Uploaded image for project: 'JBoss Enterprise Application Platform'
  1. JBoss Enterprise Application Platform
  2. JBEAP-29300

Messages lost after terminating the server with a remote JCA connection

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Done
    • Icon: Blocker Blocker
    • 8.1.0.GA-CR4, 8.1.0.GA
    • 8.1.0.Beta
    • ActiveMQ
    • None
    • False
    • Hide

      None

      Show
      None
    • False
    • Known Issue
    • Hide

      Steps to reproduce (issue is intermittent):

      git clone git@gitlab.cee.redhat.com:jbossqe-eap/messaging-testsuite.git messaging-testsuite
cd messaging-testsuite/scripts/


groovy -DEAP_ZIP_URL=<path_to_server_zip_file> PrepareServers7.groovy
export WORKSPACE=$PWD
export JBOSS_HOME_1=$WORKSPACE/server1/jboss-eap
export JBOSS_HOME_2=$WORKSPACE/server2/jboss-eap
export JBOSS_HOME_3=$WORKSPACE/server3/jboss-eap
export JBOSS_HOME_4=$WORKSPACE/server4/jboss-eap

cd ../jboss-hornetq-testsuite/
mvn --batch-mode clean test -Dtest=BytemanLodh2TestCase#testLodh3KillOnTransactionCommit -Dsurefire.failIfNoSpecifiedTests=false -Deap7.clients.version=8.1739126675-SNAPSHOT -Deap7.org.jboss.qa.hornetq.apps.clients.version=8.1739126675-SNAPSHOT | tee log
      Show
      Steps to reproduce (issue is intermittent): git clone git@gitlab.cee.redhat.com:jbossqe-eap/messaging-testsuite.git messaging-testsuite cd messaging-testsuite/scripts/ groovy -DEAP_ZIP_URL=<path_to_server_zip_file> PrepareServers7.groovy export WORKSPACE=$PWD export JBOSS_HOME_1=$WORKSPACE/server1/jboss-eap export JBOSS_HOME_2=$WORKSPACE/server2/jboss-eap export JBOSS_HOME_3=$WORKSPACE/server3/jboss-eap export JBOSS_HOME_4=$WORKSPACE/server4/jboss-eap cd ../jboss-hornetq-testsuite/ mvn --batch-mode clean test -Dtest=BytemanLodh2TestCase#testLodh3KillOnTransactionCommit -Dsurefire.failIfNoSpecifiedTests= false -Deap7.clients.version=8.1739126675-SNAPSHOT -Deap7.org.jboss.qa.hornetq.apps.clients.version=8.1739126675-SNAPSHOT | tee log

      There is intermittent failure in test scenario with with EAP servers forming Artemis cluster and EAP servers with deployed MDBs connected through remote JCA to this cluster.

      Customer impact: There may be lost messages in case of crash of EAP server in cluster with remote JCA connection.

      Test scenario:

      • There are 4 nodes. Node 1 and 3 are in Artemis cluster and started with queues InQueue and OutQueue on both of them.
      • Start producer which sends 20000 messages (mix of small and large messages) to InQueue. Once producer finishes, deploy MDBs to nodes 2, 4 which read messages from InQueue and sends them to OutQueue (in XA transaction).
      • Kill and start again node 1. Then kill and start node 3 on XA transaction commit (when org.apache.activemq.artemis.core.transaction.Transaction.commit() is called)
      • Read all messages from OutQueue.

      Expected result: Number of send and received messages is the same (20 000)
      Actual result: Sometimes there are lost messages.

      Investigation: I was tracking one of the lost messages (it's not large message) and see that it was processed on node-4 by MDB and sent to OutQueue to node-3 succesfully. From the traces it's visible that message was paged but it's not clear why it could not be read from OutQueue by receiver later on.

              ehugonne1@redhat.com Emmanuel Hugonnet
              mnovak1@redhat.com Miroslav Novak
              Miroslav Novak Miroslav Novak
              Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

                Created:
                Updated:
                Resolved: