-
Bug
-
Resolution: Done
-
Critical
-
None
Scenario:
The error shows up both with JDK-11 and JDK-17
There are three servers. Live server (Node 1) and backup server (Node 2) are in dedicated HA topology. InQueue and OutQueue are deployed on live and backup:
- Send messages to InQueue on live server
- When all messages are sent, deploy message driven bean on Node 3
- MDB sends messages form InQueue to OutQueue
- Shutdown live server
- Receive messages from OutQueue from backup server
After all this happened, transactions remain in prepared state on Node 2;
The complete error message is:
[ERROR] org.jboss.qa.hornetq.test.failover.ReplicatedDedicatedFailoverTestWithMdb.testShutdown Time elapsed: 605.326 s <<< FAILURE! java.lang.AssertionError: There are prepared transactions in HornetQ/Artemis journal - number of prepared transactions is: 2 at org.junit.Assert.fail(Assert.java:89) at org.jboss.qa.hornetq.tools.TransactionUtils.waitUntilThereAreNoPreparedHornetQTransactions(TransactionUtils.java:63) at org.jboss.qa.hornetq.tools.TransactionUtils.waitUntilThereAreNoPreparedHornetQTransactions(TransactionUtils.java:81) at org.jboss.qa.hornetq.test.failover.DedicatedFailoverTestCaseWithMdb.testFailoverWithRemoteJca(DedicatedFailoverTestCaseWithMdb.java:395) at org.jboss.qa.hornetq.test.failover.DedicatedFailoverTestCaseWithMdb.testShutdown(DedicatedFailoverTestCaseWithMdb.java:323)
- clones
-
WFLY-17362 Messaging - Transaction remained in prepared state after failover
- Closed
- is incorporated by
-
JBEAP-24383 [GSS](7.4.z) Upgrade artemis-wildfly-integration from 1.0.4 to 1.0.7
- Closed