oleg-vlsk commented on code in PR #13216: URL: https://github.com/apache/ignite/pull/13216#discussion_r3376599069
########## modules/core/src/test/java/org/apache/ignite/internal/processors/cache/query/continuous/CacheContinuousQueryLocalTxRollbackTest.java: ########## @@ -0,0 +1,263 @@ +/* + * 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.ignite.internal.processors.cache.query.continuous; + +import java.util.Map; +import java.util.concurrent.ThreadLocalRandom; +import java.util.concurrent.atomic.AtomicBoolean; +import java.util.concurrent.atomic.AtomicInteger; +import java.util.function.Function; +import java.util.stream.Collectors; +import java.util.stream.IntStream; +import javax.cache.Cache; +import javax.cache.event.CacheEntryListenerException; +import javax.cache.event.CacheEntryUpdatedListener; +import org.apache.ignite.IgniteCache; +import org.apache.ignite.cache.CacheAtomicityMode; +import org.apache.ignite.cache.CacheMode; +import org.apache.ignite.cache.CacheWriteSynchronizationMode; +import org.apache.ignite.cache.query.ContinuousQuery; +import org.apache.ignite.cache.query.QueryCursor; +import org.apache.ignite.cluster.ClusterState; +import org.apache.ignite.configuration.CacheConfiguration; +import org.apache.ignite.configuration.DataRegionConfiguration; +import org.apache.ignite.configuration.DataStorageConfiguration; +import org.apache.ignite.configuration.IgniteConfiguration; +import org.apache.ignite.failure.StopNodeFailureHandler; +import org.apache.ignite.internal.GridKernalState; +import org.apache.ignite.internal.IgniteEx; +import org.apache.ignite.internal.IgniteInternalFuture; +import org.apache.ignite.spi.discovery.tcp.TcpDiscoverySpi; +import org.apache.ignite.testframework.GridTestUtils; +import org.apache.ignite.testframework.junits.common.GridCommonAbstractTest; +import org.apache.ignite.transactions.Transaction; +import org.apache.ignite.transactions.TransactionConcurrency; +import org.apache.ignite.transactions.TransactionIsolation; +import org.junit.Test; +import org.junit.runner.RunWith; +import org.junit.runners.Parameterized; + +import static org.apache.ignite.cache.CacheMode.PARTITIONED; +import static org.apache.ignite.cache.CacheMode.REPLICATED; +import static org.apache.ignite.testframework.GridTestUtils.waitForCondition; + +/** + * Tests interaction between local continuous queries created with {@link ContinuousQuery#setLocal(boolean)} set to + * {@code true} and transaction rollback counter cleanup. + * <p> + * When a transactional update is rolled back, {@code IgniteTxHandler} closes the corresponding partition update + * counter gaps and calls {@code CacheContinuousQueryManager#skipUpdateCounter} for the rolled back counters. These + * skipped counters are needed by distributed continuous queries to release pending events after gaps caused by + * rolled back updates. + * <p> + * Local continuous queries are different: their regular update events are delivered directly to the local listener and + * do not use the distributed partition recovery path based on skipped update counters. Therefore, local-only continuous + * query listeners must not be passed into {@code skipUpdateCounter}. Otherwise, the local-only handler may reach the + * unsupported {@code loc == true && locOnly == true} path in + * {@code CacheContinuousQueryHandler.CacheContinuousQueryListener#skipUpdateCounter}. + * <p> + * This class verifies that local continuous queries are ignored by skipped-counter processing, while ordinary + * distributed continuous queries continue to work correctly. + */ +@RunWith(Parameterized.class) +public class CacheContinuousQueryLocalTxRollbackTest extends GridCommonAbstractTest { + /** */ + private static final int NODE_CNT = 3; + + /** */ + private static final int TX_THREADS = 10; + + /** */ + private static final int KEYS_PER_TX = 10; + + /** */ + @Parameterized.Parameter + public CacheMode cacheMode; + + /** */ + @Parameterized.Parameters(name = "cacheMode={0}") + public static Object[] params() { + return new Object[] {REPLICATED, PARTITIONED}; + } + + /** {@inheritDoc} */ + @Override protected IgniteConfiguration getConfiguration(String igniteInstanceName) throws Exception { + CacheConfiguration<?, ?> cacheCfg = new CacheConfiguration<>(DEFAULT_CACHE_NAME) + .setCacheMode(cacheMode) + .setWriteSynchronizationMode(CacheWriteSynchronizationMode.FULL_SYNC) + .setAtomicityMode(CacheAtomicityMode.TRANSACTIONAL); + + if (cacheMode == PARTITIONED) + cacheCfg.setBackups(NODE_CNT - 1); + + return super.getConfiguration(igniteInstanceName) + .setDataStorageConfiguration(new DataStorageConfiguration() + .setDefaultDataRegionConfiguration(new DataRegionConfiguration() + .setPersistenceEnabled(true))) + .setFailureHandler(new StopNodeFailureHandler()) + .setCacheConfiguration(cacheCfg); + } + + /** {@inheritDoc} */ + @Override protected void beforeTest() throws Exception { + super.beforeTest(); + + cleanPersistenceDir(); + } + + /** {@inheritDoc} */ + @Override protected void afterTest() throws Exception { + stopAllGrids(); + + cleanPersistenceDir(); + + super.afterTest(); + } + + /** + * Checks that local and distributed continuous queries behave correctly when transaction rollback closes + * partition update counter gaps after a node failure. + */ + @Test + public void checkLocalAndDistributedCqAfterNodeFail() throws Exception { + IgniteEx ignite = startGrids(NODE_CNT); + + ignite.cluster().state(ClusterState.ACTIVE); + + awaitPartitionMapExchange(); + + AtomicBoolean stopTxLoad = new AtomicBoolean(); + AtomicBoolean nodeFailed = new AtomicBoolean(); + + AtomicInteger updatesDistrBeforeFail = new AtomicInteger(); + AtomicInteger updatesLocBeforeFail = new AtomicInteger(); + + AtomicInteger updatesDistrAfterFail = new AtomicInteger(); + AtomicInteger updatesLocAfterFail = new AtomicInteger(); + + IgniteCache<Object, Object> cache1 = grid(1).cache(DEFAULT_CACHE_NAME); + IgniteCache<Object, Object> cache0 = grid(0).cache(DEFAULT_CACHE_NAME); + + IgniteInternalFuture<?> txLoadFut = launchTxLoad(grid(1), cache1, stopTxLoad); + + try ( + QueryCursor<Cache.Entry<Object, Object>> locCur = cache0.query( + buildContinuousQuery(true, nodeFailed, updatesLocBeforeFail, updatesLocAfterFail)); + + QueryCursor<Cache.Entry<Object, Object>> distrCur = cache0.query( + buildContinuousQuery(false, nodeFailed, updatesDistrBeforeFail, updatesDistrAfterFail)) + ) { + assertTrue( + String.format("Failed to receive expected updates before node failure [locUpdates=%s, distrUpdates=%s]", + updatesLocBeforeFail.get(), updatesDistrBeforeFail.get()), + waitForCondition(() -> updatesLocBeforeFail.get() > 0 && updatesDistrBeforeFail.get() > 0, 3_000) + ); + + failNode(NODE_CNT - 1); + + assertFalse("Cluster stopped/stopping after target node failure", + waitForCondition(() -> + ignite.context().gateway().getState() == GridKernalState.STOPPED || + ignite.context().gateway().getState() == GridKernalState.STOPPING, + 10_000) + ); + + assertEquals("Not enogh nodes sirvuved after target node failure", + NODE_CNT - 1, ignite.cluster().nodes().size()); + + for (int i : IntStream.range(0, NODE_CNT - 1).toArray()) { + IgniteEx grid = grid(i); + + assertFalse("Grid " + i + " is stopping", grid.context().isStopping()); + assertNull("Failure context is not null for grid " + i, grid.context().failure().failureContext()); + } + + nodeFailed.set(true); + + assertTrue( + String.format("Failed to receive expected updates after node failure [locUpdates=%s, distrUpdates=%s]", + updatesLocAfterFail.get(), updatesDistrAfterFail.get()), + waitForCondition(() -> updatesLocAfterFail.get() > 0 && updatesDistrAfterFail.get() > 0, 3_000) 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
