kevin-wu24 commented on code in PR #19589: URL: https://github.com/apache/kafka/pull/19589#discussion_r2254740560
########## raft/src/test/java/org/apache/kafka/raft/KafkaRaftClientAutoJoinTest.java: ########## @@ -0,0 +1,323 @@ +/* + * 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.kafka.raft; + +import org.apache.kafka.common.protocol.Errors; +import org.apache.kafka.common.record.MemoryRecords; +import org.apache.kafka.common.utils.BufferSupplier; +import org.apache.kafka.server.common.KRaftVersion; + +import org.junit.jupiter.api.Test; + +import java.util.concurrent.ThreadLocalRandom; +import java.util.stream.Stream; + +import static org.apache.kafka.raft.KafkaRaftClientTest.replicaKey; +import static org.apache.kafka.raft.RaftClientTestContext.RaftProtocol.KIP_595_PROTOCOL; +import static org.apache.kafka.raft.RaftClientTestContext.RaftProtocol.KIP_853_PROTOCOL; + +public class KafkaRaftClientAutoJoinTest { + @Test + public void testAutoRemoveOldVoter() throws Exception { + final var leader = replicaKey(randomReplicaId(), true); + final var oldFollower = replicaKey(leader.id() + 1, true); + final var newFollowerKey = replicaKey(oldFollower.id(), true); + final int epoch = 1; + final var context = new RaftClientTestContext.Builder( + newFollowerKey.id(), + newFollowerKey.directoryId().get() + ) + .withRaftProtocol(KIP_853_PROTOCOL) + .withStartingVoters( + VoterSetTest.voterSet(Stream.of(leader, oldFollower)), KRaftVersion.KRAFT_VERSION_1 + ) + .withElectedLeader(epoch, leader.id()) + .withAutoJoin(true) + .withCanBecomeVoter(true) + .build(); + + context.advanceTimeAndCompleteFetch(epoch, leader.id(), true); + + // the next request should be a remove voter request + pollAndDeliverRemoveVoter(context, oldFollower); + + // after sending a remove voter the next request should be a fetch + context.advanceTimeAndCompleteFetch(epoch, leader.id(), true); + + // the replica should send remove voter again because the fetch did not update the voter set + pollAndDeliverRemoveVoter(context, oldFollower); + } + + @Test + public void testAutoAddNewVoter() throws Exception { + final var leader = replicaKey(randomReplicaId(), true); + final var follower = replicaKey(leader.id() + 1, true); + final var newVoter = replicaKey(follower.id() + 1, true); + final int epoch = 1; + final var context = new RaftClientTestContext.Builder( + newVoter.id(), + newVoter.directoryId().get() + ) + .withRaftProtocol(KIP_853_PROTOCOL) + .withStartingVoters( + VoterSetTest.voterSet(Stream.of(leader, follower)), KRaftVersion.KRAFT_VERSION_1 + ) + .withElectedLeader(epoch, leader.id()) + .withAutoJoin(true) + .withCanBecomeVoter(true) + .build(); + + context.advanceTimeAndCompleteFetch(epoch, leader.id(), true); + + // the next request should be an add voter request + pollAndSendAddVoter(context, newVoter); + + // expire the add voter request, the next request should be a fetch + context.advanceTimeAndCompleteFetch(epoch, leader.id(), true); + + // the replica should send add voter again because the completed fetch + // did not update the voter set, and its timer has expired + final var addVoterRequest = pollAndSendAddVoter(context, newVoter); + + // deliver the add voter response, this is possible before a completed fetch because of KIP-1186 + context.deliverResponse( + addVoterRequest.correlationId(), + addVoterRequest.destination(), + RaftUtil.addVoterResponse(Errors.NONE, Errors.NONE.message()) + ); + + // verify the replica can perform a fetch to commit the new voter set while the add + // voter request is still in-flight + pollAndDeliverFetchToUpdateVoterSet(context, epoch, + VoterSetTest.voterSet(Stream.of(leader, newVoter))); + } + + @Test + public void testObserverRemovesOldVoterAndAutoJoins() throws Exception { + final var leader = replicaKey(randomReplicaId(), true); + final var oldFollower = replicaKey(leader.id() + 1, true); + final var newFollowerKey = replicaKey(oldFollower.id(), true); + final int epoch = 1; + final var context = new RaftClientTestContext.Builder( + newFollowerKey.id(), + newFollowerKey.directoryId().get() + ) + .withRaftProtocol(KIP_853_PROTOCOL) + .withStartingVoters( + VoterSetTest.voterSet(Stream.of(leader, oldFollower)), KRaftVersion.KRAFT_VERSION_1 + ) + .withElectedLeader(epoch, leader.id()) + .withAutoJoin(true) + .withCanBecomeVoter(true) + .build(); + + // advance time and complete a fetch to trigger the remove voter request + context.advanceTimeAndCompleteFetch(epoch, leader.id(), true); + + // the next request should be a remove voter request + pollAndDeliverRemoveVoter(context, oldFollower); + + // after sending a remove voter the next request should be a fetch + // this fetch will remove the old follower from the voter set + pollAndDeliverFetchToUpdateVoterSet( + context, + epoch, + VoterSetTest.voterSet(Stream.of(leader)) + ); + + // advance time and complete a fetch to trigger the add voter request + context.advanceTimeAndCompleteFetch(epoch, leader.id(), true); + + // the next request should be an add voter request + final var addVoterRequest = pollAndSendAddVoter(context, newFollowerKey); + + // deliver the add voter response, this is possible before a completed fetch because of KIP-1186 + context.deliverResponse( + addVoterRequest.correlationId(), + addVoterRequest.destination(), + RaftUtil.addVoterResponse(Errors.NONE, Errors.NONE.message()) + ); + + // verify the replica can perform a fetch to commit the new voter set while the add + // voter request is still in-flight + pollAndDeliverFetchToUpdateVoterSet( + context, + epoch, + VoterSetTest.voterSet(Stream.of(leader, newFollowerKey)) + ); + + // advance time and complete a fetch and expire the update voter set timer + // the next request should be a fetch because the log voter configuration is up-to-date + context.advanceTimeAndCompleteFetch(epoch, leader.id(), true); + context.pollUntilRequest(); + context.assertSentFetchRequest(); + } + + + @Test + public void testObserversDoNotAutoJoin() throws Exception { + final var leader = replicaKey(randomReplicaId(), true); + final var follower = replicaKey(leader.id() + 1, true); + final var newObserver = replicaKey(follower.id() + 1, true); + final int epoch = 1; + final var context = new RaftClientTestContext.Builder( + newObserver.id(), + newObserver.directoryId().get() + ) + .withRaftProtocol(KIP_853_PROTOCOL) + .withStartingVoters( + VoterSetTest.voterSet(Stream.of(leader, follower)), KRaftVersion.KRAFT_VERSION_1 + ) + .withElectedLeader(epoch, leader.id()) + .withAutoJoin(true) + .withCanBecomeVoter(false) + .build(); + + context.advanceTimeAndCompleteFetch(epoch, leader.id(), true); + + context.time.sleep(context.fetchTimeoutMs - 1); + context.pollUntilRequest(); + + // When canBecomeVoter == false, the replica should not send an add voter request + final var fetchRequest = context.assertSentFetchRequest(); + context.assertFetchRequestData(fetchRequest, epoch, 0L, 0, context.client.highWatermark()); + } + + @Test + public void testObserverDoesNotAddItselfWhenAutoJoinDisabled() throws Exception { + final var leader = replicaKey(randomReplicaId(), true); + final var follower = replicaKey(leader.id() + 1, true); + final var observer = replicaKey(follower.id() + 1, true); + final int epoch = 1; + final var context = new RaftClientTestContext.Builder( + observer.id(), + observer.directoryId().get() + ) + .withRaftProtocol(KIP_853_PROTOCOL) + .withStartingVoters( + VoterSetTest.voterSet(Stream.of(leader, follower)), KRaftVersion.KRAFT_VERSION_1 + ) + .withElectedLeader(epoch, leader.id()) + .withAutoJoin(false) + .withCanBecomeVoter(true) + .build(); + + context.advanceTimeAndCompleteFetch(epoch, leader.id(), true); + + context.time.sleep(context.fetchTimeoutMs - 1); + context.pollUntilRequest(); + + // When autoJoin == false, the replica should not send an add voter request + final var fetchRequest = context.assertSentFetchRequest(); + context.assertFetchRequestData(fetchRequest, epoch, 0L, 0, context.client.highWatermark()); + } + + @Test + public void testObserverDoesNotAutoJoinWithKRaftVersion0() throws Exception { + final var leader = replicaKey(randomReplicaId(), true); + final var follower = replicaKey(leader.id() + 1, true); + final var observer = replicaKey(follower.id() + 1, true); + final int epoch = 1; + final var context = new RaftClientTestContext.Builder( + observer.id(), + observer.directoryId().get() + ) + .withRaftProtocol(KIP_595_PROTOCOL) + .withStartingVoters( + VoterSetTest.voterSet(Stream.of(leader, follower)), KRaftVersion.KRAFT_VERSION_0 + ) + .withElectedLeader(epoch, leader.id()) + .withAutoJoin(true) + .withCanBecomeVoter(true) + .build(); + + context.advanceTimeAndCompleteFetch(epoch, leader.id(), true); + + context.time.sleep(context.fetchTimeoutMs - 1); + context.pollUntilRequest(); + + // When kraft.version == 0, the replica should not send an add voter request + final var fetchRequest = context.assertSentFetchRequest(); + + context.assertFetchRequestData(fetchRequest, epoch, 0L, 0, context.client.highWatermark()); + } + + private void pollAndDeliverRemoveVoter( + RaftClientTestContext context, + ReplicaKey oldFollower + ) throws Exception { + context.pollUntilRequest(); + final var removeRequest = context.assertSentRemoveVoterRequest(oldFollower); + context.deliverResponse( + removeRequest.correlationId(), + removeRequest.destination(), + RaftUtil.removeVoterResponse(Errors.NONE, Errors.NONE.message()) + ); + } + + private RaftRequest.Outbound pollAndSendAddVoter( + RaftClientTestContext context, + ReplicaKey newVoter + ) throws Exception { + context.pollUntilRequest(); + return context.assertSentAddVoterRequest( + newVoter, + context.client.quorum().localVoterNodeOrThrow().listeners(), + false + ); + } + + private void pollAndDeliverFetchToUpdateVoterSet( + RaftClientTestContext context, + int epoch, + VoterSet newVoterSet + ) throws Exception { + context.pollUntilRequest(); + final var fetchRequest = context.assertSentFetchRequest(); + context.assertFetchRequestData( + fetchRequest, + epoch, + context.log.endOffset().offset(), + context.log.lastFetchedEpoch(), + context.client.highWatermark() + ); + // deliver the fetch response with the updated voter set + context.deliverResponse( + fetchRequest.correlationId(), + fetchRequest.destination(), + context.fetchResponse( + epoch, + fetchRequest.destination().id(), + MemoryRecords.withVotersRecord( + context.log.endOffset().offset(), + 0, + epoch, + BufferSupplier.NO_CACHING.get(300), + newVoterSet.toVotersRecord((short) 0) + ), + context.log.endOffset().offset() + 1, Review Comment: Why does this assume the voter set only has one voter? Which voter set are you referencing? `testObserverRemovesOldVoterAndAutoJoins` has the voter set go from size 2, to size 1, and then back to size 2 by having a follower node complete the whole "auto-join" flow (i.e. remove its old self and and its new self to the voter set). -- 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: jira-unsubscr...@kafka.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
