lianetm commented on code in PR #14323: URL: https://github.com/apache/kafka/pull/14323#discussion_r1321621168
########## clients/src/main/java/org/apache/kafka/clients/consumer/internals/MembershipManagerImpl.java: ########## @@ -0,0 +1,167 @@ +/* + * 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.clients.consumer.internals; + +import org.apache.kafka.common.message.ConsumerGroupHeartbeatResponseData; +import org.apache.kafka.common.protocol.Errors; + +import java.util.ArrayList; +import java.util.List; +import java.util.Optional; + +/** + * Membership manager that maintains group membership for a single member following the new + * consumer group protocol. + * <p/> + * This keeps membership state and assignment updated in-memory, based on the heartbeat responses + * the member receives. It is also responsible for computing assignment for the group based on + * the metadata, if the member has been selected by the broker to do so. + */ +public class MembershipManagerImpl implements MembershipManager { + + private final String groupId; + private Optional<String> groupInstanceId; + private String memberId; + private int memberEpoch; + private MemberState state; + private AssignorSelection assignorSelection; + + /** + * Assignment that the member received from the server and successfully processed + */ + private ConsumerGroupHeartbeatResponseData.Assignment currentAssignment; + + /** + * List of assignments that the member received from the server but hasn't processed yet + */ + private final List<ConsumerGroupHeartbeatResponseData.Assignment> targetAssignments; Review Comment: Very interesting, we've had some conversations about this (without clear ideas yet) so I need to think about this more. In principle I don't see a reason why we couldn't process the 2 revocations above in parallel, given they relate to different partitions. It feels like the right thing to do, to take advantage of the incremental assignment when possible. That being said, it would definitely be a change in how we execute the callbacks. User's callbacks could be executed concurrently for distinct sets of partitions, which is something that does not happen today (they are invoked sequentially in the poll loop) I will update this initial code to have the clearer var definitions for now: `currentAssignment`, `targetAssignment` and `nextTargetAssignment`. I will continue to think about how to handle assignments/revocations received while processing a previous one, which seems to be where we could consider the trade-offs of starting processing a `nextTargetAssignment` even if there is a `targetAssignment` being processed (when possible) -- 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
