| /* |
| * 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.cassandra.analytics.movement; |
| |
| import java.util.Collections; |
| import java.util.Map; |
| import java.util.Optional; |
| import java.util.Set; |
| import java.util.concurrent.CountDownLatch; |
| import java.util.concurrent.TimeUnit; |
| import java.util.stream.Stream; |
| |
| import org.junit.jupiter.params.provider.Arguments; |
| |
| import org.apache.cassandra.analytics.DataGenerationUtils; |
| import org.apache.cassandra.analytics.ResiliencyTestBase; |
| import org.apache.cassandra.analytics.TestConsistencyLevel; |
| import org.apache.cassandra.analytics.TestUninterruptibles; |
| import org.apache.cassandra.distributed.api.ICluster; |
| import org.apache.cassandra.distributed.api.IInstance; |
| import org.apache.cassandra.sidecar.testing.QualifiedName; |
| import org.apache.cassandra.spark.bulkwriter.WriterOptions; |
| import org.apache.cassandra.testing.ClusterBuilderConfiguration; |
| import org.apache.cassandra.testing.utils.ClusterUtils; |
| import org.apache.spark.sql.Dataset; |
| import org.apache.spark.sql.Row; |
| import org.apache.spark.sql.SparkSession; |
| |
| import static org.apache.cassandra.testing.TestUtils.ROW_COUNT; |
| import static org.apache.cassandra.testing.TestUtils.TEST_KEYSPACE; |
| import static org.assertj.core.api.Assertions.assertThat; |
| |
| abstract class NodeMovementTestBase extends ResiliencyTestBase |
| { |
| public static final int SINGLE_DC_MOVING_NODE_IDX = 5; |
| public static final int MULTI_DC_MOVING_NODE_IDX = 3; |
| |
| IInstance movingNode; |
| Dataset<Row> df; |
| Map<? extends IInstance, Set<String>> expectedInstanceData; |
| |
| protected void runMovingNodeTest(TestConsistencyLevel cl) |
| { |
| QualifiedName table = uniqueTestTableFullName(TEST_KEYSPACE, cl.readCL, cl.writeCL); |
| bulkWriterDataFrameWriter(df, table).option(WriterOptions.BULK_WRITER_CL.name(), cl.writeCL.name()) |
| .save(); |
| // validate data right after bulk writes |
| validateData(table, cl.readCL, ROW_COUNT); |
| validateNodeSpecificData(table, expectedInstanceData, false); |
| } |
| |
| @Override |
| protected void beforeTestStart() |
| { |
| super.beforeTestStart(); |
| SparkSession spark = getOrCreateSparkSession(); |
| // Generate some artificial data for the test |
| df = DataGenerationUtils.generateCourseData(spark, ROW_COUNT); |
| // generate the expected data for the moving instance |
| expectedInstanceData = generateExpectedInstanceData(cluster, Collections.singletonList(movingNode), ROW_COUNT); |
| } |
| |
| @Override |
| protected void afterClusterProvisioned() |
| { |
| ClusterBuilderConfiguration configuration = testClusterConfiguration(); |
| int movingNodeIndex = configuration.dcCount > 1 ? MULTI_DC_MOVING_NODE_IDX : SINGLE_DC_MOVING_NODE_IDX; |
| movingNode = cluster.get(movingNodeIndex); |
| |
| IInstance seed = cluster.get(1); |
| new Thread(() -> { |
| long moveTarget = calculateMoveTargetToken(cluster, configuration.dcCount); |
| movingNode.nodetoolResult("move", "--", Long.toString(moveTarget)).asserts().success(); |
| }).start(); |
| |
| // Wait until nodes have reached expected state |
| TestUninterruptibles.awaitUninterruptiblyOrThrow(transitioningStateStart(), 2, TimeUnit.MINUTES); |
| cluster.awaitRingState(seed, movingNode, "Moving"); |
| } |
| |
| /** |
| * @return a latch to wait before the cluster provisioning is complete |
| */ |
| protected abstract CountDownLatch transitioningStateStart(); |
| |
| protected void completeTransitionAndValidateWrites(CountDownLatch transitionalStateEnd, |
| Stream<Arguments> testInputs, |
| boolean expectFailure) |
| { |
| transitionalStateEnd.countDown(); |
| |
| assertThat(movingNode).isNotNull(); |
| |
| // It is only in successful MOVE operation that we validate that the node has reached NORMAL state |
| if (!expectFailure) |
| { |
| cluster.awaitRingState(cluster.get(1), movingNode, "Normal"); |
| } |
| |
| testInputs.forEach(arguments -> { |
| TestConsistencyLevel cl = (TestConsistencyLevel) arguments.get()[0]; |
| |
| QualifiedName tableName = uniqueTestTableFullName(TEST_KEYSPACE, cl.readCL, cl.writeCL); |
| validateData(tableName, cl.readCL, ROW_COUNT); |
| validateNodeSpecificData(tableName, expectedInstanceData, false); |
| }); |
| |
| // For tests that involve MOVE failures, we make a best-effort attempt by checking if the node is either |
| // still MOVING or has flipped back to NORMAL state with the initial token that it previously held |
| if (expectFailure) |
| { |
| String initialToken = movingNode.config().getString("initial_token"); |
| Optional<ClusterUtils.RingInstanceDetails> movingInstance = |
| ClusterUtils.ring(cluster.get(1)) |
| .stream() |
| .filter(i -> i.getAddress().equals(movingNode.broadcastAddress().getAddress().getHostAddress())) |
| .findFirst(); |
| assertThat(movingInstance).isPresent(); |
| String state = movingInstance.get().getState(); |
| |
| assertThat(state.equals("Moving") || |
| (state.equals("Normal") && movingInstance.get().getToken().equals(initialToken))).isTrue(); |
| } |
| } |
| |
| static long calculateMoveTargetToken(ICluster<? extends IInstance> cluster, int dcCount) |
| { |
| IInstance seed = cluster.get(1); |
| // The target token to move the node to is calculated by adding an offset to the seed node token which |
| // is half of the range between 2 tokens. |
| // For multi-DC case (specifically 2 DCs), since neighbouring tokens can be consecutive, we use tokens 1 |
| // and 3 to calculate the offset |
| int nextIndex = (dcCount > 1) ? 3 : 2; |
| long t2 = Long.parseLong(seed.config().getString("initial_token")); |
| long t3 = Long.parseLong(cluster.get(nextIndex).config().getString("initial_token")); |
| return (t2 + ((t3 - t2) / 2)); |
| } |
| } |