Bug 1089163

Summary: [QE] (6.2.x) Intermittent failures of HornetQBackupActivationTestCase
Product: [JBoss] JBoss Enterprise Application Platform 6 Reporter: Miroslav Novak <mnovak>
Component: JMSAssignee: Jeff Mesnil <jmesnil>
Status: CLOSED WONTFIX QA Contact: Miroslav Novak <mnovak>
Severity: medium Docs Contact: Russell Dickenson <rdickens>
Priority: unspecified    
Version: 6.2.2CC: jmesnil, pjelinek, rhatlapa
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
: 1136095 (view as bug list) Environment:
Last Closed: 2014-07-07 13:39:12 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1136095    

Description Miroslav Novak 2014-04-18 07:11:29 UTC 
Description of problem:

There is intermittent failure in HornetQBackupActivationTestCase.testBackupActivation. See [1] for more details. 

Test is failing on assert in:
HornetQBackupActivationTestCase.checkJMSTopic(HornetQBackupActivationTestCase.java:487)

It seems that this check is run before topic is bound to JNDI because in server.log this topic deployed:
&#27;[0m04:13:42,203 INFO  [org.jboss.as.messaging] (ServerService Thread Pool -- 22) JBAS011601: Bound messaging object to jndi name java:jboss/exported/jms/1107b76d-f9f3-4ac1-ae91-89efad61886c

[1] https://jenkins.mw.lab.eng.bos.redhat.com/hudson/view/EAP6/view/EAP-6.2.x-CP/view/EAP-6.2.x-CP-jobs/job/eap-62x-patched-as-testsuite-Solaris-matrix-OracleJDK6/35/BITS=-d32,jdk=java16_default,label_exp=solaris10%26%26x86/testReport/org.jboss.as.test.manualmode.messaging/HornetQBackupActivationTestCase/testBackupActivation/
Comment 1 Miroslav Novak 2014-04-23 10:59:28 UTC 
Hi Jeff,

we're getting intermittent failure in HornetQBackupActivationTestCase. We're unsure whether it's test issue or this is real bug. 

From console log of backup server there is:
...
04:13:41,590 INFO  [org.hornetq.core.server] (ServerService Thread Pool -- 22) HQ221003: trying to deploy queue jms.topic.1107b76d-f9f3-4ac1-ae91-89efad61886c
...
04:13:41,690 INFO  [org.hornetq.core.server] (HQ119000: Activation for server HornetQServerImpl::serverUUID=339aa96c-b3f5-11e3-9a75-7f9e59dd90dc) HQ221010: Backup Server is now live
...
04:13:42,203 INFO  [org.jboss.as.messaging] (ServerService Thread Pool -- 22) JBAS011601: Bound messaging object to jndi name java:jboss/exported/jms/1107b76d-f9f3-4ac1-ae91-89efad61886c

so topic is bound to JNDI after "Backup Server is now live". 

This could cause that any new client connecting to backup before topic is bound could fail to lookup this topic. WDYT?

Thanks,
Mirek
Comment 2 Jeff Mesnil 2014-04-24 12:24:36 UTC 
there is indeed a timeframe when the backup server becomes live but the JMS topic is not already bound.

The root issue is that HornetQ only notifies WildFly when the hornetq core server becomes live but at this points the JMS resources have not been loaded. See https://issues.jboss.org/browse/HORNETQ-1180

For the tests, we can reduce that timeframe by adding a Thread.sleep after the backup server becomes live and before checking whether the JMS topic is bound
Comment 3 Miroslav Novak 2014-07-07 13:39:12 UTC 
No new CP release is planned for EAP 6.2.x and this issue won't be fixed in it. Closing as won't fix. 

Work on this for EAP 6.4 will be tracked on bz#970243.