.0, 0.94.4
>
>
> This is the most frequently failing test in the recent 0.94 jenkins builds.
> Looking at the code it seem the timeouts are a bit too aggressive (there's only a 500ms window for the test to detect that a ZK was removed). I think we should just double the timeouts.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

[jira] [Commented] (HBASE-7165) TestSplitLogManager.testUnassignedTimeout is flaky

Posted by "Lars Hofhansl (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/HBASE-7165?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13497791#comment-13497791 ] 

Lars Hofhansl commented on HBASE-7165:
--------------------------------------

I quadruple the TO
                
> TestSplitLogManager.testUnassignedTimeout is flaky
> --------------------------------------------------
>
>                 Key: HBASE-7165
>                 URL: https://issues.apache.org/jira/browse/HBASE-7165
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Lars Hofhansl
>            Assignee: Lars Hofhansl
>             Fix For: 0.96.0, 0.94.4
>
>         Attachments: 7165.txt
>
>
> This is the most frequently failing test in the recent 0.94 jenkins builds.
> Looking at the code it seem the timeouts are a bit too aggressive (there's only a 500ms window for the test to detect that a ZK was removed). I think we should just double the timeouts.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

[jira] [Commented] (HBASE-7165) TestSplitLogManager.testUnassignedTimeout is flaky

Posted by "stack (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/HBASE-7165?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13497752#comment-13497752 ] 

stack commented on HBASE-7165:
------------------------------

+1

Commit quadruple I'd say.
                
> TestSplitLogManager.testUnassignedTimeout is flaky
> --------------------------------------------------
>
>                 Key: HBASE-7165
>                 URL: https://issues.apache.org/jira/browse/HBASE-7165
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Lars Hofhansl
>             Fix For: 0.96.0, 0.94.4
>
>         Attachments: 7165.txt
>
>
> This is the most frequently failing test in the recent 0.94 jenkins builds.
> Looking at the code it seem the timeouts are a bit too aggressive (there's only a 500ms window for the test to detect that a ZK was removed). I think we should just double the timeouts.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

[jira] [Assigned] (HBASE-7165) TestSplitLogManager.testUnassignedTimeout is flaky

Posted by "Lars Hofhansl (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HBASE-7165?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Lars Hofhansl reassigned HBASE-7165:
------------------------------------

    Assignee: Lars Hofhansl
    
> TestSplitLogManager.testUnassignedTimeout is flaky
> --------------------------------------------------
>
>                 Key: HBASE-7165
>                 URL: https://issues.apache.org/jira/browse/HBASE-7165
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Lars Hofhansl
>            Assignee: Lars Hofhansl
>             Fix For: 0.96.0, 0.94.4
>
>         Attachments: 7165.txt
>
>
> This is the most frequently failing test in the recent 0.94 jenkins builds.
> Looking at the code it seem the timeouts are a bit too aggressive (there's only a 500ms window for the test to detect that a ZK was removed). I think we should just double the timeouts.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

[jira] [Commented] (HBASE-7165) TestSplitLogManager.testUnassignedTimeout is flaky

Posted by "Hudson (JIRA)" <ji...@apache.org>.
    [ https://issues.apache.org/jira/browse/HBASE-7165?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13497816#comment-13497816 ] 

Hudson commented on HBASE-7165:
-------------------------------

Integrated in HBase-TRUNK #3545 (See [https://builds.apache.org/job/HBase-TRUNK/3545/])
    HBASE-7165 TestSplitLogManager.testUnassignedTimeout is flaky (Revision 1409649)

     Result = FAILURE
larsh : 
Files : 
* /hbase/trunk/hbase-server/src/test/java/org/apache/hadoop/hbase/master/TestSplitLogManager.java

                
> TestSplitLogManager.testUnassignedTimeout is flaky
> --------------------------------------------------
>
>                 Key: HBASE-7165
>                 URL: https://issues.apache.org/jira/browse/HBASE-7165
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Lars Hofhansl
>            Assignee: Lars Hofhansl
>             Fix For: 0.96.0, 0.94.4
>
>         Attachments: 7165.txt
>
>
> This is the most frequently failing test in the recent 0.94 jenkins builds.
> Looking at the code it seem the timeouts are a bit too aggressive (there's only a 500ms window for the test to detect that a ZK was removed). I think we should just double the timeouts.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

[jira] [Resolved] (HBASE-7165) TestSplitLogManager.testUnassignedTimeout is flaky

Posted by "Lars Hofhansl (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HBASE-7165?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Lars Hofhansl resolved HBASE-7165.
----------------------------------

      Resolution: Fixed
    Hadoop Flags: Reviewed

Committed to 0.94 and 0.96. Let's hope that these are a bit more stable from now on.
                
> TestSplitLogManager.testUnassignedTimeout is flaky
> --------------------------------------------------
>
>                 Key: HBASE-7165
>                 URL: https://issues.apache.org/jira/browse/HBASE-7165
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Lars Hofhansl
>             Fix For: 0.96.0, 0.94.4
>
>         Attachments: 7165.txt
>
>
> This is the most frequently failing test in the recent 0.94 jenkins builds.
> Looking at the code it seem the timeouts are a bit too aggressive (there's only a 500ms window for the test to detect that a ZK was removed). I think we should just double the timeouts.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira