hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Nick Dimiduk (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-8018) Add "Flaky Testcase Dector" tool into dev-tools
Date Wed, 17 Apr 2013 19:21:17 GMT

    [ https://issues.apache.org/jira/browse/HBASE-8018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13634327#comment-13634327
] 

Nick Dimiduk commented on HBASE-8018:
-------------------------------------

{noformat}
Pre-requirement
       Please download jenkins-client from https://github.com/jeffreyz88/jenkins-tools/tree/master/jenkins-client
and put folder jenkins-client under jenkins-tools. 
{noformat}

Is this {{jenkins-client}} on your repo patched from the upstream version? Can we just pull
it in as a dependency instead of downloading it manually?

{noformat}
Usage are: 

        java -jar ./buildstats/target/buildstats.jar <Jenkins HTTP URL> <Job Name>
[number of last most recent jobs to check]
{noformat}

How about bundling a launch script that builds the classpath for the user?
                
> Add "Flaky Testcase Dector" tool into dev-tools
> -----------------------------------------------
>
>                 Key: HBASE-8018
>                 URL: https://issues.apache.org/jira/browse/HBASE-8018
>             Project: HBase
>          Issue Type: Bug
>          Components: util
>            Reporter: Jeffrey Zhong
>            Assignee: Jeffrey Zhong
>             Fix For: 0.98.0
>
>         Attachments: hbase-8018.patch
>
>
> Recently I'm working on some hbase test case failures and I think it's useful if we can
see a report of all failed test cases from most recent runs. So we can easily see how flaky
a test case is. I wrote a tool some time back and below are some reports against different
branches from today's run. You can get the tools' source at https://github.com/jeffreyz88/jenkins-tools.
If we can run it daily and send out an email then we can quickly notice what may break after
recent check-ins. 
> Notes: 
> 1) " 873  874  875  876  877  878  879  880  881" are builds which has failed test cases
in current or previous runs
> 2) 1 means "PASSED", 0 means "NOT RUN AT ALL", -1 means "FAILED"
> HBase-0.95 (from last 10 runs configurable)
> Failed Test Cases                21   22   23   24   25   27
> ========================================================
> org.apache.hadoop.hbase.catalog.testmetamigrationconvertingtopb.org.apache.hadoop.hbase.catalog.testmetamigrationconvertingtopb
   0    0    0    0   -1   -1
> org.apache.hadoop.hbase.coprocessor.example.testbulkdeleteprotocol.testbulkdeletecolumn
  -1    0    0    0    0    0
> org.apache.hadoop.hbase.coprocessor.example.testrowcountendpoint.org.apache.hadoop.hbase.coprocessor.example.testrowcountendpoint
  -1    0    0    0    0    0
> org.apache.hadoop.hbase.coprocessor.example.testzookeeperscanpolicyobserver.org.apache.hadoop.hbase.coprocessor.example.testzookeeperscanpolicyobserver
  -1    0    0    0    0    0
> org.apache.hadoop.hbase.master.testmasterfailover.testmasterfailoverwithmockedritondeadrs
   1   -1   -1    0    1    1
> org.apache.hadoop.hbase.regionserver.testsplittransactiononcluster.testshouldthrowioexceptionifstorefilesizeisemptyandshouldsuccessfullyexecuterollback
   1    1    1    1    1   -1
> org.apache.hadoop.hbase.regionserver.testsplittransactiononcluster.testshutdownfixupwhendaughterhassplit
   1    1    1   -1   -1   -1
> org.apache.hadoop.hbase.regionserver.wal.testhlog.testlogcleaning    0    1    1    1
  -1    0
> org.apache.hadoop.hbase.replication.testmasterreplication.testcyclicreplication    1
   1    1    1    1   -1
> org.apache.hadoop.hbase.replication.testreplicationqueuefailover.queuefailover    1 
 -1    0    1   -1    0
> org.apache.hadoop.hbase.replication.testreplicationqueuefailovercompressed.queuefailover
   0    1   -1    0   -1    0
> org.apache.hadoop.hbase.security.access.testaccesscontroller.org.apache.hadoop.hbase.security.access.testaccesscontroller
   0   -1    0    0    0    0
> As you can see we have a few test cases don't run successfully at all or recently.
> HBase-TRUNK (from last 10 runs)
> Failed Test Cases              3908 3909 3910 3912 3913 3914 3915 3916
> ========================================================
> org.apache.hadoop.hbase.catalog.testmetamigrationconvertingtopb.org.apache.hadoop.hbase.catalog.testmetamigrationconvertingtopb
   0   -1    0    0    0    0   -1   -1
> org.apache.hadoop.hbase.client.testadmin.testcloseregionwhenservernameisempty    1  
 1    1    1    1    1   -1    0
> org.apache.hadoop.hbase.client.testscannertimeout.test3686a    1    1   -1    0    1
   1    1    1
> org.apache.hadoop.hbase.client.testsnapshotcloneindependence.testofflinesnapshotregionoperationsindependent
   0    1   -1    0    1    1    1    1
> org.apache.hadoop.hbase.client.testsnapshotcloneindependence.testonlinesnapshotregionoperationsindependent
   1    1   -1    0    1    1    1    1
> org.apache.hadoop.hbase.master.testassignmentmanageroncluster.testmoveregion    1   
1   -1    0    1    1    1    1
> org.apache.hadoop.hbase.master.testdistributedlogsplitting.testdelayeddeleteonfailure
   1    1   -1    0    1    1    1    1
> org.apache.hadoop.hbase.master.testmasterfailover.testmasterfailoverwithmockedritondeadrs
   1    1    1    1    1   -1    0    1
> org.apache.hadoop.hbase.master.testtablelockmanager.testdelete    1    1   -1    0  
 1    1    1    1
> org.apache.hadoop.hbase.procedure.testzkprocedurecontrollers.testsimplezkcohortmembercontroller
   1    1    1    1    1    1   -1    0
> org.apache.hadoop.hbase.regionserver.testsplittransactiononcluster.testexistingznodeblockssplitandwerollback
   1    1    1    1    1   -1    0    1
> org.apache.hadoop.hbase.regionserver.testsplittransactiononcluster.testshouldthrowioexceptionifstorefilesizeisemptyandshouldsuccessfullyexecuterollback
   1    1   -1    0    1    1    1    1
> org.apache.hadoop.hbase.regionserver.testsplittransactiononcluster.testshutdownfixupwhendaughterhassplit
  -1   -1   -1    0   -1    0   -1    0
> org.apache.hadoop.hbase.regionserver.wal.testhlog.testlogcleaning    0   -1   -1    0
   1    1    1    1
> org.apache.hadoop.hbase.replication.testmultislavereplication.testmultislavereplication
   1    1   -1    0    1    1    1    1
> org.apache.hadoop.hbase.replication.testreplicationqueuefailover.queuefailover    1 
  1    1   -1    0   -1    0    1
> org.apache.hadoop.hbase.replication.testreplicationqueuefailovercompressed.queuefailover
   1    1    1   -1    0   -1    0    1
> org.apache.hadoop.hbase.security.access.testaccesscontrolfilter.testqualifieraccess 
 -1    0    1   -1    0    1    1    1
> org.apache.hadoop.hbase.util.testhbasefsck.testdegenerateregions    1   -1    0    1
   1    1    1    1
> org.apache.hadoop.hbase.util.testhbasefsck.testfixassignmentsandnohdfschecking    1 
 -1    0    1    1    1    1    1
> org.apache.hadoop.hbase.util.testhbasefsck.testfixbytable    1   -1    0    1    1  
 1    1    1
> org.apache.hadoop.hbase.util.testhbasefsck.testhbckthreadpooling    1   -1    0    1
   1    1    1    1
> org.apache.hadoop.hbase.util.testhbasefsck.testlingeringreferencefile    1   -1    0
   1    1    1    1    1
> org.apache.hadoop.hbase.util.testhbasefsck.testmissingfirstregion    1   -1    0    1
   1    1    1    1
> org.apache.hadoop.hbase.util.testhbasefsck.testmissinglastregion    1   -1    0    1
   1    1    1    1
> org.apache.hadoop.hbase.util.testhbasefsck.testnotinhdfs    1   -1    0    1    1   
1    1    1
> org.apache.hadoop.hbase.util.testhbasefsck.testnotinmetaordeployedhole    1   -1    0
   1    1    1    1    1
> org.apache.hadoop.hbase.util.testhbasefsck.testoverlapandorphan    1   -1    0    1 
  1    1    1    1
> org.apache.hadoop.hbase.util.testhbasefsck.testregionhole    1   -1    0    1    1  
 1    1    1
> org.apache.hadoop.hbase.util.testhbasefsck.testregionshouldnotbedeployed    1   -1  
 0    1    1    1    1    1
> org.apache.hadoop.hbase.util.testhbasefsck.testsidelineoverlapregion    1   -1    0 
  1    1    1    1    1
> org.apache.hadoop.hbase.util.testhbasefsck.testsplitdaughtersnotinmeta    1   -1    0
   1    1    1    1    1
> More test cases are not in good shape.
> HBase-0.94(from last 10 runs)
> Failed Test Cases               873  874  875  876  877  878  879  880  881
> ========================================================
> org.apache.hadoop.hbase.backup.testhfilearchiving.testcleaningrace   -1    0    1   
1    1    1    1    1    1
> org.apache.hadoop.hbase.catalog.testmetareadereditor.testgetregion    1    1   -1   
0    1    1    1    1    1
> org.apache.hadoop.hbase.catalog.testmetareadereditor.testretrying    1    1   -1    0
   1    1    1    1    1
> org.apache.hadoop.hbase.catalog.testmetareadereditor.testscanmetafortable    1    1 
 -1    0    1    1    1    1    1
> org.apache.hadoop.hbase.catalog.testmetareadereditor.testtableexists    1    1   -1 
  0    1    1    1    1    1
> org.apache.hadoop.hbase.client.testmetamigrationremovinghtd.org.apache.hadoop.hbase.client.testmetamigrationremovinghtd
   0   -1    0    0    0    0    0    0    0
> org.apache.hadoop.hbase.client.testmultiparallel.testflushcommitsnoabort    1    1  
 1    1    1   -1    0    1    1
> org.apache.hadoop.hbase.client.testmultiparallel.testflushcommitswithabort    1    1
   1    1    1   -1    0    1    1
> org.apache.hadoop.hbase.client.testshell.testrunshelltests    1    1    1    1    1 
  1   -1    0    1
> org.apache.hadoop.hbase.master.testmasterfailover.testmasterfailoverwithmockedritondeadrs
   1    1   -1   -1    0    1    1    1    1
> org.apache.hadoop.hbase.regionserver.testatomicoperation.testmultirowmutationmultithreads
   1    1    1   -1    0    1    1    1    1
> org.apache.hadoop.hbase.regionserver.testsplittransactiononcluster.testshutdownfixupwhendaughterhassplit
   1   -1    0    1    1    1    1    1    1 
> As you can see 0.94 is stable and testmetamigrationremovinghtd  hasn't been run for a
while either being removed or due to other failures. Some test cases aren't stable but not
too bad. 
> Thanks,
> -Jeffrey

--
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

Mime
View raw message