[
https://issues.apache.org/jira/browse/ARTEMIS-4527?focusedWorklogId=901526&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-901526
]
ASF GitHub Bot logged work on ARTEMIS-4527:
-------------------------------------------
Author: ASF GitHub Bot
Created on: 24/Jan/24 19:26
Start Date: 24/Jan/24 19:26
Worklog Time Spent: 10m
Work Description: AntonRoskvist commented on code in PR #4705:
URL: https://github.com/apache/activemq-artemis/pull/4705#discussion_r1465438976
##########
tests/integration-tests/src/test/java/org/apache/activemq/artemis/tests/integration/cluster/distribution/ClusterNotificationsTest.java:
##########
@@ -0,0 +1,124 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ * <p>
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * <p>
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.activemq.artemis.tests.integration.cluster.distribution;
+
+import javax.jms.Connection;
+import javax.jms.MessageListener;
+import javax.jms.Queue;
+import javax.jms.Session;
+import java.util.ArrayList;
+import java.util.List;
+import java.util.concurrent.CountDownLatch;
+import java.util.concurrent.ExecutorService;
+import java.util.concurrent.Executors;
+import java.util.concurrent.TimeUnit;
+
+import
org.apache.activemq.artemis.core.server.cluster.impl.MessageLoadBalancingType;
+import org.apache.activemq.artemis.core.settings.impl.AddressSettings;
+import org.apache.activemq.artemis.jms.client.ActiveMQConnectionFactory;
+import org.apache.activemq.artemis.jms.client.ActiveMQDestination;
+import org.apache.activemq.artemis.logs.AssertionLoggerHandler;
+import org.junit.Test;
+import org.springframework.jms.connection.CachingConnectionFactory;
+import org.springframework.jms.listener.AbstractJmsListeningContainer;
+import org.springframework.jms.listener.DefaultMessageListenerContainer;
+import org.springframework.util.backoff.FixedBackOff;
+
+public class ClusterNotificationsTest extends ClusterTestBase {
+
+ protected boolean isNetty() {
+ return true;
+ }
+
+ @Test
+ public void testClusterNotificationsContinuity() throws Exception {
+ final MessageLoadBalancingType lbType =
MessageLoadBalancingType.OFF_WITH_REDISTRIBUTION;
+ final String namePrefix = "TEST.QUEUE.";
+ final int queueCount = 500;
+
+ AssertionLoggerHandler loggerHandler = new AssertionLoggerHandler();
+ runAfter(loggerHandler::close);
+
+ setupServer(0, isFileStorage(), isNetty());
+ setupServer(1, isFileStorage(), isNetty());
+ setupServer(2, isFileStorage(), isNetty());
+
+ setupClusterConnection("cluster0", "", lbType, 1, isNetty(), 0, 1, 2);
+ setupClusterConnection("cluster0", "", lbType, 1, isNetty(), 1, 2, 0);
+ setupClusterConnection("cluster0", "", lbType, 1, isNetty(), 2, 0, 1);
+
+ startServers(0, 1, 2);
+
+ AddressSettings as = new AddressSettings().setRedistributionDelay(0);
+ servers[0].getAddressSettingsRepository().addMatch(namePrefix + "#", as);
+ servers[1].getAddressSettingsRepository().addMatch(namePrefix + "#", as);
+ servers[2].getAddressSettingsRepository().addMatch(namePrefix + "#", as);
+
+ waitForTopology(servers[2], 3);
+
+ ExecutorService executorService = Executors.newCachedThreadPool();
+ CountDownLatch latch = new CountDownLatch(queueCount * 2);
+
+ try (ActiveMQConnectionFactory factory = new
ActiveMQConnectionFactory("tcp://localhost:61616")) {
+
+ List<DefaultMessageListenerContainer> containers = new ArrayList<>();
+ runAfter(() ->
containers.parallelStream().forEach(AbstractJmsListeningContainer::shutdown));
+ runAfter(() -> executorService.shutdown());
+
+ for (int i = 0; i < queueCount; i++) {
+ Queue queue = ActiveMQDestination.createQueue(namePrefix + i +
"-");
+
+ executorService.submit(() -> {
+ try (Connection connection = factory.createConnection();
+ Session session =
connection.createSession(Session.SESSION_TRANSACTED)) {
+
+
session.createProducer(queue).send(session.createTextMessage("Message"));
+ session.commit();
+ latch.countDown();
+ } catch (Exception e) { }
+ });
+
+ DefaultMessageListenerContainer container = new
DefaultMessageListenerContainer();
+ container.setCacheLevelName("CACHE_NONE");
+ container.setSessionTransacted(true);
+ container.setSessionAcknowledgeModeName("SESSION_TRANSACTED");
+ container.setConcurrentConsumers(20);
+ container.setConnectionFactory(new
CachingConnectionFactory(factory));
+ container.setDestinationName(queue.getQueueName());
+ container.setBackOff(new FixedBackOff(1000, 2));
+ container.setReceiveTimeout(100);
+ container.setMessageListener((MessageListener) msg -> {
+ latch.countDown();
+ });
+
+ container.initialize();
Review Comment:
The problem here is (I guess) that in order for the test to trigger the
issue this PR is about, underlying hardware most likely has to be working close
to capacity. (This is not the case with an "actual" cluster though for some
reason).
If you are interested in just trying this out I'd advice lowering queue
count and consumer count on the Spring DMLC (while I see if I can find some
formula to scale this based on available system resources or something).
Change would be:
:52 final int queueCount = 500; --> 100?
:100 container.setConcurrentConsumers(20); --> 5?
Issue Time Tracking
-------------------
Worklog Id: (was: 901526)
Time Spent: 2h 50m (was: 2h 40m)
> Redistributor race when consumerCount reaches 0 in cluster
> ----------------------------------------------------------
>
> Key: ARTEMIS-4527
> URL: https://issues.apache.org/jira/browse/ARTEMIS-4527
> Project: ActiveMQ Artemis
> Issue Type: Bug
> Reporter: Anton Roskvist
> Priority: Major
> Time Spent: 2h 50m
> Remaining Estimate: 0h
>
> This is a very rare bug caused by cluster notifications arriving in the wrong
> order in some very specific circumstances
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
