Loading...

Type: Bug
Resolution: Unresolved
Priority: Blocker
Fix Version/s: 38.0.1.Final
Affects Version/s: 38.0.0.Beta1
Component/s: JMS
Labels:
None

Steps to Reproduce:
Hide

Steps to reproduce:

git clone git@gitlab.cee.redhat.com:mnovak/messaging-testsuite.git cd messaging-testsuite/scripts/ git checkout missing-messages groovy -DEAP_ZIP_URL=<path-to-wildfly-server-zip> 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=$WORKSPACE/server4/jboss-eap cd ../jboss-hornetq-testsuite/ mvn clean install -Dtest=Lodh2TestCase#testSimpleLodh2killMdbLocalRemote -Deap7.org.jboss.qa.hornetq.apps.clients.version=8.1760015969-8.x-8524-202510090117-SNAPSHOT -Dversion.artemis=2.42.0 -Dsurefire.failIfNoSpecifiedTests=false
Show
Steps to reproduce: git clone git@gitlab.cee.redhat.com:mnovak/messaging-testsuite.git cd messaging-testsuite/scripts/ git checkout missing-messages groovy -DEAP_ZIP_URL=<path-to-wildfly-server-zip> PrepareServers7.groovy export WORKSPACE=$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 clean install -Dtest=Lodh2TestCase#testSimpleLodh2killMdbLocalRemote -Deap7.org.jboss.qa.hornetq.apps.clients.version=8.1760015969-8.x-8524-202510090117-SNAPSHOT -Dversion.artemis=2.42.0 -Dsurefire.failIfNoSpecifiedTests= false
[QE] How to address?:
---
[QE] Why QE missed?:
---

There can be unrecovered transactions on Artemis when remote server with MDB crashed and is restarted again.

Test Scenario:

Start Cluster A (Node 1 and Node 3) and deploy queue OutQueue.
Start Cluster B (Node 2 and Node 4) and deploy queue InQueue.
Start a producer that sends 1000 mixed-size (large and normal) messages to InQueue.
Wait for the producer to complete message sending.
Deploy MDBs on Node 2 and Node 4 which consume messages from InQueue (node 2 and 4) and send to remote OutQueue (node 1 and 3) in XA transaction.
Kill Node 2 while MDBs are actively processing messages.
Restart Node 2.
Wait until all messages have been processed. Provide enough time for recovery of XA transactions.
Start a consumer to consume all messages from OutQueue on node 1.

Expected result: All messages will be consumed from OutQueue from node 1.
Actual results: Some messages are missing. There are unrecovered Artemis transactions on nodes 1 and 3.

Investigation showed that there are some (Artemis) transactions in prepared state on node 1 or 3 (holding those messages). There are also warnings during periodic recovery on node 2 like:

12:01:00,641 WARN  [com.arjuna.ats.jta] (Periodic Recovery) ARJUNA016037: Could not find new XAResource to use for recovering non-serializable XAResource XAResourceRecord < resource:null, txid:< formatId=131077, gtrid_length=32, bqual_length=36, tx_uid=0:ffff7f000001:23447aae:68ee36b9:124d, node_name=5026, branch_uid
=0:ffff7f000001:23447aae:68ee36b9:1250, subordinatenodename=null, eis_name=java:/LocalJmsXA NodeId:7d43c294-a8f2-11f0-af76-fa163efd220b >, heuristic: TwoPhaseOutcome.FINISH_OK, product: ActiveMQ Artemis/2.42.0, jndiName: java:/LocalJmsXA NodeId:7d43c294-a8f2-11f0-af76-fa163efd220b com.arjuna.ats.internal.jta.resource
s.arjunacore.XAResourceRecord@498a716a >
12:01:00,642 WARN  [com.arjuna.ats.jta] (Periodic Recovery) ARJUNA016037: Could not find new XAResource to use for recovering non-serializable XAResource XAResourceRecord < resource:null, txid:< formatId=131077, gtrid_length=32, bqual_length=36, tx_uid=0:ffff7f000001:23447aae:68ee36b9:124d, node_name=5026, branch_uid
=0:ffff7f000001:23447aae:68ee36b9:12de, subordinatenodename=null, eis_name=java:/JmsXA NodeId:8374d7cc-a8f2-11f0-ae53-fa163efd220b >, heuristic: TwoPhaseOutcome.FINISH_OK, product: ActiveMQ Artemis/2.0, jndiName: java:/JmsXA NodeId:8374d7cc-a8f2-11f0-ae53-fa163efd220b com.arjuna.ats.internal.jta.resources.arjunacore.
XAResourceRecord@32a4244d >
12:01:00,642 WARN  [com.arjuna.ats.jta] (Periodic Recovery) ARJUNA016038: No XAResource to recover < formatId=131077, gtrid_length=32, bqual_length=36, tx_uid=0:ffff7f000001:23447aae:68ee36b9:124d, node_name=5026, branch_uid=0:ffff7f000001:23447aae:68ee36b9:1250, subordinatenodename=null, eis_name=java:/LocalJmsXA No
deId:7d43c294-a8f2-11f0-af76-fa163efd220b >
12:01:00,642 WARN  [com.arjuna.ats.jta] (Periodic Recovery) ARJUNA016038: No XAResource to recover < formatId=131077, gtrid_length=32, bqual_length=36, tx_uid=0:ffff7f000001:23447aae:68ee36b9:124d, node_name=5026, branch_uid=0:ffff7f000001:23447aae:68ee36b9:12de, subordinatenodename=null, eis_name=java:/JmsXA NodeId:8374d7cc-a8f2-11f0-ae53-fa163efd220b >

It indicates that recovery manager knows that there are unfinished transactions but cannot do anything as it's not able to find correct XAResource.

Details

Description

Attachments

Easy Agile Planning Poker

Activity

People

Dates