hadoop-common-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Xuan Gong (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HADOOP-9057) TestMetricsSystemImpl.testInitFirst fails intermittently
Date Fri, 30 Nov 2012 20:57:58 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-9057?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13507645#comment-13507645
] 

Xuan Gong commented on HADOOP-9057:
-----------------------------------

The timeout in mockito verify is a great idea. It actually verify the callback before we call
the stop function. 

But I think that we may need add another test case to verify the stop(). I think that the
purpose of this stop()  is when I call this function, all the sink threads should stop no
matter they are idle or they are doing something. 

And actually the test is fail, on the other side, it proved that our stop function is actually
worked correctly. When we call the stop, everything is stopped no matter what they are doing
right now. Based on that, I runned this test several times, I found that sometimes it will
be successful, sometimes half rigth half wrong, sometimes all fails.
That is why orignally we try to add Thread.sleep() and try to buy some times for the sink
threads to consume all the items in queue.

Maybe you can take a look at the HADOOP-8981, this is the duplication issue. I uploaded patches
verion 3, including the test case that using the mockito timeout and another test case to
verify stop function. 
                
> TestMetricsSystemImpl.testInitFirst fails intermittently
> --------------------------------------------------------
>
>                 Key: HADOOP-9057
>                 URL: https://issues.apache.org/jira/browse/HADOOP-9057
>             Project: Hadoop Common
>          Issue Type: Bug
>    Affects Versions: 1-win
>            Reporter: Ivan Mitic
>            Assignee: Ivan Mitic
>         Attachments: HADOOP-9057.branch-1-win.metricsrace.patch
>
>
> Error Message
>  Wanted but not invoked: metricsSink.putMetrics(<Capturing argument>); -> at
org.apache.hadoop.metrics2.impl.TestMetricsSystemImpl.testInitFirst(TestMetricsSystemImpl.java:80)
Actually, there were zero interactions with this mock. 
> Stacktrace
> Wanted but not invoked:
> metricsSink.putMetrics(<Capturing argument>);
> -> at org.apache.hadoop.metrics2.impl.TestMetricsSystemImpl.testInitFirst(TestMetricsSystemImpl.java:80)
> Actually, there were zero interactions with this mock.
> 	at org.apache.hadoop.metrics2.impl.TestMetricsSystemImpl.testInitFirst(TestMetricsSystemImpl.java:80)
> 	at org.mockito.internal.runners.JUnit45AndHigherRunnerImpl.run(JUnit45AndHigherRunnerImpl.java:37)
> 	at org.mockito.runners.MockitoJUnitRunner.run(MockitoJUnitRunner.java:62)
> Standard Output
> 2012-10-04 11:43:55,641 INFO  impl.MetricsConfig (MetricsConfig.java:loadFirst(99)) -
loaded properties from hadoop-metrics2-test.properties

--
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