[ 
https://issues.apache.org/jira/browse/SOLR-17000?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17771644#comment-17771644
 ] 

ASF subversion and git services commented on SOLR-17000:
--------------------------------------------------------

Commit 98056d4ee254c6a04daaa8ff8842e3cea254b60e in solr's branch 
refs/heads/branch_9x from Chris M. Hostetter
[ https://gitbox.apache.org/repos/asf?p=solr.git;h=98056d4ee25 ]

SOLR-17000: Fix ExecutorUtilTest to eliminate concurrency assumptions

(cherry picked from commit 70abd8db8980a91913f28771247a3c8418e2840e)


> ExecutorUtilTest failures due to bad concurrency assumptions in test logic
> --------------------------------------------------------------------------
>
>                 Key: SOLR-17000
>                 URL: https://issues.apache.org/jira/browse/SOLR-17000
>             Project: Solr
>          Issue Type: Test
>      Security Level: Public(Default Security Level. Issues are Public) 
>            Reporter: Chris M. Hostetter
>            Assignee: Chris M. Hostetter
>            Priority: Major
>         Attachments: SOLR-17000-1.patch, SOLR-17000.patch, 
> apache_solr_Solr-check-9.3_647.log.txt, apache_solr_Solr-check-9.3_665.log.txt
>
>
> The basic logic of {{ExecutorUtilTest.testExecutorUtilAwaitsTerminationEnds}} 
> in psuedo code is...
> {code:java}
> Future f = executorService.submit(newTaskSleepAndIgnoreInterupts(300ms))      
>  // L45
> executorService.shutdownNow();                                                
>  // L46
> assertThrows(RuntimeException,                                                
>  // L47
>              ExecutorUtil.awaitTermination(executorService, 100ms)
> // Thread should not have finished in await termination.                      
>  
> assertFalse(f.isDone());                                                      
>  // L53
> {code}
> There are at least two concurrency assumptions here that are not guaranteed 
> to be true, and occasionally cause jenkins failures...
>  * There is no guarantee that the task submitted on line 45 will start before 
> the {{shutdownNow()}} call on line 46 – which means {{awaitTermination()}} 
> can succeed w/o throwing an exception:
> {noformat}
> // apache_solr_Solr-check-9.3_665.log.txt
> org.apache.solr.common.util.ExecutorUtilTest > 
> testExecutorUtilAwaitsTerminationEnds FAILED
>     java.lang.AssertionError: expected java.lang.RuntimeException to be 
> thrown, but nothing was thrown
>         at 
> __randomizedtesting.SeedInfo.seed([7169EEE284A03087:DFB4E65167A857BF]:0)
>         at org.junit.Assert.assertThrows(Assert.java:1028)
>         at org.junit.Assert.assertThrows(Assert.java:981)
>         at 
> org.apache.solr.common.util.ExecutorUtilTest.testExecutorUtilAwaitsTerminationEnds(ExecutorUtilTest.java:47)
> {noformat}
>  * From the perspective of the test thread, it's guaranteed that _at least_ 
> 100ms has elapsed (since line 45) by the time {{awaitTermination()}} throws 
> it's exception, but that doesn't preclude the possibility that by the time 
> the test thread gets to line 53, a full 300ms may have elapsed, and the 
> background thread has already finished the task, so that the  
> {{Future.isDone()}} call may return true:
> {noformat}
> // apache_solr_Solr-check-9.3_647.log.txt
> org.apache.solr.common.util.ExecutorUtilTest > 
> testExecutorUtilAwaitsTerminationEnds FAILED
> java.lang.AssertionError
> at __randomizedtesting.SeedInfo.seed([F7AFD5785E583017:5972DDCBBD50572F]:0)
> at org.junit.Assert.fail(Assert.java:87)
> at org.junit.Assert.assertTrue(Assert.java:42)
> at org.junit.Assert.assertFalse(Assert.java:65)
> at org.junit.Assert.assertFalse(Assert.java:75)
> at 
> org.apache.solr.common.util.ExecutorUtilTest.testExecutorUtilAwaitsTerminationEnds(ExecutorUtilTest.java:53)
> {noformat}



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@solr.apache.org
For additional commands, e-mail: issues-h...@solr.apache.org

Reply via email to