[ 
https://issues.apache.org/jira/browse/SUREFIRE-1487?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Tibor Digana updated SUREFIRE-1487:
-----------------------------------
    Fix Version/s: 3.0.0-M1

> ParallelComputerBuilderTest fails on overloaded system because internal delay 
> are shorter than blocking time of JVM
> -------------------------------------------------------------------------------------------------------------------
>
>                 Key: SUREFIRE-1487
>                 URL: https://issues.apache.org/jira/browse/SUREFIRE-1487
>             Project: Maven Surefire
>          Issue Type: Task
>          Components: Junit 4.7+ (parallel) support
>            Reporter: Tibor Digana
>            Priority: Major
>             Fix For: 3.0.0-M1
>
>
> The delays are 500 ms and variations 250 ms.
> The JVM was blocked for 675 ms.
> From these measurements, the new delay should be 3-times longer at least.
> [windows] Tests in error: 
> [windows]   
> separatePoolsWithSuiteAndSequentialClasses(org.apache.maven.surefire.junitcore.pc.ParallelComputerBuilderTest):
>  (..)
> [windows]   
> parallelMethodsReuseOneOrTwoThreads(org.apache.maven.surefire.junitcore.pc.ParallelComputerBuilderTest):
>  (..)
> [windows]   
> suiteAndClassInOnePool(org.apache.maven.surefire.junitcore.pc.ParallelComputerBuilderTest):
>  (..)
> suiteAndClassInOnePool:
> Expected: (between <1450L> and <1750L> or between <1950L> and <2250L> or 
> between <2450L> and <2750L>)
>       but: was <3175L>
> parallelMethodsReuseOneOrTwoThreads:
> Expected: between <1450L> and <1750L>
>       but: was <1837L>
> separatePoolsWithSuiteAndSequentialClasses:
> Expected: between <1450L> and <1750L>
>       but: was <3562L>
> In this case the JVM was blocked for 2.062 second and the delays should be 9 
> times longer. The test {{ParallelComputerBuilderTest}} has small number of 
> tests, so we can prolong the delays 10 times.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to