[jboss-jira] [JBoss JIRA] (DROOLS-2592) PMML - tests in class ScorecardReasonCodeTest fail randomly

[Tibor Zimányi (JIRA)]

    [ https://issues.jboss.org/browse/DROOLS-2592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13631858#comment-13631858 ] 

Tibor Zimányi commented on DROOLS-2592:
---------------------------------------

After debugging, it is a race condition caused by "after" construct used in [1]. It happens sometimes that the execution is so fast, that the timestamps of both compared facts are the same. If I use after[0] with no-loop on the rule, it fixes the issue. I will create a PR with the fix. 

[1] https://github.com/kiegroup/drools/blob/133890cbdcdcc0dd9ea4f3a775fcae81aabdb1c2/kie-pmml/src/main/resources/org/kie/pmml/pmml_4_2/templates/models/scorecard/scorecardOutputRankCode.drlt#L74

> PMML - tests in class ScorecardReasonCodeTest fail randomly
> -----------------------------------------------------------
>
>                 Key: DROOLS-2592
>                 URL: https://issues.jboss.org/browse/DROOLS-2592
>             Project: Drools
>          Issue Type: Bug
>          Components: core engine
>    Affects Versions: 7.7.0.Final
>            Reporter: Tibor Zimányi
>            Assignee: Tibor Zimányi
>            Priority: Critical
>              Labels: drools-core
>         Attachments: output.zip, results_bad.zip, results_good.zip, test.sh
>
>
> When the ScorecardReasonCodeTest.testPointsBelow is run repeatedly, it sometimes fail with following assertion fail [1]. Setting this as critical, because the random fail causes build problems in the CI environment. Same applies for testPointsAbove, testReasonCodesCombination and probably others (I tested those 3 mentioned). 
> [1] https://gist.github.com/baldimir/4c8ac4563fd17d13e4646fd60dd3a357



--
This message was sent by Atlassian JIRA
(v7.5.0#75005)