[JBoss JIRA] (ISPN-2645) LocalKeyAffinityServiceTest.testFilteredRemoveServers fails randomly
by Dan Berindei (JIRA)
Dan Berindei created ISPN-2645:
----------------------------------
Summary: LocalKeyAffinityServiceTest.testFilteredRemoveServers fails randomly
Key: ISPN-2645
URL: https://issues.jboss.org/browse/ISPN-2645
Project: Infinispan
Issue Type: Bug
Components: Distributed Cache, Test Suite
Affects Versions: 5.2.0.Beta5
Reporter: Dan Berindei
Assignee: Mircea Markus
Fix For: 6.0.0.Final
LocalKeyAffinityServiceTest tests that KeyAffinnityServiceImpl's internal queues eventually fill up, and it fails:
{noformat}
java.lang.AssertionError: expected [true] but found [false]
at org.testng.Assert.fail(Assert.java:89)
at org.testng.Assert.failNotEquals(Assert.java:489)
at org.testng.Assert.assertTrue(Assert.java:37)
at org.testng.Assert.assertTrue(Assert.java:47)
at org.infinispan.test.AbstractInfinispanTest.eventually(AbstractInfinispanTest.java:76)
at org.infinispan.test.AbstractInfinispanTest.eventually(AbstractInfinispanTest.java:59)
at org.infinispan.affinity.BaseKeyAffinityServiceTest.assertEventualFullCapacity(BaseKeyAffinityServiceTest.java:98)
at org.infinispan.affinity.BaseFilterKeyAffinityServiceTest.assertUnaffected(BaseFilterKeyAffinityServiceTest.java:108)
at org.infinispan.affinity.BaseFilterKeyAffinityServiceTest.testRemoveServers(BaseFilterKeyAffinityServiceTest.java:82)
at org.infinispan.affinity.LocalKeyAffinityServiceTest.testFilteredRemoveServers(LocalKeyAffinityServiceTest.java:66)
{noformat}
I have added some TRACE logs to KeyAffinityServiceImpl and it looks like the queue fills up immediately, but it could be that a new queue is created and the test is checking the wrong queue object.
--
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
12 years
[JBoss JIRA] (ISPN-2629) Dist Exec testsuite fails in case of TopologyAware nodes.
by Vladimir Blagojevic (JIRA)
[ https://issues.jboss.org/browse/ISPN-2629?page=com.atlassian.jira.plugin.... ]
Vladimir Blagojevic updated ISPN-2629:
--------------------------------------
Fix Version/s: 5.2.0.Beta6
> Dist Exec testsuite fails in case of TopologyAware nodes.
> ---------------------------------------------------------
>
> Key: ISPN-2629
> URL: https://issues.jboss.org/browse/ISPN-2629
> Project: Infinispan
> Issue Type: Bug
> Components: Distributed Execution and Map/Reduce
> Reporter: Anna Manukyan
> Assignee: Vladimir Blagojevic
> Fix For: 5.2.0.Beta6
>
> Attachments: DistributedExecutorWithTopologyAwareNodesTest.java, TEST-org.infinispan.distexec.DistributedExecutorWithTopologyAwareNodesTest.xml
>
>
> Many tests from dist exec test suite fail in case of TopologyAware nodes.
> It has been found that with both types of Cache configuration (old and new API), the issue appears.
> You can find the test and the stacktraces attached.
--
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
12 years
[JBoss JIRA] (ISPN-2629) Dist Exec testsuite fails in case of TopologyAware nodes.
by Vladimir Blagojevic (JIRA)
[ https://issues.jboss.org/browse/ISPN-2629?page=com.atlassian.jira.plugin.... ]
Vladimir Blagojevic commented on ISPN-2629:
-------------------------------------------
See pull for diff. Essentially the problem was that we did not update cancellation tests to use RpcManager#getMembers but instead used old Transport#getMembers
> Dist Exec testsuite fails in case of TopologyAware nodes.
> ---------------------------------------------------------
>
> Key: ISPN-2629
> URL: https://issues.jboss.org/browse/ISPN-2629
> Project: Infinispan
> Issue Type: Bug
> Components: Distributed Execution and Map/Reduce
> Reporter: Anna Manukyan
> Assignee: Vladimir Blagojevic
> Attachments: DistributedExecutorWithTopologyAwareNodesTest.java, TEST-org.infinispan.distexec.DistributedExecutorWithTopologyAwareNodesTest.xml
>
>
> Many tests from dist exec test suite fail in case of TopologyAware nodes.
> It has been found that with both types of Cache configuration (old and new API), the issue appears.
> You can find the test and the stacktraces attached.
--
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
12 years
[JBoss JIRA] (ISPN-2603) Intermittent test failure: AsyncAPITest
by Mircea Markus (JIRA)
[ https://issues.jboss.org/browse/ISPN-2603?page=com.atlassian.jira.plugin.... ]
Mircea Markus resolved ISPN-2603.
---------------------------------
Resolution: Done
Fixed as part of ISPN-2552
> Intermittent test failure: AsyncAPITest
> ---------------------------------------
>
> Key: ISPN-2603
> URL: https://issues.jboss.org/browse/ISPN-2603
> Project: Infinispan
> Issue Type: Feature Request
> Affects Versions: 5.2.0.Beta5
> Reporter: Mircea Markus
> Assignee: Mircea Markus
> Labels: test
> Fix For: 5.2.0.Beta6
>
>
> AsyncAPITest.testAsyncMethodWithLifespanAndMaxIdle has this logic:
> {code:java}
> // putIfAbsent lifespan only
> f = c.putAsync("k", "v3");
> assertNull(f.get());
> f = c.putIfAbsentAsync("k", "v4", 1000, TimeUnit.MILLISECONDS);
> markStartTime();
> assert f != null;
> assert !f.isCancelled();
> assertEquals("v3", f.get()); //here's the problem!!
> {code}
> the test assumes that the first put(v3_ executes before the second one (v4) which is not correct.
--
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
12 years
[JBoss JIRA] (ISPN-2641) Use RpcManager#getMembers instead of Transport#getMembers if necessary
by Dan Berindei (JIRA)
[ https://issues.jboss.org/browse/ISPN-2641?page=com.atlassian.jira.plugin.... ]
Dan Berindei commented on ISPN-2641:
------------------------------------
We should also investigate how the primary lock owner is computed for replicated caches - the coordinator may not be a member.
> Use RpcManager#getMembers instead of Transport#getMembers if necessary
> ----------------------------------------------------------------------
>
> Key: ISPN-2641
> URL: https://issues.jboss.org/browse/ISPN-2641
> Project: Infinispan
> Issue Type: Bug
> Components: Distributed Cache
> Affects Versions: 5.2.0.Beta5
> Reporter: Vladimir Blagojevic
> Assignee: Mircea Markus
> Fix For: 5.2.0.Final
>
>
> Recently we have added RpcManager#getMembers method which returns cache members rather that entire cluster members returned by Transport#getMembers. We should review the entire codebase very carefully and replace instances of Transport#getMembers with RpcManager#getMembers if necessary.
--
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
12 years