smjn commented on code in PR #17270: URL: https://github.com/apache/kafka/pull/17270#discussion_r1817311509
########## share/src/main/java/org/apache/kafka/server/share/persister/PersisterStateManager.java: ########## @@ -0,0 +1,965 @@ +/* + * 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.server.share.persister; + +import org.apache.kafka.clients.ClientResponse; +import org.apache.kafka.clients.CommonClientConfigs; +import org.apache.kafka.clients.KafkaClient; +import org.apache.kafka.clients.RequestCompletionHandler; +import org.apache.kafka.common.Node; +import org.apache.kafka.common.Uuid; +import org.apache.kafka.common.internals.Topic; +import org.apache.kafka.common.message.FindCoordinatorRequestData; +import org.apache.kafka.common.message.FindCoordinatorResponseData; +import org.apache.kafka.common.message.ReadShareGroupStateRequestData; +import org.apache.kafka.common.message.ReadShareGroupStateResponseData; +import org.apache.kafka.common.message.WriteShareGroupStateRequestData; +import org.apache.kafka.common.message.WriteShareGroupStateResponseData; +import org.apache.kafka.common.protocol.ApiKeys; +import org.apache.kafka.common.protocol.Errors; +import org.apache.kafka.common.requests.AbstractRequest; +import org.apache.kafka.common.requests.AbstractResponse; +import org.apache.kafka.common.requests.FindCoordinatorRequest; +import org.apache.kafka.common.requests.FindCoordinatorResponse; +import org.apache.kafka.common.requests.ReadShareGroupStateRequest; +import org.apache.kafka.common.requests.ReadShareGroupStateResponse; +import org.apache.kafka.common.requests.WriteShareGroupStateRequest; +import org.apache.kafka.common.requests.WriteShareGroupStateResponse; +import org.apache.kafka.common.utils.ExponentialBackoff; +import org.apache.kafka.common.utils.Time; +import org.apache.kafka.common.utils.Utils; +import org.apache.kafka.server.share.SharePartitionKey; +import org.apache.kafka.server.util.InterBrokerSendThread; +import org.apache.kafka.server.util.RequestAndCompletionHandler; +import org.apache.kafka.server.util.timer.Timer; +import org.apache.kafka.server.util.timer.TimerTask; + +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +import java.util.ArrayList; +import java.util.Collection; +import java.util.Collections; +import java.util.HashMap; +import java.util.HashSet; +import java.util.LinkedList; +import java.util.List; +import java.util.Map; +import java.util.Optional; +import java.util.Random; +import java.util.Set; +import java.util.concurrent.CompletableFuture; +import java.util.concurrent.ConcurrentLinkedQueue; +import java.util.concurrent.atomic.AtomicBoolean; +import java.util.function.Consumer; +import java.util.stream.Collectors; + +/** + * This class encapsulates various handler classes corresponding to share + * state RPCs. It also holds an {@link InterBrokerSendThread} specialization + * which manages the sending the RPC requests over the network. + * This class is for the exclusive purpose of being used with {@link DefaultStatePersister} + * but can be extended for other {@link Persister} implementations as well. + */ +public class PersisterStateManager { Review Comment: 1. Handler objects remain in scope until the request is completed due to normal response or error. The error can happen, if we get a non-retryable error from the API or max retry attempts are exceeded while retrying a retryable error. 2. no they should not be reordered - the requests will ultimately end up in linkedlists (the ultimate value part of the nested map). The coord+rpc+group are just lookups. Whats inside them is definitely ordered (via linkedlist) 3. They will be sent to the wrong coord - but then retry logic should kick in. 4. InterbrokerSendThread maintains internal kafkaclient which is periodically invoked. It is not very active though. Hence wakeup calls have been added after specific events to force the client to fetch new requests from send thread, if it can. All other specializations of interbrokersend thread in the code base either use the event based (wakeup) strategy or keep calling poll method periodically which does the same thing. Using wakeup removes the need of sleeping. -- 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]
