You are viewing a plain text version of this content. The canonical link for it is here.
Posted to notifications@geode.apache.org by GitBox <gi...@apache.org> on 2021/06/25 13:07:58 UTC
[GitHub] [geode] jvarenina opened a new pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
jvarenina opened a new pull request #6646:
URL: https://github.com/apache/geode/pull/6646
Issue reproduces when following conditions are fulfilled:
- Redundant partition region must be configured
- Number of servers must be greater than number of redundant copies of partition region
- Parallel gateway sender must be configured on partition region
- Client must register CQs for the region
- Transactions must be used with put operations
- Events must be enqueued in parallel gateway senders (remote site is unavailable)
Server that is hosting primary bucket will send TXCommitMessage to the secondary
server, and also to the server that is hosting CQ subscription queue (if CQ condition is fulfilled).
The problem occurs when the server that is hosting CQ subscription queue does not host
the bucket for which event is received. In that case the server will store the event in
bucketToTempQueueMap, because it assumes that the bucket is in the process of the
creation, which is not correct.
The solution:
At reception of CommitProcessForTXIdMessage the events received with TXCommitMessage will not
be stored in temporary queue, if targeted bucket region is not hosted in the server. Following
checks are performed on receiving server in order to identify those events:
1. Check that targeted bucket region in not available locally on server
2. Check from filterRoutingInfo that the receiving server hosts CQ/RegionInterests queue for that event
If above checks are fulfilled then tailKey is set to -1 on receiving server to avoid enqueuing of
event in temporary queue.
Thank you for submitting a contribution to Apache Geode.
In order to streamline the review of the contribution we ask you
to ensure the following steps have been taken:
### For all changes:
- [x] Is there a JIRA ticket associated with this PR? Is it referenced in the commit message?
- [x] Has your PR been rebased against the latest commit within the target branch (typically `develop`)?
- [x] Is your initial contribution a single, squashed commit?
- [x] Does `gradlew build` run cleanly?
- [x] Have you written or updated unit tests to verify your changes?
- [ ] If adding new dependencies to the code, are these dependencies licensed in a way that is compatible for inclusion under [ASF 2.0](http://www.apache.org/legal/resolved.html#category-a)?
### Note:
Please ensure that once the PR is submitted, check Concourse for build issues and
submit an update to your PR as soon as possible. If you need help, please send an
email to dev@geode.apache.org.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] pivotal-eshu commented on a change in pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
pivotal-eshu commented on a change in pull request #6646:
URL: https://github.com/apache/geode/pull/6646#discussion_r659076400
##########
File path: geode-core/src/main/java/org/apache/geode/internal/cache/TXCommitMessage.java
##########
@@ -1301,14 +1301,22 @@ protected void txApplyEntryOp(FarSideEntryOp entryOp, List<EntryEventImpl> pendi
/*
* This happens when we don't have the bucket and are getting adjunct notification
*/
+ long tailKey = entryOp.tailKey;
+ if (entryOp.filterRoutingInfo != null) {
+ if (entryOp.filterRoutingInfo.getMembers()
+ .contains(this.internalRegion.getDistributionManager().getId())) {
Review comment:
can just use internalRegion.getMyId() instead here.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] jvarenina commented on a change in pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
jvarenina commented on a change in pull request #6646:
URL: https://github.com/apache/geode/pull/6646#discussion_r661235056
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ // check that bucketToTempQueueMap is empty on all members
+ for (MemberVM member : allMembers) {
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ }
+
+ // Choose server with bucket to be stopped
+ MemberVM serverToStop = getServerToStop(allMembers);
+ int bucketId = serverToStop.invoke(this::getPrimaryBucketList).iterator().next();
+ serverToStop.stop(true);
+
+ // remove from list the member that has been previously stopped
+ allMembers.remove(serverToStop);
+
+ startDurableClient();
+ createDurableCQs("SELECT * FROM " + Region.SEPARATOR + REGION_NAME);
+
+ // configure hook on running members
+ for (MemberVM member : allMembers) {
+ configureHooksOnRunningMember(member, bucketId);
+ }
+
+ // perform rebalance operation to trigger bucket redundancy recovery on running servers
+ String command = new CommandStringBuilder(CliStrings.REBALANCE)
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ for (MemberVM member : allMembers) {
+ // All members after redundancy is recovered should have empty temporary queue
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ // If hook has been triggered on member, then check if member temporarily queued
+ // events while getting initial image from primary server
+ if (member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isHookTriggered)) {
+ assertThat(member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isTempQueueUsed))
+ .isTrue();
+ }
+ }
+
+ // reset all hooks
+ for (MemberVM member : allMembers) {
+ resetHooks(member);
+ }
+ }
+
+ /**
+ * This hook will run some transaction traffic after server requested initial image from
+ * primary bucket server, and then it will check if events are enqueued in temporary queue.
+ * It is expected that events are enqueued in this case because bucket is not yet available.
+ */
+ void configureHooksOnRunningMember(MemberVM server, int bucketId) {
+ server.invoke(() -> InitialImageOperation.setGIITestHook(new InitialImageOperation.GIITestHook(
+ InitialImageOperation.GIITestHookType.BeforeRequestRVV,
+ "_B__" + REGION_NAME + "_" + bucketId) {
+ private static final long serialVersionUID = -3790198435185240444L;
+
+ @Override
+ public void reset() {}
+
+ @Override
+ public void run() {
+ doPutsInServer();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED = true;
+ if (sizeOfBucketToTempQueueMap("ln") != 0) {
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED = true;
+ }
+ }
+ }));
+ }
+
+ void resetHooks(MemberVM server) {
+ server.invoke(() -> InitialImageOperation.resetAllGIITestHooks());
+ }
+
+ public static boolean isHookTriggered() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED;
+ }
+
+ public static boolean isTempQueueUsed() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED;
+ }
+
+ MemberVM getServerToStop(List<MemberVM> list) {
+ for (MemberVM member : list) {
+ if (!member.invoke(this::getPrimaryBucketList).isEmpty()) {
+ return member;
+ }
+ }
+ return null;
+ }
+
+ private void startDurableClient()
+ throws Exception {
+ int locatorPort = locatorSite2.getPort();
+ clientSite2DurableSubscription = clusterStartupRule.startClientVM(7, ccf -> ccf
+ .withPoolSubscription(true).withLocatorConnection(locatorPort).withCacheSetup(c -> c
+ .set("durable-client-id", DURABLE_CLIENT_ID)));
+ }
+
+ private void createDurableCQs(String... queries) {
+ clientSite2DurableSubscription.invoke(() -> {
+ assertThat(ClusterStartupRule.getClientCache()).isNotNull();
+ QueryService queryService = ClusterStartupRule.getClientCache().getQueryService();
+ CqAttributesFactory cqAttributesFactory = new CqAttributesFactory();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.cqListener =
+ new CqListenerTestReceivedEvents();
+ cqAttributesFactory.addCqListener(cqListener);
+
+ for (String query : queries) {
+ CqQuery cq = queryService.newCq(query, cqAttributesFactory.create(), true);
+ cq.execute();
+ }
+ ClusterStartupRule.getClientCache().readyForEvents();
+ });
+ }
+
+ private Set<Integer> getPrimaryBucketList() {
+ PartitionedRegion region = (PartitionedRegion) Objects
+ .requireNonNull(ClusterStartupRule.getCache()).getRegion(REGION_NAME);
+ return new HashSet<>(region.getDataStore().getAllLocalPrimaryBucketIds());
+ }
+
+ /**
+ * Check that all events are received on durable client that registered cq's
+ */
+ private void checkCqEvents(int expectedNumberOfEvents) {
+ // Check if number of events is correct
+ clientSite2DurableSubscription.invoke(() -> await().untilAsserted(() -> assertThat(
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.cqListener.getNumEvents())
+ .isEqualTo(expectedNumberOfEvents)));
+ }
+
+ /**
+ * Checks that the bucketToTempQueueMap for a partitioned region
+ * that holds events for buckets that are not available locally, is empty.
+ */
+ public static void validateBucketToTempQueueMap(String senderId, boolean shouldBeEmpty) {
+ final int finalSize = sizeOfBucketToTempQueueMap(senderId);
+ if (shouldBeEmpty) {
+ assertThat(finalSize).isEqualTo(0);
+ } else {
+ assertThat(finalSize).isNotEqualTo(0);
+ }
+ }
+
+ public static int sizeOfBucketToTempQueueMap(String senderId) {
+ GatewaySender sender = getGatewaySender(senderId);
+ int size = 0;
+ Set<RegionQueue> queues = ((AbstractGatewaySender) sender).getQueues();
+ for (RegionQueue queue : queues) {
+ PartitionedRegion region =
+ (PartitionedRegion) queue.getRegion();
+ int buckets = region.getTotalNumberOfBuckets();
+ for (int bucket = 0; bucket < buckets; bucket++) {
+ BlockingQueue<GatewaySenderEventImpl> newQueue =
+ ((ConcurrentParallelGatewaySenderQueue) queue).getBucketTmpQueue(bucket);
+ if (newQueue != null) {
+ size += newQueue.size();
+ }
+ }
+ }
+ return size;
+ }
+
+ private static GatewaySender getGatewaySender(String senderId) {
+ Set<GatewaySender> senders =
+ Objects.requireNonNull(ClusterStartupRule.getCache()).getGatewaySenders();
+ GatewaySender sender = null;
+ for (GatewaySender s : senders) {
+ if (s.getId().equals(senderId)) {
+ sender = s;
+ break;
+ }
+ }
+ return sender;
+ }
+
+ void configureSites(String totalBucketNum, boolean addColocatedRegion)
+ throws Exception {
+
+ Properties props = new Properties();
+ props.setProperty(DISTRIBUTED_SYSTEM_ID, DISTRIBUTED_SYSTEM_ID_SITE1);
+ MemberVM locatorSite1 = clusterStartupRule.startLocatorVM(1, props);
+
+ props.setProperty(DISTRIBUTED_SYSTEM_ID, DISTRIBUTED_SYSTEM_ID_SITE2);
+ props.setProperty(REMOTE_LOCATORS, "localhost[" + locatorSite1.getPort() + "]");
+ locatorSite2 = clusterStartupRule.startLocatorVM(2, props);
+
+ // start servers for site #2
+ server1Site2 = clusterStartupRule.startServerVM(3, locatorSite2.getPort());
+ server2Site2 = clusterStartupRule.startServerVM(4, locatorSite2.getPort());
+ server3Site2 = clusterStartupRule.startServerVM(5, locatorSite2.getPort());
+
+ // create parallel gateway-sender on site #2
+ connectGfshToSite(locatorSite2);
+ String command = new CommandStringBuilder(CliStrings.CREATE_GATEWAYSENDER)
+ .addOption(CliStrings.CREATE_GATEWAYSENDER__ID, "ln")
+ .addOption(CliStrings.CREATE_GATEWAYSENDER__REMOTEDISTRIBUTEDSYSTEMID, "1")
+ .addOption(CliStrings.CREATE_GATEWAYSENDER__PARALLEL, "true")
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ server1Site2.invoke(() -> verifySenderState("ln", true, false));
+ server2Site2.invoke(() -> verifySenderState("ln", true, false));
+ server3Site2.invoke(() -> verifySenderState("ln", true, false));
+
+ locatorSite2.invoke(
+ () -> validateGatewaySenderMXBeanProxy(getMember(server1Site2.getVM()), "ln", true, false));
+ locatorSite2.invoke(
+ () -> validateGatewaySenderMXBeanProxy(getMember(server2Site2.getVM()), "ln", true, false));
+ locatorSite2.invoke(
+ () -> validateGatewaySenderMXBeanProxy(getMember(server3Site2.getVM()), "ln", true, false));
+
+ // create partition region on site #2
+ CommandStringBuilder csb = new CommandStringBuilder(CliStrings.CREATE_REGION);
+ csb.addOption(CliStrings.CREATE_REGION__REGION, REGION_NAME);
+ csb.addOption(CliStrings.CREATE_REGION__REGIONSHORTCUT, "PARTITION");
+ csb.addOption(CliStrings.CREATE_REGION__GATEWAYSENDERID, "ln");
+ csb.addOption(CliStrings.CREATE_REGION__REDUNDANTCOPIES, "1");
+ csb.addOption(CliStrings.CREATE_REGION__TOTALNUMBUCKETS, totalBucketNum);
+ gfsh.executeAndAssertThat(csb.toString()).statusIsSuccess();
+
+ if (addColocatedRegion) {
+ csb = new CommandStringBuilder(CliStrings.CREATE_REGION);
+ csb.addOption(CliStrings.CREATE_REGION__REGION, REGION_COLOCATED);
+ csb.addOption(CliStrings.CREATE_REGION__REGIONSHORTCUT, "PARTITION");
+ csb.addOption(CliStrings.CREATE_REGION__GATEWAYSENDERID, "ln");
+ csb.addOption(CliStrings.CREATE_REGION__REDUNDANTCOPIES, "1");
+ csb.addOption(CliStrings.CREATE_REGION__TOTALNUMBUCKETS, totalBucketNum);
+ csb.addOption(CliStrings.CREATE_REGION__COLOCATEDWITH, REGION_NAME);
+ gfsh.executeAndAssertThat(csb.toString()).statusIsSuccess();
+ }
+
+ // Start client
+ clientSite2 =
+ clusterStartupRule.startClientVM(6, c -> c.withLocatorConnection(locatorSite2.getPort()));
+ clientSite2.invoke(() -> {
+ ClusterStartupRule.clientCacheRule.createProxyRegion(REGION_NAME);
+ if (addColocatedRegion) {
+ ClusterStartupRule.clientCacheRule.createProxyRegion(REGION_COLOCATED);
+ }
+ });
+ }
+
+ void connectGfshToSite(MemberVM locator) throws Exception {
+ if (gfsh.isConnected()) {
+ gfsh.disconnect();
+ }
+ gfsh.connectAndVerify(locator);
+ }
+
+ void verifyGatewaySenderState(boolean isRunning, boolean isPaused) {
Review comment:
Done.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] jvarenina commented on a change in pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
jvarenina commented on a change in pull request #6646:
URL: https://github.com/apache/geode/pull/6646#discussion_r661235177
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ // check that bucketToTempQueueMap is empty on all members
+ for (MemberVM member : allMembers) {
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ }
+
+ // Choose server with bucket to be stopped
+ MemberVM serverToStop = getServerToStop(allMembers);
+ int bucketId = serverToStop.invoke(this::getPrimaryBucketList).iterator().next();
+ serverToStop.stop(true);
+
+ // remove from list the member that has been previously stopped
+ allMembers.remove(serverToStop);
+
+ startDurableClient();
+ createDurableCQs("SELECT * FROM " + Region.SEPARATOR + REGION_NAME);
+
+ // configure hook on running members
+ for (MemberVM member : allMembers) {
+ configureHooksOnRunningMember(member, bucketId);
+ }
+
+ // perform rebalance operation to trigger bucket redundancy recovery on running servers
+ String command = new CommandStringBuilder(CliStrings.REBALANCE)
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ for (MemberVM member : allMembers) {
+ // All members after redundancy is recovered should have empty temporary queue
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ // If hook has been triggered on member, then check if member temporarily queued
+ // events while getting initial image from primary server
+ if (member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isHookTriggered)) {
+ assertThat(member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isTempQueueUsed))
+ .isTrue();
+ }
+ }
+
+ // reset all hooks
+ for (MemberVM member : allMembers) {
+ resetHooks(member);
+ }
+ }
+
+ /**
+ * This hook will run some transaction traffic after server requested initial image from
+ * primary bucket server, and then it will check if events are enqueued in temporary queue.
+ * It is expected that events are enqueued in this case because bucket is not yet available.
+ */
+ void configureHooksOnRunningMember(MemberVM server, int bucketId) {
+ server.invoke(() -> InitialImageOperation.setGIITestHook(new InitialImageOperation.GIITestHook(
+ InitialImageOperation.GIITestHookType.BeforeRequestRVV,
+ "_B__" + REGION_NAME + "_" + bucketId) {
+ private static final long serialVersionUID = -3790198435185240444L;
+
+ @Override
+ public void reset() {}
+
+ @Override
+ public void run() {
+ doPutsInServer();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED = true;
+ if (sizeOfBucketToTempQueueMap("ln") != 0) {
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED = true;
+ }
+ }
+ }));
+ }
+
+ void resetHooks(MemberVM server) {
+ server.invoke(() -> InitialImageOperation.resetAllGIITestHooks());
+ }
+
+ public static boolean isHookTriggered() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED;
+ }
+
+ public static boolean isTempQueueUsed() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED;
+ }
+
+ MemberVM getServerToStop(List<MemberVM> list) {
+ for (MemberVM member : list) {
+ if (!member.invoke(this::getPrimaryBucketList).isEmpty()) {
+ return member;
+ }
+ }
+ return null;
+ }
+
+ private void startDurableClient()
+ throws Exception {
+ int locatorPort = locatorSite2.getPort();
+ clientSite2DurableSubscription = clusterStartupRule.startClientVM(7, ccf -> ccf
+ .withPoolSubscription(true).withLocatorConnection(locatorPort).withCacheSetup(c -> c
+ .set("durable-client-id", DURABLE_CLIENT_ID)));
Review comment:
Yes, done.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] jvarenina commented on a change in pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
jvarenina commented on a change in pull request #6646:
URL: https://github.com/apache/geode/pull/6646#discussion_r661234438
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
Review comment:
Yes, you are right! Done.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] jvarenina commented on a change in pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
jvarenina commented on a change in pull request #6646:
URL: https://github.com/apache/geode/pull/6646#discussion_r659530427
##########
File path: geode-core/src/main/java/org/apache/geode/internal/cache/TXCommitMessage.java
##########
@@ -1301,14 +1301,22 @@ protected void txApplyEntryOp(FarSideEntryOp entryOp, List<EntryEventImpl> pendi
/*
* This happens when we don't have the bucket and are getting adjunct notification
*/
+ long tailKey = entryOp.tailKey;
+ if (entryOp.filterRoutingInfo != null) {
+ if (entryOp.filterRoutingInfo.getMembers()
+ .contains(this.internalRegion.getDistributionManager().getId())) {
Review comment:
Yes, I changed that. Thanks for the review!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] jvarenina commented on a change in pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
jvarenina commented on a change in pull request #6646:
URL: https://github.com/apache/geode/pull/6646#discussion_r661229191
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ // check that bucketToTempQueueMap is empty on all members
+ for (MemberVM member : allMembers) {
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ }
+
+ // Choose server with bucket to be stopped
+ MemberVM serverToStop = getServerToStop(allMembers);
+ int bucketId = serverToStop.invoke(this::getPrimaryBucketList).iterator().next();
+ serverToStop.stop(true);
+
+ // remove from list the member that has been previously stopped
+ allMembers.remove(serverToStop);
+
+ startDurableClient();
+ createDurableCQs("SELECT * FROM " + Region.SEPARATOR + REGION_NAME);
+
+ // configure hook on running members
+ for (MemberVM member : allMembers) {
+ configureHooksOnRunningMember(member, bucketId);
+ }
+
+ // perform rebalance operation to trigger bucket redundancy recovery on running servers
+ String command = new CommandStringBuilder(CliStrings.REBALANCE)
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ for (MemberVM member : allMembers) {
+ // All members after redundancy is recovered should have empty temporary queue
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ // If hook has been triggered on member, then check if member temporarily queued
+ // events while getting initial image from primary server
+ if (member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isHookTriggered)) {
+ assertThat(member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isTempQueueUsed))
+ .isTrue();
+ }
+ }
+
+ // reset all hooks
+ for (MemberVM member : allMembers) {
+ resetHooks(member);
+ }
+ }
+
+ /**
+ * This hook will run some transaction traffic after server requested initial image from
+ * primary bucket server, and then it will check if events are enqueued in temporary queue.
+ * It is expected that events are enqueued in this case because bucket is not yet available.
+ */
+ void configureHooksOnRunningMember(MemberVM server, int bucketId) {
+ server.invoke(() -> InitialImageOperation.setGIITestHook(new InitialImageOperation.GIITestHook(
+ InitialImageOperation.GIITestHookType.BeforeRequestRVV,
+ "_B__" + REGION_NAME + "_" + bucketId) {
+ private static final long serialVersionUID = -3790198435185240444L;
+
+ @Override
+ public void reset() {}
+
+ @Override
+ public void run() {
+ doPutsInServer();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED = true;
+ if (sizeOfBucketToTempQueueMap("ln") != 0) {
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED = true;
+ }
+ }
+ }));
+ }
+
+ void resetHooks(MemberVM server) {
+ server.invoke(() -> InitialImageOperation.resetAllGIITestHooks());
+ }
+
+ public static boolean isHookTriggered() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED;
+ }
+
+ public static boolean isTempQueueUsed() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED;
+ }
+
+ MemberVM getServerToStop(List<MemberVM> list) {
+ for (MemberVM member : list) {
+ if (!member.invoke(this::getPrimaryBucketList).isEmpty()) {
+ return member;
+ }
+ }
+ return null;
+ }
+
+ private void startDurableClient()
+ throws Exception {
+ int locatorPort = locatorSite2.getPort();
+ clientSite2DurableSubscription = clusterStartupRule.startClientVM(7, ccf -> ccf
+ .withPoolSubscription(true).withLocatorConnection(locatorPort).withCacheSetup(c -> c
+ .set("durable-client-id", DURABLE_CLIENT_ID)));
+ }
+
+ private void createDurableCQs(String... queries) {
+ clientSite2DurableSubscription.invoke(() -> {
+ assertThat(ClusterStartupRule.getClientCache()).isNotNull();
+ QueryService queryService = ClusterStartupRule.getClientCache().getQueryService();
+ CqAttributesFactory cqAttributesFactory = new CqAttributesFactory();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.cqListener =
+ new CqListenerTestReceivedEvents();
+ cqAttributesFactory.addCqListener(cqListener);
+
+ for (String query : queries) {
+ CqQuery cq = queryService.newCq(query, cqAttributesFactory.create(), true);
+ cq.execute();
+ }
+ ClusterStartupRule.getClientCache().readyForEvents();
+ });
+ }
+
+ private Set<Integer> getPrimaryBucketList() {
+ PartitionedRegion region = (PartitionedRegion) Objects
+ .requireNonNull(ClusterStartupRule.getCache()).getRegion(REGION_NAME);
+ return new HashSet<>(region.getDataStore().getAllLocalPrimaryBucketIds());
+ }
+
+ /**
+ * Check that all events are received on durable client that registered cq's
+ */
+ private void checkCqEvents(int expectedNumberOfEvents) {
+ // Check if number of events is correct
+ clientSite2DurableSubscription.invoke(() -> await().untilAsserted(() -> assertThat(
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.cqListener.getNumEvents())
+ .isEqualTo(expectedNumberOfEvents)));
+ }
+
+ /**
+ * Checks that the bucketToTempQueueMap for a partitioned region
+ * that holds events for buckets that are not available locally, is empty.
+ */
+ public static void validateBucketToTempQueueMap(String senderId, boolean shouldBeEmpty) {
+ final int finalSize = sizeOfBucketToTempQueueMap(senderId);
+ if (shouldBeEmpty) {
+ assertThat(finalSize).isEqualTo(0);
+ } else {
+ assertThat(finalSize).isNotEqualTo(0);
+ }
+ }
+
+ public static int sizeOfBucketToTempQueueMap(String senderId) {
+ GatewaySender sender = getGatewaySender(senderId);
+ int size = 0;
+ Set<RegionQueue> queues = ((AbstractGatewaySender) sender).getQueues();
+ for (RegionQueue queue : queues) {
+ PartitionedRegion region =
+ (PartitionedRegion) queue.getRegion();
+ int buckets = region.getTotalNumberOfBuckets();
+ for (int bucket = 0; bucket < buckets; bucket++) {
+ BlockingQueue<GatewaySenderEventImpl> newQueue =
+ ((ConcurrentParallelGatewaySenderQueue) queue).getBucketTmpQueue(bucket);
+ if (newQueue != null) {
+ size += newQueue.size();
+ }
+ }
+ }
+ return size;
+ }
+
+ private static GatewaySender getGatewaySender(String senderId) {
+ Set<GatewaySender> senders =
+ Objects.requireNonNull(ClusterStartupRule.getCache()).getGatewaySenders();
+ GatewaySender sender = null;
+ for (GatewaySender s : senders) {
+ if (s.getId().equals(senderId)) {
+ sender = s;
+ break;
+ }
+ }
+ return sender;
+ }
+
+ void configureSites(String totalBucketNum, boolean addColocatedRegion)
+ throws Exception {
+
+ Properties props = new Properties();
+ props.setProperty(DISTRIBUTED_SYSTEM_ID, DISTRIBUTED_SYSTEM_ID_SITE1);
+ MemberVM locatorSite1 = clusterStartupRule.startLocatorVM(1, props);
+
+ props.setProperty(DISTRIBUTED_SYSTEM_ID, DISTRIBUTED_SYSTEM_ID_SITE2);
+ props.setProperty(REMOTE_LOCATORS, "localhost[" + locatorSite1.getPort() + "]");
+ locatorSite2 = clusterStartupRule.startLocatorVM(2, props);
+
+ // start servers for site #2
+ server1Site2 = clusterStartupRule.startServerVM(3, locatorSite2.getPort());
+ server2Site2 = clusterStartupRule.startServerVM(4, locatorSite2.getPort());
+ server3Site2 = clusterStartupRule.startServerVM(5, locatorSite2.getPort());
+
+ // create parallel gateway-sender on site #2
+ connectGfshToSite(locatorSite2);
+ String command = new CommandStringBuilder(CliStrings.CREATE_GATEWAYSENDER)
+ .addOption(CliStrings.CREATE_GATEWAYSENDER__ID, "ln")
+ .addOption(CliStrings.CREATE_GATEWAYSENDER__REMOTEDISTRIBUTEDSYSTEMID, "1")
+ .addOption(CliStrings.CREATE_GATEWAYSENDER__PARALLEL, "true")
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ server1Site2.invoke(() -> verifySenderState("ln", true, false));
+ server2Site2.invoke(() -> verifySenderState("ln", true, false));
+ server3Site2.invoke(() -> verifySenderState("ln", true, false));
+
+ locatorSite2.invoke(
+ () -> validateGatewaySenderMXBeanProxy(getMember(server1Site2.getVM()), "ln", true, false));
+ locatorSite2.invoke(
+ () -> validateGatewaySenderMXBeanProxy(getMember(server2Site2.getVM()), "ln", true, false));
+ locatorSite2.invoke(
+ () -> validateGatewaySenderMXBeanProxy(getMember(server3Site2.getVM()), "ln", true, false));
+
+ // create partition region on site #2
+ CommandStringBuilder csb = new CommandStringBuilder(CliStrings.CREATE_REGION);
+ csb.addOption(CliStrings.CREATE_REGION__REGION, REGION_NAME);
+ csb.addOption(CliStrings.CREATE_REGION__REGIONSHORTCUT, "PARTITION");
+ csb.addOption(CliStrings.CREATE_REGION__GATEWAYSENDERID, "ln");
+ csb.addOption(CliStrings.CREATE_REGION__REDUNDANTCOPIES, "1");
+ csb.addOption(CliStrings.CREATE_REGION__TOTALNUMBUCKETS, totalBucketNum);
+ gfsh.executeAndAssertThat(csb.toString()).statusIsSuccess();
+
+ if (addColocatedRegion) {
+ csb = new CommandStringBuilder(CliStrings.CREATE_REGION);
+ csb.addOption(CliStrings.CREATE_REGION__REGION, REGION_COLOCATED);
+ csb.addOption(CliStrings.CREATE_REGION__REGIONSHORTCUT, "PARTITION");
+ csb.addOption(CliStrings.CREATE_REGION__GATEWAYSENDERID, "ln");
+ csb.addOption(CliStrings.CREATE_REGION__REDUNDANTCOPIES, "1");
+ csb.addOption(CliStrings.CREATE_REGION__TOTALNUMBUCKETS, totalBucketNum);
+ csb.addOption(CliStrings.CREATE_REGION__COLOCATEDWITH, REGION_NAME);
+ gfsh.executeAndAssertThat(csb.toString()).statusIsSuccess();
+ }
+
+ // Start client
+ clientSite2 =
+ clusterStartupRule.startClientVM(6, c -> c.withLocatorConnection(locatorSite2.getPort()));
+ clientSite2.invoke(() -> {
+ ClusterStartupRule.clientCacheRule.createProxyRegion(REGION_NAME);
+ if (addColocatedRegion) {
+ ClusterStartupRule.clientCacheRule.createProxyRegion(REGION_COLOCATED);
+ }
+ });
+ }
+
+ void connectGfshToSite(MemberVM locator) throws Exception {
+ if (gfsh.isConnected()) {
+ gfsh.disconnect();
+ }
+ gfsh.connectAndVerify(locator);
+ }
+
+ void verifyGatewaySenderState(boolean isRunning, boolean isPaused) {
+ locatorSite2.invoke(
+ () -> validateGatewaySenderMXBeanProxy(getMember(server1Site2.getVM()), "ln", isRunning,
+ isPaused));
+ locatorSite2.invoke(
+ () -> validateGatewaySenderMXBeanProxy(getMember(server2Site2.getVM()), "ln", isRunning,
+ isPaused));
+ server1Site2.invoke(() -> verifySenderState("ln", isRunning, isPaused));
+ server2Site2.invoke(() -> verifySenderState("ln", isRunning, isPaused));
+ server3Site2.invoke(() -> verifySenderState("ln", isRunning, isPaused));
+ }
+
+ Set<String> doPutsInRangeTransactionSingleKey(int start, int stop, String regionName) {
+ Region<String, String> region =
+ ClusterStartupRule.clientCacheRule.getCache().getRegion(regionName);
+ Set<String> keys = new HashSet<>();
+
+ CacheTransactionManager transactionManager =
+ Objects.requireNonNull(ClusterStartupRule.getClientCache()).getCacheTransactionManager();
+ for (int i = start; i < stop; i++) {
+ transactionManager.begin();
+ region.put(i + "key", i + "value");
+ transactionManager.commit();
+ keys.add(i + "key");
+ }
+ return keys;
+ }
+
+ Set<String> doPutsInRangeTransactionWithTwoPutOperation(int start, int stop) {
+ Region<String, String> region =
+ ClusterStartupRule.clientCacheRule.getCache().getRegion(REGION_NAME);
+
+ Region<String, String> regionC =
+ ClusterStartupRule.clientCacheRule.getCache().getRegion(REGION_COLOCATED);
+
+ Set<String> keys = new HashSet<>();
+ CacheTransactionManager transactionManager =
+ Objects.requireNonNull(ClusterStartupRule.getClientCache()).getCacheTransactionManager();
+
+ for (int i = start; i < stop; i++) {
+ transactionManager.begin();
+ region.put(i + "key", i + "value");
+ regionC.put(i + "key", i + "value");
+ transactionManager.commit();
+ keys.add(i + "key");
+ }
+ return keys;
+ }
+
+ Set<String> doPutsInRange(int start, int stop) {
+ Region<String, String> region =
+ ClusterStartupRule.clientCacheRule.getCache().getRegion(REGION_NAME);
+ Set<String> keys = new HashSet<>();
+
+ for (int i = start; i < stop; i++) {
+ region.put(i + "key", i + "value");
+ keys.add(i + "key");
+ }
+ return keys;
+ }
+
+ void doPutsInServer() {
Review comment:
Yes, done.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] jvarenina commented on a change in pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
jvarenina commented on a change in pull request #6646:
URL: https://github.com/apache/geode/pull/6646#discussion_r661204574
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ // check that bucketToTempQueueMap is empty on all members
+ for (MemberVM member : allMembers) {
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ }
+
+ // Choose server with bucket to be stopped
+ MemberVM serverToStop = getServerToStop(allMembers);
+ int bucketId = serverToStop.invoke(this::getPrimaryBucketList).iterator().next();
+ serverToStop.stop(true);
+
+ // remove from list the member that has been previously stopped
+ allMembers.remove(serverToStop);
+
+ startDurableClient();
+ createDurableCQs("SELECT * FROM " + Region.SEPARATOR + REGION_NAME);
+
+ // configure hook on running members
+ for (MemberVM member : allMembers) {
+ configureHooksOnRunningMember(member, bucketId);
+ }
+
+ // perform rebalance operation to trigger bucket redundancy recovery on running servers
+ String command = new CommandStringBuilder(CliStrings.REBALANCE)
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ for (MemberVM member : allMembers) {
+ // All members after redundancy is recovered should have empty temporary queue
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ // If hook has been triggered on member, then check if member temporarily queued
+ // events while getting initial image from primary server
+ if (member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isHookTriggered)) {
Review comment:
Hi @DonalEvans ,
Thanks for the review!
The hook **will be always triggered** on at least on one of the servers that will restore bucket redundancy. Hook is installed on all running servers and then rebalance command is triggered. This code loops through all servers to find the servers that actually restored redundancy (the ones that triggered hook isHookTriggered == true), and then verify that events has been stored to temporary gateway sender queue during recovery of bucket redundancy. So I don't think that this test case could be flaky, since hook will be always triggered on at least one server.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] nabarunnag commented on pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
nabarunnag commented on pull request #6646:
URL: https://github.com/apache/geode/pull/6646#issuecomment-869969928
- Tests are running. Please do not push this to develop for at least 24 hours.
- Also, do remember to squash the commits to one single commit while pushing to develop
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] nabarunnag merged pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
nabarunnag merged pull request #6646:
URL: https://github.com/apache/geode/pull/6646
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] pivotal-eshu commented on a change in pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
pivotal-eshu commented on a change in pull request #6646:
URL: https://github.com/apache/geode/pull/6646#discussion_r659076400
##########
File path: geode-core/src/main/java/org/apache/geode/internal/cache/TXCommitMessage.java
##########
@@ -1301,14 +1301,22 @@ protected void txApplyEntryOp(FarSideEntryOp entryOp, List<EntryEventImpl> pendi
/*
* This happens when we don't have the bucket and are getting adjunct notification
*/
+ long tailKey = entryOp.tailKey;
+ if (entryOp.filterRoutingInfo != null) {
+ if (entryOp.filterRoutingInfo.getMembers()
+ .contains(this.internalRegion.getDistributionManager().getId())) {
Review comment:
can just use .contains(internalRegion.getMyId()) instead here.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] DonalEvans commented on a change in pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
DonalEvans commented on a change in pull request #6646:
URL: https://github.com/apache/geode/pull/6646#discussion_r662416405
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ // check that bucketToTempQueueMap is empty on all members
+ for (MemberVM member : allMembers) {
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ }
+
+ // Choose server with bucket to be stopped
+ MemberVM serverToStop = getServerToStop(allMembers);
+ int bucketId = serverToStop.invoke(this::getPrimaryBucketList).iterator().next();
+ serverToStop.stop(true);
+
+ // remove from list the member that has been previously stopped
+ allMembers.remove(serverToStop);
+
+ startDurableClient();
+ createDurableCQs("SELECT * FROM " + Region.SEPARATOR + REGION_NAME);
+
+ // configure hook on running members
+ for (MemberVM member : allMembers) {
+ configureHooksOnRunningMember(member, bucketId);
+ }
+
+ // perform rebalance operation to trigger bucket redundancy recovery on running servers
+ String command = new CommandStringBuilder(CliStrings.REBALANCE)
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ for (MemberVM member : allMembers) {
+ // All members after redundancy is recovered should have empty temporary queue
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ // If hook has been triggered on member, then check if member temporarily queued
+ // events while getting initial image from primary server
+ if (member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isHookTriggered)) {
Review comment:
Aah, I see, thank you for the explanation.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] jvarenina commented on a change in pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
jvarenina commented on a change in pull request #6646:
URL: https://github.com/apache/geode/pull/6646#discussion_r661234777
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ // check that bucketToTempQueueMap is empty on all members
+ for (MemberVM member : allMembers) {
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ }
+
+ // Choose server with bucket to be stopped
+ MemberVM serverToStop = getServerToStop(allMembers);
+ int bucketId = serverToStop.invoke(this::getPrimaryBucketList).iterator().next();
+ serverToStop.stop(true);
+
+ // remove from list the member that has been previously stopped
+ allMembers.remove(serverToStop);
+
+ startDurableClient();
+ createDurableCQs("SELECT * FROM " + Region.SEPARATOR + REGION_NAME);
+
+ // configure hook on running members
+ for (MemberVM member : allMembers) {
+ configureHooksOnRunningMember(member, bucketId);
+ }
+
+ // perform rebalance operation to trigger bucket redundancy recovery on running servers
+ String command = new CommandStringBuilder(CliStrings.REBALANCE)
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ for (MemberVM member : allMembers) {
+ // All members after redundancy is recovered should have empty temporary queue
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ // If hook has been triggered on member, then check if member temporarily queued
+ // events while getting initial image from primary server
+ if (member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isHookTriggered)) {
+ assertThat(member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isTempQueueUsed))
+ .isTrue();
+ }
+ }
+
+ // reset all hooks
+ for (MemberVM member : allMembers) {
+ resetHooks(member);
+ }
+ }
+
+ /**
+ * This hook will run some transaction traffic after server requested initial image from
+ * primary bucket server, and then it will check if events are enqueued in temporary queue.
+ * It is expected that events are enqueued in this case because bucket is not yet available.
+ */
+ void configureHooksOnRunningMember(MemberVM server, int bucketId) {
+ server.invoke(() -> InitialImageOperation.setGIITestHook(new InitialImageOperation.GIITestHook(
+ InitialImageOperation.GIITestHookType.BeforeRequestRVV,
+ "_B__" + REGION_NAME + "_" + bucketId) {
+ private static final long serialVersionUID = -3790198435185240444L;
+
+ @Override
+ public void reset() {}
+
+ @Override
+ public void run() {
+ doPutsInServer();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED = true;
+ if (sizeOfBucketToTempQueueMap("ln") != 0) {
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED = true;
+ }
+ }
+ }));
+ }
+
+ void resetHooks(MemberVM server) {
+ server.invoke(() -> InitialImageOperation.resetAllGIITestHooks());
+ }
+
+ public static boolean isHookTriggered() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED;
+ }
+
+ public static boolean isTempQueueUsed() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED;
+ }
+
+ MemberVM getServerToStop(List<MemberVM> list) {
+ for (MemberVM member : list) {
+ if (!member.invoke(this::getPrimaryBucketList).isEmpty()) {
+ return member;
+ }
+ }
+ return null;
+ }
+
+ private void startDurableClient()
+ throws Exception {
+ int locatorPort = locatorSite2.getPort();
+ clientSite2DurableSubscription = clusterStartupRule.startClientVM(7, ccf -> ccf
+ .withPoolSubscription(true).withLocatorConnection(locatorPort).withCacheSetup(c -> c
+ .set("durable-client-id", DURABLE_CLIENT_ID)));
+ }
+
+ private void createDurableCQs(String... queries) {
+ clientSite2DurableSubscription.invoke(() -> {
+ assertThat(ClusterStartupRule.getClientCache()).isNotNull();
+ QueryService queryService = ClusterStartupRule.getClientCache().getQueryService();
+ CqAttributesFactory cqAttributesFactory = new CqAttributesFactory();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.cqListener =
+ new CqListenerTestReceivedEvents();
+ cqAttributesFactory.addCqListener(cqListener);
+
+ for (String query : queries) {
+ CqQuery cq = queryService.newCq(query, cqAttributesFactory.create(), true);
+ cq.execute();
+ }
+ ClusterStartupRule.getClientCache().readyForEvents();
+ });
+ }
+
+ private Set<Integer> getPrimaryBucketList() {
+ PartitionedRegion region = (PartitionedRegion) Objects
+ .requireNonNull(ClusterStartupRule.getCache()).getRegion(REGION_NAME);
+ return new HashSet<>(region.getDataStore().getAllLocalPrimaryBucketIds());
+ }
+
+ /**
+ * Check that all events are received on durable client that registered cq's
+ */
+ private void checkCqEvents(int expectedNumberOfEvents) {
+ // Check if number of events is correct
+ clientSite2DurableSubscription.invoke(() -> await().untilAsserted(() -> assertThat(
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.cqListener.getNumEvents())
+ .isEqualTo(expectedNumberOfEvents)));
+ }
+
+ /**
+ * Checks that the bucketToTempQueueMap for a partitioned region
+ * that holds events for buckets that are not available locally, is empty.
+ */
+ public static void validateBucketToTempQueueMap(String senderId, boolean shouldBeEmpty) {
+ final int finalSize = sizeOfBucketToTempQueueMap(senderId);
+ if (shouldBeEmpty) {
+ assertThat(finalSize).isEqualTo(0);
+ } else {
+ assertThat(finalSize).isNotEqualTo(0);
+ }
+ }
+
+ public static int sizeOfBucketToTempQueueMap(String senderId) {
+ GatewaySender sender = getGatewaySender(senderId);
Review comment:
Yes, I changed that.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] nabarunnag commented on pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
nabarunnag commented on pull request #6646:
URL: https://github.com/apache/geode/pull/6646#issuecomment-869969928
- Tests are running. Please do not push this to develop for at least 24 hours.
- Also, do remember to squash the commits to one single commit while pushing to develop
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] DonalEvans commented on a change in pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
DonalEvans commented on a change in pull request #6646:
URL: https://github.com/apache/geode/pull/6646#discussion_r660830905
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ // check that bucketToTempQueueMap is empty on all members
+ for (MemberVM member : allMembers) {
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ }
+
+ // Choose server with bucket to be stopped
+ MemberVM serverToStop = getServerToStop(allMembers);
+ int bucketId = serverToStop.invoke(this::getPrimaryBucketList).iterator().next();
+ serverToStop.stop(true);
+
+ // remove from list the member that has been previously stopped
+ allMembers.remove(serverToStop);
+
+ startDurableClient();
+ createDurableCQs("SELECT * FROM " + Region.SEPARATOR + REGION_NAME);
+
+ // configure hook on running members
+ for (MemberVM member : allMembers) {
+ configureHooksOnRunningMember(member, bucketId);
+ }
+
+ // perform rebalance operation to trigger bucket redundancy recovery on running servers
+ String command = new CommandStringBuilder(CliStrings.REBALANCE)
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ for (MemberVM member : allMembers) {
+ // All members after redundancy is recovered should have empty temporary queue
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ // If hook has been triggered on member, then check if member temporarily queued
+ // events while getting initial image from primary server
+ if (member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isHookTriggered)) {
+ assertThat(member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isTempQueueUsed))
+ .isTrue();
+ }
+ }
+
+ // reset all hooks
+ for (MemberVM member : allMembers) {
+ resetHooks(member);
+ }
+ }
+
+ /**
+ * This hook will run some transaction traffic after server requested initial image from
+ * primary bucket server, and then it will check if events are enqueued in temporary queue.
+ * It is expected that events are enqueued in this case because bucket is not yet available.
+ */
+ void configureHooksOnRunningMember(MemberVM server, int bucketId) {
+ server.invoke(() -> InitialImageOperation.setGIITestHook(new InitialImageOperation.GIITestHook(
+ InitialImageOperation.GIITestHookType.BeforeRequestRVV,
+ "_B__" + REGION_NAME + "_" + bucketId) {
+ private static final long serialVersionUID = -3790198435185240444L;
+
+ @Override
+ public void reset() {}
+
+ @Override
+ public void run() {
+ doPutsInServer();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED = true;
+ if (sizeOfBucketToTempQueueMap("ln") != 0) {
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED = true;
+ }
+ }
+ }));
+ }
+
+ void resetHooks(MemberVM server) {
+ server.invoke(() -> InitialImageOperation.resetAllGIITestHooks());
+ }
+
+ public static boolean isHookTriggered() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED;
+ }
+
+ public static boolean isTempQueueUsed() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED;
+ }
+
+ MemberVM getServerToStop(List<MemberVM> list) {
+ for (MemberVM member : list) {
+ if (!member.invoke(this::getPrimaryBucketList).isEmpty()) {
+ return member;
+ }
+ }
+ return null;
+ }
+
+ private void startDurableClient()
+ throws Exception {
+ int locatorPort = locatorSite2.getPort();
+ clientSite2DurableSubscription = clusterStartupRule.startClientVM(7, ccf -> ccf
+ .withPoolSubscription(true).withLocatorConnection(locatorPort).withCacheSetup(c -> c
+ .set("durable-client-id", DURABLE_CLIENT_ID)));
+ }
+
+ private void createDurableCQs(String... queries) {
+ clientSite2DurableSubscription.invoke(() -> {
+ assertThat(ClusterStartupRule.getClientCache()).isNotNull();
+ QueryService queryService = ClusterStartupRule.getClientCache().getQueryService();
+ CqAttributesFactory cqAttributesFactory = new CqAttributesFactory();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.cqListener =
+ new CqListenerTestReceivedEvents();
+ cqAttributesFactory.addCqListener(cqListener);
+
+ for (String query : queries) {
+ CqQuery cq = queryService.newCq(query, cqAttributesFactory.create(), true);
+ cq.execute();
+ }
+ ClusterStartupRule.getClientCache().readyForEvents();
+ });
+ }
+
+ private Set<Integer> getPrimaryBucketList() {
+ PartitionedRegion region = (PartitionedRegion) Objects
+ .requireNonNull(ClusterStartupRule.getCache()).getRegion(REGION_NAME);
+ return new HashSet<>(region.getDataStore().getAllLocalPrimaryBucketIds());
+ }
+
+ /**
+ * Check that all events are received on durable client that registered cq's
+ */
+ private void checkCqEvents(int expectedNumberOfEvents) {
+ // Check if number of events is correct
+ clientSite2DurableSubscription.invoke(() -> await().untilAsserted(() -> assertThat(
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.cqListener.getNumEvents())
+ .isEqualTo(expectedNumberOfEvents)));
+ }
+
+ /**
+ * Checks that the bucketToTempQueueMap for a partitioned region
+ * that holds events for buckets that are not available locally, is empty.
+ */
+ public static void validateBucketToTempQueueMap(String senderId, boolean shouldBeEmpty) {
Review comment:
The value of `shouldBeEmpty` is always `true`. It might be simpler to rename this method "validateBucketToTempQueueMapIsEmpty" and inline the value.
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
Review comment:
It looks like this test never uses more than 7 VMs, so it probably doesn't need to start 9 every time.
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ // check that bucketToTempQueueMap is empty on all members
+ for (MemberVM member : allMembers) {
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ }
+
+ // Choose server with bucket to be stopped
+ MemberVM serverToStop = getServerToStop(allMembers);
+ int bucketId = serverToStop.invoke(this::getPrimaryBucketList).iterator().next();
+ serverToStop.stop(true);
+
+ // remove from list the member that has been previously stopped
+ allMembers.remove(serverToStop);
+
+ startDurableClient();
+ createDurableCQs("SELECT * FROM " + Region.SEPARATOR + REGION_NAME);
+
+ // configure hook on running members
+ for (MemberVM member : allMembers) {
+ configureHooksOnRunningMember(member, bucketId);
+ }
+
+ // perform rebalance operation to trigger bucket redundancy recovery on running servers
+ String command = new CommandStringBuilder(CliStrings.REBALANCE)
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ for (MemberVM member : allMembers) {
+ // All members after redundancy is recovered should have empty temporary queue
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ // If hook has been triggered on member, then check if member temporarily queued
+ // events while getting initial image from primary server
+ if (member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isHookTriggered)) {
+ assertThat(member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isTempQueueUsed))
+ .isTrue();
+ }
+ }
+
+ // reset all hooks
+ for (MemberVM member : allMembers) {
+ resetHooks(member);
+ }
+ }
+
+ /**
+ * This hook will run some transaction traffic after server requested initial image from
+ * primary bucket server, and then it will check if events are enqueued in temporary queue.
+ * It is expected that events are enqueued in this case because bucket is not yet available.
+ */
+ void configureHooksOnRunningMember(MemberVM server, int bucketId) {
+ server.invoke(() -> InitialImageOperation.setGIITestHook(new InitialImageOperation.GIITestHook(
+ InitialImageOperation.GIITestHookType.BeforeRequestRVV,
+ "_B__" + REGION_NAME + "_" + bucketId) {
+ private static final long serialVersionUID = -3790198435185240444L;
+
+ @Override
+ public void reset() {}
+
+ @Override
+ public void run() {
+ doPutsInServer();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED = true;
+ if (sizeOfBucketToTempQueueMap("ln") != 0) {
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED = true;
+ }
+ }
+ }));
+ }
+
+ void resetHooks(MemberVM server) {
+ server.invoke(() -> InitialImageOperation.resetAllGIITestHooks());
+ }
+
+ public static boolean isHookTriggered() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED;
+ }
+
+ public static boolean isTempQueueUsed() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED;
+ }
+
+ MemberVM getServerToStop(List<MemberVM> list) {
+ for (MemberVM member : list) {
+ if (!member.invoke(this::getPrimaryBucketList).isEmpty()) {
+ return member;
+ }
+ }
+ return null;
+ }
+
+ private void startDurableClient()
+ throws Exception {
+ int locatorPort = locatorSite2.getPort();
+ clientSite2DurableSubscription = clusterStartupRule.startClientVM(7, ccf -> ccf
+ .withPoolSubscription(true).withLocatorConnection(locatorPort).withCacheSetup(c -> c
+ .set("durable-client-id", DURABLE_CLIENT_ID)));
+ }
+
+ private void createDurableCQs(String... queries) {
+ clientSite2DurableSubscription.invoke(() -> {
+ assertThat(ClusterStartupRule.getClientCache()).isNotNull();
+ QueryService queryService = ClusterStartupRule.getClientCache().getQueryService();
+ CqAttributesFactory cqAttributesFactory = new CqAttributesFactory();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.cqListener =
+ new CqListenerTestReceivedEvents();
+ cqAttributesFactory.addCqListener(cqListener);
+
+ for (String query : queries) {
+ CqQuery cq = queryService.newCq(query, cqAttributesFactory.create(), true);
+ cq.execute();
+ }
+ ClusterStartupRule.getClientCache().readyForEvents();
+ });
+ }
+
+ private Set<Integer> getPrimaryBucketList() {
+ PartitionedRegion region = (PartitionedRegion) Objects
+ .requireNonNull(ClusterStartupRule.getCache()).getRegion(REGION_NAME);
+ return new HashSet<>(region.getDataStore().getAllLocalPrimaryBucketIds());
+ }
+
+ /**
+ * Check that all events are received on durable client that registered cq's
+ */
+ private void checkCqEvents(int expectedNumberOfEvents) {
+ // Check if number of events is correct
+ clientSite2DurableSubscription.invoke(() -> await().untilAsserted(() -> assertThat(
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.cqListener.getNumEvents())
+ .isEqualTo(expectedNumberOfEvents)));
+ }
+
+ /**
+ * Checks that the bucketToTempQueueMap for a partitioned region
+ * that holds events for buckets that are not available locally, is empty.
+ */
+ public static void validateBucketToTempQueueMap(String senderId, boolean shouldBeEmpty) {
+ final int finalSize = sizeOfBucketToTempQueueMap(senderId);
+ if (shouldBeEmpty) {
+ assertThat(finalSize).isEqualTo(0);
+ } else {
+ assertThat(finalSize).isNotEqualTo(0);
+ }
+ }
+
+ public static int sizeOfBucketToTempQueueMap(String senderId) {
+ GatewaySender sender = getGatewaySender(senderId);
+ int size = 0;
+ Set<RegionQueue> queues = ((AbstractGatewaySender) sender).getQueues();
+ for (RegionQueue queue : queues) {
+ PartitionedRegion region =
+ (PartitionedRegion) queue.getRegion();
+ int buckets = region.getTotalNumberOfBuckets();
+ for (int bucket = 0; bucket < buckets; bucket++) {
+ BlockingQueue<GatewaySenderEventImpl> newQueue =
+ ((ConcurrentParallelGatewaySenderQueue) queue).getBucketTmpQueue(bucket);
+ if (newQueue != null) {
+ size += newQueue.size();
+ }
+ }
+ }
+ return size;
+ }
+
+ private static GatewaySender getGatewaySender(String senderId) {
+ Set<GatewaySender> senders =
+ Objects.requireNonNull(ClusterStartupRule.getCache()).getGatewaySenders();
+ GatewaySender sender = null;
+ for (GatewaySender s : senders) {
+ if (s.getId().equals(senderId)) {
+ sender = s;
+ break;
+ }
+ }
+ return sender;
+ }
+
+ void configureSites(String totalBucketNum, boolean addColocatedRegion)
+ throws Exception {
+
+ Properties props = new Properties();
+ props.setProperty(DISTRIBUTED_SYSTEM_ID, DISTRIBUTED_SYSTEM_ID_SITE1);
+ MemberVM locatorSite1 = clusterStartupRule.startLocatorVM(1, props);
+
+ props.setProperty(DISTRIBUTED_SYSTEM_ID, DISTRIBUTED_SYSTEM_ID_SITE2);
+ props.setProperty(REMOTE_LOCATORS, "localhost[" + locatorSite1.getPort() + "]");
+ locatorSite2 = clusterStartupRule.startLocatorVM(2, props);
+
+ // start servers for site #2
+ server1Site2 = clusterStartupRule.startServerVM(3, locatorSite2.getPort());
+ server2Site2 = clusterStartupRule.startServerVM(4, locatorSite2.getPort());
+ server3Site2 = clusterStartupRule.startServerVM(5, locatorSite2.getPort());
+
+ // create parallel gateway-sender on site #2
+ connectGfshToSite(locatorSite2);
+ String command = new CommandStringBuilder(CliStrings.CREATE_GATEWAYSENDER)
+ .addOption(CliStrings.CREATE_GATEWAYSENDER__ID, "ln")
+ .addOption(CliStrings.CREATE_GATEWAYSENDER__REMOTEDISTRIBUTEDSYSTEMID, "1")
+ .addOption(CliStrings.CREATE_GATEWAYSENDER__PARALLEL, "true")
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ server1Site2.invoke(() -> verifySenderState("ln", true, false));
+ server2Site2.invoke(() -> verifySenderState("ln", true, false));
+ server3Site2.invoke(() -> verifySenderState("ln", true, false));
+
+ locatorSite2.invoke(
+ () -> validateGatewaySenderMXBeanProxy(getMember(server1Site2.getVM()), "ln", true, false));
+ locatorSite2.invoke(
+ () -> validateGatewaySenderMXBeanProxy(getMember(server2Site2.getVM()), "ln", true, false));
+ locatorSite2.invoke(
+ () -> validateGatewaySenderMXBeanProxy(getMember(server3Site2.getVM()), "ln", true, false));
+
+ // create partition region on site #2
+ CommandStringBuilder csb = new CommandStringBuilder(CliStrings.CREATE_REGION);
+ csb.addOption(CliStrings.CREATE_REGION__REGION, REGION_NAME);
+ csb.addOption(CliStrings.CREATE_REGION__REGIONSHORTCUT, "PARTITION");
+ csb.addOption(CliStrings.CREATE_REGION__GATEWAYSENDERID, "ln");
+ csb.addOption(CliStrings.CREATE_REGION__REDUNDANTCOPIES, "1");
+ csb.addOption(CliStrings.CREATE_REGION__TOTALNUMBUCKETS, totalBucketNum);
+ gfsh.executeAndAssertThat(csb.toString()).statusIsSuccess();
+
+ if (addColocatedRegion) {
+ csb = new CommandStringBuilder(CliStrings.CREATE_REGION);
+ csb.addOption(CliStrings.CREATE_REGION__REGION, REGION_COLOCATED);
+ csb.addOption(CliStrings.CREATE_REGION__REGIONSHORTCUT, "PARTITION");
+ csb.addOption(CliStrings.CREATE_REGION__GATEWAYSENDERID, "ln");
+ csb.addOption(CliStrings.CREATE_REGION__REDUNDANTCOPIES, "1");
+ csb.addOption(CliStrings.CREATE_REGION__TOTALNUMBUCKETS, totalBucketNum);
+ csb.addOption(CliStrings.CREATE_REGION__COLOCATEDWITH, REGION_NAME);
+ gfsh.executeAndAssertThat(csb.toString()).statusIsSuccess();
+ }
+
+ // Start client
+ clientSite2 =
+ clusterStartupRule.startClientVM(6, c -> c.withLocatorConnection(locatorSite2.getPort()));
+ clientSite2.invoke(() -> {
+ ClusterStartupRule.clientCacheRule.createProxyRegion(REGION_NAME);
+ if (addColocatedRegion) {
+ ClusterStartupRule.clientCacheRule.createProxyRegion(REGION_COLOCATED);
+ }
+ });
+ }
+
+ void connectGfshToSite(MemberVM locator) throws Exception {
+ if (gfsh.isConnected()) {
+ gfsh.disconnect();
+ }
+ gfsh.connectAndVerify(locator);
+ }
+
+ void verifyGatewaySenderState(boolean isRunning, boolean isPaused) {
+ locatorSite2.invoke(
+ () -> validateGatewaySenderMXBeanProxy(getMember(server1Site2.getVM()), "ln", isRunning,
+ isPaused));
+ locatorSite2.invoke(
+ () -> validateGatewaySenderMXBeanProxy(getMember(server2Site2.getVM()), "ln", isRunning,
+ isPaused));
+ server1Site2.invoke(() -> verifySenderState("ln", isRunning, isPaused));
+ server2Site2.invoke(() -> verifySenderState("ln", isRunning, isPaused));
+ server3Site2.invoke(() -> verifySenderState("ln", isRunning, isPaused));
+ }
+
+ Set<String> doPutsInRangeTransactionSingleKey(int start, int stop, String regionName) {
+ Region<String, String> region =
+ ClusterStartupRule.clientCacheRule.getCache().getRegion(regionName);
+ Set<String> keys = new HashSet<>();
+
+ CacheTransactionManager transactionManager =
+ Objects.requireNonNull(ClusterStartupRule.getClientCache()).getCacheTransactionManager();
+ for (int i = start; i < stop; i++) {
+ transactionManager.begin();
+ region.put(i + "key", i + "value");
+ transactionManager.commit();
+ keys.add(i + "key");
+ }
+ return keys;
+ }
+
+ Set<String> doPutsInRangeTransactionWithTwoPutOperation(int start, int stop) {
+ Region<String, String> region =
+ ClusterStartupRule.clientCacheRule.getCache().getRegion(REGION_NAME);
+
+ Region<String, String> regionC =
+ ClusterStartupRule.clientCacheRule.getCache().getRegion(REGION_COLOCATED);
+
+ Set<String> keys = new HashSet<>();
+ CacheTransactionManager transactionManager =
+ Objects.requireNonNull(ClusterStartupRule.getClientCache()).getCacheTransactionManager();
+
+ for (int i = start; i < stop; i++) {
+ transactionManager.begin();
+ region.put(i + "key", i + "value");
+ regionC.put(i + "key", i + "value");
+ transactionManager.commit();
+ keys.add(i + "key");
+ }
+ return keys;
+ }
+
+ Set<String> doPutsInRange(int start, int stop) {
+ Region<String, String> region =
+ ClusterStartupRule.clientCacheRule.getCache().getRegion(REGION_NAME);
+ Set<String> keys = new HashSet<>();
+
+ for (int i = start; i < stop; i++) {
+ region.put(i + "key", i + "value");
+ keys.add(i + "key");
+ }
+ return keys;
+ }
+
+ void doPutsInServer() {
Review comment:
Could this method name be changed to reflect the fact that the puts are being done in transactions please?
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ // check that bucketToTempQueueMap is empty on all members
+ for (MemberVM member : allMembers) {
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ }
+
+ // Choose server with bucket to be stopped
+ MemberVM serverToStop = getServerToStop(allMembers);
+ int bucketId = serverToStop.invoke(this::getPrimaryBucketList).iterator().next();
+ serverToStop.stop(true);
+
+ // remove from list the member that has been previously stopped
+ allMembers.remove(serverToStop);
+
+ startDurableClient();
+ createDurableCQs("SELECT * FROM " + Region.SEPARATOR + REGION_NAME);
+
+ // configure hook on running members
+ for (MemberVM member : allMembers) {
+ configureHooksOnRunningMember(member, bucketId);
+ }
+
+ // perform rebalance operation to trigger bucket redundancy recovery on running servers
+ String command = new CommandStringBuilder(CliStrings.REBALANCE)
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ for (MemberVM member : allMembers) {
+ // All members after redundancy is recovered should have empty temporary queue
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ // If hook has been triggered on member, then check if member temporarily queued
+ // events while getting initial image from primary server
+ if (member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isHookTriggered)) {
Review comment:
If the hook is not triggered every time the test runs, then this test may become flaky in the future. Do we have a way of ensuring that the hook is always triggered, or an idea of how often it is/isn't from run to run?
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ // check that bucketToTempQueueMap is empty on all members
+ for (MemberVM member : allMembers) {
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ }
+
+ // Choose server with bucket to be stopped
+ MemberVM serverToStop = getServerToStop(allMembers);
+ int bucketId = serverToStop.invoke(this::getPrimaryBucketList).iterator().next();
+ serverToStop.stop(true);
+
+ // remove from list the member that has been previously stopped
+ allMembers.remove(serverToStop);
+
+ startDurableClient();
+ createDurableCQs("SELECT * FROM " + Region.SEPARATOR + REGION_NAME);
+
+ // configure hook on running members
+ for (MemberVM member : allMembers) {
+ configureHooksOnRunningMember(member, bucketId);
+ }
+
+ // perform rebalance operation to trigger bucket redundancy recovery on running servers
+ String command = new CommandStringBuilder(CliStrings.REBALANCE)
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ for (MemberVM member : allMembers) {
+ // All members after redundancy is recovered should have empty temporary queue
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ // If hook has been triggered on member, then check if member temporarily queued
+ // events while getting initial image from primary server
+ if (member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isHookTriggered)) {
+ assertThat(member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isTempQueueUsed))
+ .isTrue();
+ }
+ }
+
+ // reset all hooks
+ for (MemberVM member : allMembers) {
+ resetHooks(member);
+ }
+ }
+
+ /**
+ * This hook will run some transaction traffic after server requested initial image from
+ * primary bucket server, and then it will check if events are enqueued in temporary queue.
+ * It is expected that events are enqueued in this case because bucket is not yet available.
+ */
+ void configureHooksOnRunningMember(MemberVM server, int bucketId) {
+ server.invoke(() -> InitialImageOperation.setGIITestHook(new InitialImageOperation.GIITestHook(
+ InitialImageOperation.GIITestHookType.BeforeRequestRVV,
+ "_B__" + REGION_NAME + "_" + bucketId) {
+ private static final long serialVersionUID = -3790198435185240444L;
+
+ @Override
+ public void reset() {}
+
+ @Override
+ public void run() {
+ doPutsInServer();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED = true;
+ if (sizeOfBucketToTempQueueMap("ln") != 0) {
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED = true;
+ }
+ }
+ }));
+ }
+
+ void resetHooks(MemberVM server) {
+ server.invoke(() -> InitialImageOperation.resetAllGIITestHooks());
+ }
+
+ public static boolean isHookTriggered() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED;
+ }
+
+ public static boolean isTempQueueUsed() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED;
+ }
+
+ MemberVM getServerToStop(List<MemberVM> list) {
+ for (MemberVM member : list) {
+ if (!member.invoke(this::getPrimaryBucketList).isEmpty()) {
+ return member;
+ }
+ }
+ return null;
+ }
+
+ private void startDurableClient()
+ throws Exception {
+ int locatorPort = locatorSite2.getPort();
+ clientSite2DurableSubscription = clusterStartupRule.startClientVM(7, ccf -> ccf
+ .withPoolSubscription(true).withLocatorConnection(locatorPort).withCacheSetup(c -> c
+ .set("durable-client-id", DURABLE_CLIENT_ID)));
Review comment:
Could this use the `ConfigurationProperties.DURABLE_CLIENT_ID` constant rather than a hard-coded String?
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ // check that bucketToTempQueueMap is empty on all members
+ for (MemberVM member : allMembers) {
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ }
+
+ // Choose server with bucket to be stopped
+ MemberVM serverToStop = getServerToStop(allMembers);
+ int bucketId = serverToStop.invoke(this::getPrimaryBucketList).iterator().next();
+ serverToStop.stop(true);
+
+ // remove from list the member that has been previously stopped
+ allMembers.remove(serverToStop);
+
+ startDurableClient();
+ createDurableCQs("SELECT * FROM " + Region.SEPARATOR + REGION_NAME);
+
+ // configure hook on running members
+ for (MemberVM member : allMembers) {
+ configureHooksOnRunningMember(member, bucketId);
+ }
+
+ // perform rebalance operation to trigger bucket redundancy recovery on running servers
+ String command = new CommandStringBuilder(CliStrings.REBALANCE)
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ for (MemberVM member : allMembers) {
+ // All members after redundancy is recovered should have empty temporary queue
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ // If hook has been triggered on member, then check if member temporarily queued
+ // events while getting initial image from primary server
+ if (member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isHookTriggered)) {
+ assertThat(member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isTempQueueUsed))
+ .isTrue();
+ }
+ }
+
+ // reset all hooks
+ for (MemberVM member : allMembers) {
+ resetHooks(member);
+ }
+ }
+
+ /**
+ * This hook will run some transaction traffic after server requested initial image from
+ * primary bucket server, and then it will check if events are enqueued in temporary queue.
+ * It is expected that events are enqueued in this case because bucket is not yet available.
+ */
+ void configureHooksOnRunningMember(MemberVM server, int bucketId) {
+ server.invoke(() -> InitialImageOperation.setGIITestHook(new InitialImageOperation.GIITestHook(
+ InitialImageOperation.GIITestHookType.BeforeRequestRVV,
+ "_B__" + REGION_NAME + "_" + bucketId) {
+ private static final long serialVersionUID = -3790198435185240444L;
+
+ @Override
+ public void reset() {}
+
+ @Override
+ public void run() {
+ doPutsInServer();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED = true;
+ if (sizeOfBucketToTempQueueMap("ln") != 0) {
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED = true;
+ }
+ }
+ }));
+ }
+
+ void resetHooks(MemberVM server) {
+ server.invoke(() -> InitialImageOperation.resetAllGIITestHooks());
+ }
+
+ public static boolean isHookTriggered() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED;
+ }
+
+ public static boolean isTempQueueUsed() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED;
+ }
+
+ MemberVM getServerToStop(List<MemberVM> list) {
+ for (MemberVM member : list) {
+ if (!member.invoke(this::getPrimaryBucketList).isEmpty()) {
+ return member;
+ }
+ }
+ return null;
+ }
+
+ private void startDurableClient()
+ throws Exception {
+ int locatorPort = locatorSite2.getPort();
+ clientSite2DurableSubscription = clusterStartupRule.startClientVM(7, ccf -> ccf
+ .withPoolSubscription(true).withLocatorConnection(locatorPort).withCacheSetup(c -> c
+ .set("durable-client-id", DURABLE_CLIENT_ID)));
+ }
+
+ private void createDurableCQs(String... queries) {
+ clientSite2DurableSubscription.invoke(() -> {
+ assertThat(ClusterStartupRule.getClientCache()).isNotNull();
+ QueryService queryService = ClusterStartupRule.getClientCache().getQueryService();
+ CqAttributesFactory cqAttributesFactory = new CqAttributesFactory();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.cqListener =
+ new CqListenerTestReceivedEvents();
+ cqAttributesFactory.addCqListener(cqListener);
+
+ for (String query : queries) {
+ CqQuery cq = queryService.newCq(query, cqAttributesFactory.create(), true);
+ cq.execute();
+ }
+ ClusterStartupRule.getClientCache().readyForEvents();
+ });
+ }
+
+ private Set<Integer> getPrimaryBucketList() {
+ PartitionedRegion region = (PartitionedRegion) Objects
+ .requireNonNull(ClusterStartupRule.getCache()).getRegion(REGION_NAME);
+ return new HashSet<>(region.getDataStore().getAllLocalPrimaryBucketIds());
+ }
+
+ /**
+ * Check that all events are received on durable client that registered cq's
+ */
+ private void checkCqEvents(int expectedNumberOfEvents) {
+ // Check if number of events is correct
+ clientSite2DurableSubscription.invoke(() -> await().untilAsserted(() -> assertThat(
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.cqListener.getNumEvents())
+ .isEqualTo(expectedNumberOfEvents)));
+ }
+
+ /**
+ * Checks that the bucketToTempQueueMap for a partitioned region
+ * that holds events for buckets that are not available locally, is empty.
+ */
+ public static void validateBucketToTempQueueMap(String senderId, boolean shouldBeEmpty) {
+ final int finalSize = sizeOfBucketToTempQueueMap(senderId);
+ if (shouldBeEmpty) {
+ assertThat(finalSize).isEqualTo(0);
+ } else {
+ assertThat(finalSize).isNotEqualTo(0);
+ }
+ }
+
+ public static int sizeOfBucketToTempQueueMap(String senderId) {
+ GatewaySender sender = getGatewaySender(senderId);
Review comment:
Can this method call be replaced with a call to `Cache.getGatewaySender(String id)`? It seems to do the same thing as the `getGatewaySender()` method implemented in this class.
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ // check that bucketToTempQueueMap is empty on all members
+ for (MemberVM member : allMembers) {
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ }
+
+ // Choose server with bucket to be stopped
+ MemberVM serverToStop = getServerToStop(allMembers);
+ int bucketId = serverToStop.invoke(this::getPrimaryBucketList).iterator().next();
+ serverToStop.stop(true);
+
+ // remove from list the member that has been previously stopped
+ allMembers.remove(serverToStop);
+
+ startDurableClient();
+ createDurableCQs("SELECT * FROM " + Region.SEPARATOR + REGION_NAME);
+
+ // configure hook on running members
+ for (MemberVM member : allMembers) {
+ configureHooksOnRunningMember(member, bucketId);
+ }
+
+ // perform rebalance operation to trigger bucket redundancy recovery on running servers
+ String command = new CommandStringBuilder(CliStrings.REBALANCE)
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ for (MemberVM member : allMembers) {
+ // All members after redundancy is recovered should have empty temporary queue
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ // If hook has been triggered on member, then check if member temporarily queued
+ // events while getting initial image from primary server
+ if (member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isHookTriggered)) {
+ assertThat(member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isTempQueueUsed))
+ .isTrue();
+ }
+ }
+
+ // reset all hooks
+ for (MemberVM member : allMembers) {
+ resetHooks(member);
+ }
+ }
+
+ /**
+ * This hook will run some transaction traffic after server requested initial image from
+ * primary bucket server, and then it will check if events are enqueued in temporary queue.
+ * It is expected that events are enqueued in this case because bucket is not yet available.
+ */
+ void configureHooksOnRunningMember(MemberVM server, int bucketId) {
+ server.invoke(() -> InitialImageOperation.setGIITestHook(new InitialImageOperation.GIITestHook(
+ InitialImageOperation.GIITestHookType.BeforeRequestRVV,
+ "_B__" + REGION_NAME + "_" + bucketId) {
+ private static final long serialVersionUID = -3790198435185240444L;
+
+ @Override
+ public void reset() {}
+
+ @Override
+ public void run() {
+ doPutsInServer();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED = true;
+ if (sizeOfBucketToTempQueueMap("ln") != 0) {
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED = true;
+ }
+ }
+ }));
+ }
+
+ void resetHooks(MemberVM server) {
+ server.invoke(() -> InitialImageOperation.resetAllGIITestHooks());
+ }
+
+ public static boolean isHookTriggered() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED;
+ }
+
+ public static boolean isTempQueueUsed() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED;
+ }
+
+ MemberVM getServerToStop(List<MemberVM> list) {
+ for (MemberVM member : list) {
+ if (!member.invoke(this::getPrimaryBucketList).isEmpty()) {
+ return member;
+ }
+ }
+ return null;
+ }
+
+ private void startDurableClient()
+ throws Exception {
+ int locatorPort = locatorSite2.getPort();
+ clientSite2DurableSubscription = clusterStartupRule.startClientVM(7, ccf -> ccf
+ .withPoolSubscription(true).withLocatorConnection(locatorPort).withCacheSetup(c -> c
+ .set("durable-client-id", DURABLE_CLIENT_ID)));
+ }
+
+ private void createDurableCQs(String... queries) {
+ clientSite2DurableSubscription.invoke(() -> {
+ assertThat(ClusterStartupRule.getClientCache()).isNotNull();
+ QueryService queryService = ClusterStartupRule.getClientCache().getQueryService();
+ CqAttributesFactory cqAttributesFactory = new CqAttributesFactory();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.cqListener =
+ new CqListenerTestReceivedEvents();
+ cqAttributesFactory.addCqListener(cqListener);
+
+ for (String query : queries) {
+ CqQuery cq = queryService.newCq(query, cqAttributesFactory.create(), true);
+ cq.execute();
+ }
+ ClusterStartupRule.getClientCache().readyForEvents();
+ });
+ }
+
+ private Set<Integer> getPrimaryBucketList() {
+ PartitionedRegion region = (PartitionedRegion) Objects
+ .requireNonNull(ClusterStartupRule.getCache()).getRegion(REGION_NAME);
+ return new HashSet<>(region.getDataStore().getAllLocalPrimaryBucketIds());
+ }
+
+ /**
+ * Check that all events are received on durable client that registered cq's
+ */
+ private void checkCqEvents(int expectedNumberOfEvents) {
+ // Check if number of events is correct
+ clientSite2DurableSubscription.invoke(() -> await().untilAsserted(() -> assertThat(
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.cqListener.getNumEvents())
+ .isEqualTo(expectedNumberOfEvents)));
+ }
+
+ /**
+ * Checks that the bucketToTempQueueMap for a partitioned region
+ * that holds events for buckets that are not available locally, is empty.
+ */
+ public static void validateBucketToTempQueueMap(String senderId, boolean shouldBeEmpty) {
+ final int finalSize = sizeOfBucketToTempQueueMap(senderId);
+ if (shouldBeEmpty) {
+ assertThat(finalSize).isEqualTo(0);
+ } else {
+ assertThat(finalSize).isNotEqualTo(0);
+ }
+ }
+
+ public static int sizeOfBucketToTempQueueMap(String senderId) {
+ GatewaySender sender = getGatewaySender(senderId);
+ int size = 0;
+ Set<RegionQueue> queues = ((AbstractGatewaySender) sender).getQueues();
+ for (RegionQueue queue : queues) {
+ PartitionedRegion region =
+ (PartitionedRegion) queue.getRegion();
+ int buckets = region.getTotalNumberOfBuckets();
+ for (int bucket = 0; bucket < buckets; bucket++) {
+ BlockingQueue<GatewaySenderEventImpl> newQueue =
+ ((ConcurrentParallelGatewaySenderQueue) queue).getBucketTmpQueue(bucket);
+ if (newQueue != null) {
+ size += newQueue.size();
+ }
+ }
+ }
+ return size;
+ }
+
+ private static GatewaySender getGatewaySender(String senderId) {
+ Set<GatewaySender> senders =
+ Objects.requireNonNull(ClusterStartupRule.getCache()).getGatewaySenders();
+ GatewaySender sender = null;
+ for (GatewaySender s : senders) {
+ if (s.getId().equals(senderId)) {
+ sender = s;
+ break;
+ }
+ }
+ return sender;
+ }
+
+ void configureSites(String totalBucketNum, boolean addColocatedRegion)
+ throws Exception {
+
+ Properties props = new Properties();
+ props.setProperty(DISTRIBUTED_SYSTEM_ID, DISTRIBUTED_SYSTEM_ID_SITE1);
+ MemberVM locatorSite1 = clusterStartupRule.startLocatorVM(1, props);
+
+ props.setProperty(DISTRIBUTED_SYSTEM_ID, DISTRIBUTED_SYSTEM_ID_SITE2);
+ props.setProperty(REMOTE_LOCATORS, "localhost[" + locatorSite1.getPort() + "]");
+ locatorSite2 = clusterStartupRule.startLocatorVM(2, props);
+
+ // start servers for site #2
+ server1Site2 = clusterStartupRule.startServerVM(3, locatorSite2.getPort());
+ server2Site2 = clusterStartupRule.startServerVM(4, locatorSite2.getPort());
+ server3Site2 = clusterStartupRule.startServerVM(5, locatorSite2.getPort());
+
+ // create parallel gateway-sender on site #2
+ connectGfshToSite(locatorSite2);
+ String command = new CommandStringBuilder(CliStrings.CREATE_GATEWAYSENDER)
+ .addOption(CliStrings.CREATE_GATEWAYSENDER__ID, "ln")
+ .addOption(CliStrings.CREATE_GATEWAYSENDER__REMOTEDISTRIBUTEDSYSTEMID, "1")
+ .addOption(CliStrings.CREATE_GATEWAYSENDER__PARALLEL, "true")
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ server1Site2.invoke(() -> verifySenderState("ln", true, false));
+ server2Site2.invoke(() -> verifySenderState("ln", true, false));
+ server3Site2.invoke(() -> verifySenderState("ln", true, false));
+
+ locatorSite2.invoke(
+ () -> validateGatewaySenderMXBeanProxy(getMember(server1Site2.getVM()), "ln", true, false));
+ locatorSite2.invoke(
+ () -> validateGatewaySenderMXBeanProxy(getMember(server2Site2.getVM()), "ln", true, false));
+ locatorSite2.invoke(
+ () -> validateGatewaySenderMXBeanProxy(getMember(server3Site2.getVM()), "ln", true, false));
+
+ // create partition region on site #2
+ CommandStringBuilder csb = new CommandStringBuilder(CliStrings.CREATE_REGION);
+ csb.addOption(CliStrings.CREATE_REGION__REGION, REGION_NAME);
+ csb.addOption(CliStrings.CREATE_REGION__REGIONSHORTCUT, "PARTITION");
+ csb.addOption(CliStrings.CREATE_REGION__GATEWAYSENDERID, "ln");
+ csb.addOption(CliStrings.CREATE_REGION__REDUNDANTCOPIES, "1");
+ csb.addOption(CliStrings.CREATE_REGION__TOTALNUMBUCKETS, totalBucketNum);
+ gfsh.executeAndAssertThat(csb.toString()).statusIsSuccess();
+
+ if (addColocatedRegion) {
+ csb = new CommandStringBuilder(CliStrings.CREATE_REGION);
+ csb.addOption(CliStrings.CREATE_REGION__REGION, REGION_COLOCATED);
+ csb.addOption(CliStrings.CREATE_REGION__REGIONSHORTCUT, "PARTITION");
+ csb.addOption(CliStrings.CREATE_REGION__GATEWAYSENDERID, "ln");
+ csb.addOption(CliStrings.CREATE_REGION__REDUNDANTCOPIES, "1");
+ csb.addOption(CliStrings.CREATE_REGION__TOTALNUMBUCKETS, totalBucketNum);
+ csb.addOption(CliStrings.CREATE_REGION__COLOCATEDWITH, REGION_NAME);
+ gfsh.executeAndAssertThat(csb.toString()).statusIsSuccess();
+ }
+
+ // Start client
+ clientSite2 =
+ clusterStartupRule.startClientVM(6, c -> c.withLocatorConnection(locatorSite2.getPort()));
+ clientSite2.invoke(() -> {
+ ClusterStartupRule.clientCacheRule.createProxyRegion(REGION_NAME);
+ if (addColocatedRegion) {
+ ClusterStartupRule.clientCacheRule.createProxyRegion(REGION_COLOCATED);
+ }
+ });
+ }
+
+ void connectGfshToSite(MemberVM locator) throws Exception {
+ if (gfsh.isConnected()) {
+ gfsh.disconnect();
+ }
+ gfsh.connectAndVerify(locator);
+ }
+
+ void verifyGatewaySenderState(boolean isRunning, boolean isPaused) {
Review comment:
This method always uses `isRunning = true` and `isPaused = false`. For simplicity, t might be better to change the method name to "verifyGatewaySenderIsRunningAndNotPaused" and inline the values.
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
Review comment:
Rather than relying on puts to create buckets, I think you can instead use:
```
server1Site2.invoke(() -> PartitionRegionHelper.assignBucketsToPartitions(
Objects.requireNonNull(ClusterStartupRule.getCache()).getRegion(REGION_NAME)));
```
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ // check that bucketToTempQueueMap is empty on all members
+ for (MemberVM member : allMembers) {
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ }
+
+ // Choose server with bucket to be stopped
+ MemberVM serverToStop = getServerToStop(allMembers);
+ int bucketId = serverToStop.invoke(this::getPrimaryBucketList).iterator().next();
+ serverToStop.stop(true);
+
+ // remove from list the member that has been previously stopped
+ allMembers.remove(serverToStop);
+
+ startDurableClient();
+ createDurableCQs("SELECT * FROM " + Region.SEPARATOR + REGION_NAME);
Review comment:
This CQ can be replaced with the `REGION_CQ` constant defined in this class.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] jvarenina commented on a change in pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
jvarenina commented on a change in pull request #6646:
URL: https://github.com/apache/geode/pull/6646#discussion_r661234930
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ // check that bucketToTempQueueMap is empty on all members
+ for (MemberVM member : allMembers) {
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ }
+
+ // Choose server with bucket to be stopped
+ MemberVM serverToStop = getServerToStop(allMembers);
+ int bucketId = serverToStop.invoke(this::getPrimaryBucketList).iterator().next();
+ serverToStop.stop(true);
+
+ // remove from list the member that has been previously stopped
+ allMembers.remove(serverToStop);
+
+ startDurableClient();
+ createDurableCQs("SELECT * FROM " + Region.SEPARATOR + REGION_NAME);
+
+ // configure hook on running members
+ for (MemberVM member : allMembers) {
+ configureHooksOnRunningMember(member, bucketId);
+ }
+
+ // perform rebalance operation to trigger bucket redundancy recovery on running servers
+ String command = new CommandStringBuilder(CliStrings.REBALANCE)
+ .getCommandString();
+ gfsh.executeAndAssertThat(command).statusIsSuccess();
+
+ for (MemberVM member : allMembers) {
+ // All members after redundancy is recovered should have empty temporary queue
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ // If hook has been triggered on member, then check if member temporarily queued
+ // events while getting initial image from primary server
+ if (member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isHookTriggered)) {
+ assertThat(member.invoke(ParallelGatewaySenderAndCQDurableClientDUnitTest::isTempQueueUsed))
+ .isTrue();
+ }
+ }
+
+ // reset all hooks
+ for (MemberVM member : allMembers) {
+ resetHooks(member);
+ }
+ }
+
+ /**
+ * This hook will run some transaction traffic after server requested initial image from
+ * primary bucket server, and then it will check if events are enqueued in temporary queue.
+ * It is expected that events are enqueued in this case because bucket is not yet available.
+ */
+ void configureHooksOnRunningMember(MemberVM server, int bucketId) {
+ server.invoke(() -> InitialImageOperation.setGIITestHook(new InitialImageOperation.GIITestHook(
+ InitialImageOperation.GIITestHookType.BeforeRequestRVV,
+ "_B__" + REGION_NAME + "_" + bucketId) {
+ private static final long serialVersionUID = -3790198435185240444L;
+
+ @Override
+ public void reset() {}
+
+ @Override
+ public void run() {
+ doPutsInServer();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED = true;
+ if (sizeOfBucketToTempQueueMap("ln") != 0) {
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED = true;
+ }
+ }
+ }));
+ }
+
+ void resetHooks(MemberVM server) {
+ server.invoke(() -> InitialImageOperation.resetAllGIITestHooks());
+ }
+
+ public static boolean isHookTriggered() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_HOOK_TRIGGERED;
+ }
+
+ public static boolean isTempQueueUsed() {
+ return ParallelGatewaySenderAndCQDurableClientDUnitTest.IS_TEMP_QUEUE_USED;
+ }
+
+ MemberVM getServerToStop(List<MemberVM> list) {
+ for (MemberVM member : list) {
+ if (!member.invoke(this::getPrimaryBucketList).isEmpty()) {
+ return member;
+ }
+ }
+ return null;
+ }
+
+ private void startDurableClient()
+ throws Exception {
+ int locatorPort = locatorSite2.getPort();
+ clientSite2DurableSubscription = clusterStartupRule.startClientVM(7, ccf -> ccf
+ .withPoolSubscription(true).withLocatorConnection(locatorPort).withCacheSetup(c -> c
+ .set("durable-client-id", DURABLE_CLIENT_ID)));
+ }
+
+ private void createDurableCQs(String... queries) {
+ clientSite2DurableSubscription.invoke(() -> {
+ assertThat(ClusterStartupRule.getClientCache()).isNotNull();
+ QueryService queryService = ClusterStartupRule.getClientCache().getQueryService();
+ CqAttributesFactory cqAttributesFactory = new CqAttributesFactory();
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.cqListener =
+ new CqListenerTestReceivedEvents();
+ cqAttributesFactory.addCqListener(cqListener);
+
+ for (String query : queries) {
+ CqQuery cq = queryService.newCq(query, cqAttributesFactory.create(), true);
+ cq.execute();
+ }
+ ClusterStartupRule.getClientCache().readyForEvents();
+ });
+ }
+
+ private Set<Integer> getPrimaryBucketList() {
+ PartitionedRegion region = (PartitionedRegion) Objects
+ .requireNonNull(ClusterStartupRule.getCache()).getRegion(REGION_NAME);
+ return new HashSet<>(region.getDataStore().getAllLocalPrimaryBucketIds());
+ }
+
+ /**
+ * Check that all events are received on durable client that registered cq's
+ */
+ private void checkCqEvents(int expectedNumberOfEvents) {
+ // Check if number of events is correct
+ clientSite2DurableSubscription.invoke(() -> await().untilAsserted(() -> assertThat(
+ ParallelGatewaySenderAndCQDurableClientDUnitTest.cqListener.getNumEvents())
+ .isEqualTo(expectedNumberOfEvents)));
+ }
+
+ /**
+ * Checks that the bucketToTempQueueMap for a partitioned region
+ * that holds events for buckets that are not available locally, is empty.
+ */
+ public static void validateBucketToTempQueueMap(String senderId, boolean shouldBeEmpty) {
Review comment:
Done
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] jvarenina commented on a change in pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
jvarenina commented on a change in pull request #6646:
URL: https://github.com/apache/geode/pull/6646#discussion_r661234192
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
+
+ @Rule
+ public transient GfshCommandRule gfsh = new GfshCommandRule();
+
+ private static CqListenerTestReceivedEvents cqListener;
+
+ private MemberVM locatorSite2;
+ private MemberVM server1Site2;
+ private MemberVM server2Site2;
+ private MemberVM server3Site2;
+
+ private ClientVM clientSite2;
+ private ClientVM clientSite2DurableSubscription;
+
+ public static boolean IS_TEMP_QUEUE_USED = false;
+ public static boolean IS_HOOK_TRIGGERED = false;
+
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE1 = "1";
+ private static final String DISTRIBUTED_SYSTEM_ID_SITE2 = "2";
+ private static final String REGION_NAME = "test1";
+ private static final String REGION_COLOCATED = "colocatedRegion";
+ private static final String REGION_CQ = "SELECT * FROM " + Region.SEPARATOR + REGION_NAME;
+ private static final String COLO_REGION_CQ =
+ "SELECT * FROM " + Region.SEPARATOR + REGION_COLOCATED;
+
+ /**
+ * Issue reproduces when following conditions are fulfilled:
+ * - Redundant partition region must configured
+ * - Number of servers must be greater than number of redundant copies of partition region
+ * - Parallel gateway sender must be configured on partition region
+ * - Client must register CQs for the region
+ * - Transactions must be used with put operations
+ * - Events must enqueued in parallel gateway senders (remote site is unavailable)
+ *
+ * Server that is hosting primary bucket will send TXCommitMessage to the server that
+ * is hosting secondary bucket, and also to the server that is hosting CQ subscription
+ * queue (if CQ condition is fulfilled). The problem is that server which is hosting CQ
+ * subscription queue is not hosting the bucket for which event it actually intended.
+ * In this case the server will store this event in bucketToTempQueueMap because it assumes
+ * that the bucket is in the process of the creation, which is not correct.
+ */
+ @Test
+ public void testSubscriptionQueueWan() throws Exception {
+ configureSites("113", false);
+ startDurableClient();
+ createDurableCQs(REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionSingleKey(0, 30, REGION_NAME));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size());
+ }
+
+ @Test
+ public void testSubscriptionQueueWanColocatedRegionsMultipleOperations() throws Exception {
+ configureSites("113", true);
+ startDurableClient();
+ createDurableCQs(REGION_CQ, COLO_REGION_CQ);
+
+ verifyGatewaySenderState(true, false);
+
+ // Do some puts and check that data has been enqueued
+ Set<String> keysQueue =
+ clientSite2.invoke(() -> doPutsInRangeTransactionWithTwoPutOperation(0, 30));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size() * 2));
+
+ server1Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server2Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ server3Site2.invoke(() -> validateBucketToTempQueueMap("ln", true));
+
+ // Check that durable client has received all events
+ checkCqEvents(keysQueue.size() * 2);
+ }
+
+ /**
+ * This test case verifies that the server during bucket recovery enqueues all events intended for
+ * that bucket in temporary queue, and that after bucket redundancy is restored events are
+ * transferred from temporary queue to real bucket queue.
+ */
+ @Test
+ public void testSubscriptionQueueWanTrafficWhileRebalanced() throws Exception {
+ configureSites("3", false);
+ verifyGatewaySenderState(true, false);
+
+ List<MemberVM> allMembers = new ArrayList<>();
+ allMembers.add(server1Site2);
+ allMembers.add(server2Site2);
+ allMembers.add(server3Site2);
+
+ // Do some puts so that all bucket are created
+ Set<String> keysQueue = clientSite2.invoke(() -> doPutsInRange(0, 100));
+ server1Site2.invoke(() -> checkQueueSize("ln", keysQueue.size()));
+
+ // check that bucketToTempQueueMap is empty on all members
+ for (MemberVM member : allMembers) {
+ member.invoke(() -> validateBucketToTempQueueMap("ln", true));
+ }
+
+ // Choose server with bucket to be stopped
+ MemberVM serverToStop = getServerToStop(allMembers);
+ int bucketId = serverToStop.invoke(this::getPrimaryBucketList).iterator().next();
+ serverToStop.stop(true);
+
+ // remove from list the member that has been previously stopped
+ allMembers.remove(serverToStop);
+
+ startDurableClient();
+ createDurableCQs("SELECT * FROM " + Region.SEPARATOR + REGION_NAME);
Review comment:
Done
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
[GitHub] [geode] jvarenina commented on a change in pull request #6646: GEODE-9248: Server hosting CQ queue uneccessary fills bucketToTempQueue
Posted by GitBox <gi...@apache.org>.
jvarenina commented on a change in pull request #6646:
URL: https://github.com/apache/geode/pull/6646#discussion_r661235477
##########
File path: geode-wan/src/distributedTest/java/org/apache/geode/internal/cache/wan/parallel/ParallelGatewaySenderAndCQDurableClientDUnitTest.java
##########
@@ -0,0 +1,530 @@
+/*
+ * 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
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * 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.geode.internal.cache.wan.parallel;
+
+import static org.apache.geode.distributed.ConfigurationProperties.DISTRIBUTED_SYSTEM_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.DURABLE_CLIENT_ID;
+import static org.apache.geode.distributed.ConfigurationProperties.REMOTE_LOCATORS;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.getMember;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.validateGatewaySenderMXBeanProxy;
+import static org.apache.geode.internal.cache.wan.wancommand.WANCommandUtils.verifySenderState;
+import static org.apache.geode.test.awaitility.GeodeAwaitility.await;
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.Serializable;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Objects;
+import java.util.Properties;
+import java.util.Set;
+import java.util.concurrent.BlockingQueue;
+
+import org.junit.Rule;
+import org.junit.Test;
+import org.junit.experimental.categories.Category;
+
+import org.apache.geode.cache.CacheTransactionManager;
+import org.apache.geode.cache.Region;
+import org.apache.geode.cache.query.CqAttributesFactory;
+import org.apache.geode.cache.query.CqEvent;
+import org.apache.geode.cache.query.CqListener;
+import org.apache.geode.cache.query.CqQuery;
+import org.apache.geode.cache.query.QueryService;
+import org.apache.geode.cache.wan.GatewaySender;
+import org.apache.geode.internal.cache.InitialImageOperation;
+import org.apache.geode.internal.cache.InternalCache;
+import org.apache.geode.internal.cache.PartitionedRegion;
+import org.apache.geode.internal.cache.RegionQueue;
+import org.apache.geode.internal.cache.wan.AbstractGatewaySender;
+import org.apache.geode.internal.cache.wan.GatewaySenderEventImpl;
+import org.apache.geode.management.internal.cli.util.CommandStringBuilder;
+import org.apache.geode.management.internal.i18n.CliStrings;
+import org.apache.geode.test.dunit.rules.ClientVM;
+import org.apache.geode.test.dunit.rules.ClusterStartupRule;
+import org.apache.geode.test.dunit.rules.MemberVM;
+import org.apache.geode.test.junit.categories.WanTest;
+import org.apache.geode.test.junit.rules.GfshCommandRule;
+
+@Category({WanTest.class})
+public class ParallelGatewaySenderAndCQDurableClientDUnitTest implements Serializable {
+
+ @Rule
+ public ClusterStartupRule clusterStartupRule = new ClusterStartupRule(9);
Review comment:
Changed it to 7. Thanks.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: notifications-unsubscribe@geode.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org