[jbossts-issues] [JBoss JIRA] (JBTM-1611) Failing qa testcase org.jboss.jbossts.qa.junit.testgroup.TestGroup_crashrecovery09, 10, 12 with jacorb and HornetQ object store enabled

Michael Musgrove (JIRA) jira-events at lists.jboss.org
Fri Apr 26 15:15:54 EDT 2013

    [ https://issues.jboss.org/browse/JBTM-1611?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12770477#comment-12770477 ] 

Michael Musgrove commented on JBTM-1611:

I have been unable to replicate the issue on my laptop using the command:
ant -f run-tests.xml -Dtest.name=crashrecovery09 -Dtest.methods="CrashRecovery09_Test01" -Dprofile=hornetq onetest
and similarly for the other two failing methods in this test group and the other two test groups (crashrecover10 and crashrecover12). Therefore it looks like the issue is environmental.

I have just debugged one of the other QA crash recovery issues (JBTM-1601) and there I found that the client wasn't waiting long enough for recovery to kick in. Extending the wait period before checking for success fixed that issue but unfortunately these test groups uses a different method to wait for recovery. In this case the clients trigger a remote recovery scan. The internal interface does allow the client to trigger multiple passes but the first two test groups have the number of scans hardcoded to 1 and the last group does 2 passes so its not configurable. To test the theory we could either:

# run the test on a fast machine, or
# I patch the QA test suite so that scan counter is configurable and introduce a pause between scans

If either of these approaches work we can put the fix into 4.17.5 
> Failing qa testcase org.jboss.jbossts.qa.junit.testgroup.TestGroup_crashrecovery09,10,12 with jacorb and HornetQ object store enabled
> -------------------------------------------------------------------------------------------------------------------------------------
>                 Key: JBTM-1611
>                 URL: https://issues.jboss.org/browse/JBTM-1611
>             Project: JBoss Transaction Manager
>          Issue Type: Bug
>      Security Level: Public(Everyone can see) 
>          Components: Testing
>    Affects Versions: 4.17.3
>            Reporter: Ondřej Chaloupka
>            Assignee: Michael Musgrove
>             Fix For: 4.17.5
> During testing EAP 6.1.0.ER3 I'm hitting problem on QA tests of narayana project for branch 4.17 when running qa tests with jacorb and hornetq object store (happening for NIO and AIO).
> Tests which are failing:
> org.jboss.jbossts.qa.junit.testgroup.TestGroup_crashrecovery09
> - CrashRecovery09_Test01
> - CrashRecovery09_Test03
> - CrashRecovery09_Test05
> org.jboss.jbossts.qa.junit.testgroup.TestGroup_crashrecovery10
> - CrashRecovery10_Test01
> - CrashRecovery10_Test03
> - CrashRecovery10_Test05
> org.jboss.jbossts.qa.junit.testgroup.TestGroup_crashrecovery12
> - CrashRecovery12_Test03
> - CrashRecovery12_Test06
> Detailed log could be found in job:
> https://jenkins.mw.lab.eng.bos.redhat.com/hudson/job/eap-61-jbossts-qa-test-hqobjectstore-nio/13/artifact/archive.zip
> I was trying to get more info on the test CrashRecovery09 but I can say just that the value honlder has value 0 instead of 1 on org.jboss.jbossts.qa.CrashRecovery09Clients.Client01a (line 84).

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

More information about the jbossts-issues mailing list