[ 
https://issues.apache.org/jira/browse/HBASE-14702?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

stack updated HBASE-14702:
--------------------------
    Attachment: 14702.txt

Changing the Mockito assert to be atLeast instead of equals.

> TestZKProcedureControllers.testZKCoordinatorControllerWithSingleMemberCohort 
> is a flakey
> ----------------------------------------------------------------------------------------
>
>                 Key: HBASE-14702
>                 URL: https://issues.apache.org/jira/browse/HBASE-14702
>             Project: HBase
>          Issue Type: Sub-task
>          Components: flakey, test
>            Reporter: stack
>         Attachments: 14702.txt
>
>
> Tests fail in TestZKProcedureControllers with some regularity. Here is an 
> example: 
> https://builds.apache.org/view/H-L/view/HBase/job/HBase-1.3/309/jdk=latest1.8,label=Hadoop/testReport/org.apache.hadoop.hbase.procedure/TestZKProcedureControllers/testZKCoordinatorControllerWithSingleMemberCohort/
> {code}
> org.mockito.exceptions.verification.TooManyActualInvocations: 
> procedureMember.submitSubprocedure(<any>);
> Wanted 1 time:
> -> at 
> org.apache.hadoop.hbase.procedure.TestZKProcedureControllers.verifyCohort(TestZKProcedureControllers.java:364)
> But was 2 times. Undesired invocation:
> -> at 
> org.apache.hadoop.hbase.procedure.ZKProcedureMemberRpcs.startNewSubprocedure(ZKProcedureMemberRpcs.java:214)
>       at 
> org.apache.hadoop.hbase.procedure.TestZKProcedureControllers.verifyCohort(TestZKProcedureControllers.java:364)
>       at 
> org.apache.hadoop.hbase.procedure.TestZKProcedureControllers.runMockCommitWithOrchestratedControllers(TestZKProcedureControllers.java:235)
>       at 
> org.apache.hadoop.hbase.procedure.TestZKProcedureControllers.testZKCoordinatorControllerWithSingleMemberCohort(TestZKProcedureControllers.java:158)
> {code}
> Tests in this suite fail always in this Mockito verify step. Always double 
> the number of calls expected. Strikes me as a Mockito bug or Mockito usage 
> interesting behavior. Unless someone intercedes, I'm just going to loosen 
> this assert to be at least N rather than exactly N.
> Lets see if it fails again soon... so I have another example to paste in here.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to