[
https://issues.jboss.org/browse/ISPN-3819?page=com.atlassian.jira.plugin....
]
Gustavo Fernandes commented on ISPN-3819:
-----------------------------------------
Apparently there is a competition to register ScopedKeys into the Cluster Registry. The
issue can be easily reproducible 50% of the time by adding a Random wait in the class
QueryInterceptor at line 260, so that it is easier for 2 or more threads to try to insert
the same key in the ClusterRegisty at approximately the same time:
{code}
for (Class<?> type : classes) {
if (!clusterRegistry.containsKey(type)) {
try {
Thread.sleep(new Random().nextInt(500));
} catch (InterruptedException e) {
e.printStackTrace();
}
if (toAdd==null)
toAdd = new ArrayList<Class<?>>(classes.length);
toAdd.add(type);
}
}
{code}
Test
org.infinispan.query.blackbox.ClusteredCacheWithLongIndexNameTest.testAdditionOfNewNode
fails randomly on RHEL
-------------------------------------------------------------------------------------------------------------------
Key: ISPN-3819
URL:
https://issues.jboss.org/browse/ISPN-3819
Project: Infinispan
Issue Type: Bug
Security Level: Public(Everyone can see)
Components: Embedded Querying
Affects Versions: 6.0.0.Final
Reporter: Vitalii Chepeliuk
Assignee: Gustavo Fernandes
Labels: testsuite_stability
Description of failure please look in following jobs with different JDKs
https://jenkins.mw.lab.eng.bos.redhat.com/hudson/view/JDG/view/FUNC/job/e...
https://jenkins.mw.lab.eng.bos.redhat.com/hudson/view/JDG/view/FUNC/job/e...
https://jenkins.mw.lab.eng.bos.redhat.com/hudson/view/JDG/view/FUNC/job/e...
--
This message was sent by Atlassian JIRA
(v6.2.6#6264)