LoganZhuZzz commented on code in PR #20089: URL: https://github.com/apache/kafka/pull/20089#discussion_r2191482373
########## coordinator-common/src/main/java/org/apache/kafka/coordinator/common/runtime/CoordinatorLoaderImpl.java: ########## @@ -0,0 +1,338 @@ +/* + * 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.coordinator.common.runtime; + +import org.apache.kafka.common.TopicPartition; +import org.apache.kafka.common.errors.NotLeaderOrFollowerException; +import org.apache.kafka.common.record.ControlRecordType; +import org.apache.kafka.common.record.FileRecords; +import org.apache.kafka.common.record.MemoryRecords; +import org.apache.kafka.common.record.MutableRecordBatch; +import org.apache.kafka.common.record.Record; +import org.apache.kafka.common.record.Records; +import org.apache.kafka.common.requests.TransactionResult; +import org.apache.kafka.common.utils.Time; +import org.apache.kafka.server.storage.log.FetchIsolation; +import org.apache.kafka.server.util.KafkaScheduler; +import org.apache.kafka.storage.internals.log.FetchDataInfo; +import org.apache.kafka.storage.internals.log.UnifiedLog; + +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +import java.io.IOException; +import java.nio.ByteBuffer; +import java.util.Optional; +import java.util.concurrent.CompletableFuture; +import java.util.concurrent.ScheduledFuture; +import java.util.concurrent.atomic.AtomicBoolean; +import java.util.function.Function; + +/** + * Coordinator loader which reads records from a partition and replays them + * to a group coordinator. + * + * @param <T> The record type. + */ +public class CoordinatorLoaderImpl<T> implements CoordinatorLoader<T> { + + private static final Logger LOG = LoggerFactory.getLogger(CoordinatorLoaderImpl.class); + + private final Time time; + private final Function<TopicPartition, Optional<UnifiedLog>> partitionLogSupplier; + private final Function<TopicPartition, Optional<Long>> partitionLogEndOffsetSupplier; + private final Deserializer<T> deserializer; + private final int loadBufferSize; + + private final AtomicBoolean isRunning = new AtomicBoolean(true); + private final KafkaScheduler scheduler = new KafkaScheduler(1); + + public CoordinatorLoaderImpl( + Time time, + Function<TopicPartition, Optional<UnifiedLog>> partitionLogSupplier, + Function<TopicPartition, Optional<Long>> partitionLogEndOffsetSupplier, + Deserializer<T> deserializer, + int loadBufferSize + ) { + this.time = time; + this.partitionLogSupplier = partitionLogSupplier; + this.partitionLogEndOffsetSupplier = partitionLogEndOffsetSupplier; + this.deserializer = deserializer; + this.loadBufferSize = loadBufferSize; + this.scheduler.startup(); + } + + /** + * Loads the coordinator by reading all the records from the TopicPartition + * and applying them to the Replayable object. + * + * @param tp The TopicPartition to read from. + * @param coordinator The object to apply records to. + */ + @Override + public CompletableFuture<LoadSummary> load(TopicPartition tp, CoordinatorPlayback<T> coordinator) { + final CompletableFuture<LoadSummary> future = new CompletableFuture<>(); + long startTimeMs = time.milliseconds(); + try { + ScheduledFuture<?> result = scheduler.scheduleOnce(String.format("Load coordinator from %s", tp), + () -> doLoad(tp, coordinator, future, startTimeMs)); + if (result.isCancelled()) { + future.completeExceptionally(new RuntimeException("Coordinator loader is closed.")); + } + } catch (Exception e) { + future.completeExceptionally(e); + } + return future; + } + + private void doLoad( + TopicPartition tp, + CoordinatorPlayback<T> coordinator, + CompletableFuture<LoadSummary> future, + long startTimeMs + ) { + long schedulerQueueTimeMs = time.milliseconds() - startTimeMs; + try { + Optional<UnifiedLog> logOpt = partitionLogSupplier.apply(tp); + if (logOpt.isEmpty()) { + future.completeExceptionally(new NotLeaderOrFollowerException( + "Could not load records from " + tp + " because the log does not exist.")); + return; + } + + UnifiedLog log = logOpt.get(); + + // Buffer may not be needed if records are read from memory. + ByteBuffer buffer = ByteBuffer.allocate(0); + long currentOffset = log.logStartOffset(); + LoadStats stats = new LoadStats(); + + long previousHighWatermark = -1L; + while (shouldFetchNextBatch(currentOffset, logEndOffset(tp), stats.readAtLeastOneRecord)) { + FetchDataInfo fetchDataInfo = log.read(currentOffset, loadBufferSize, FetchIsolation.LOG_END, true); + + stats.readAtLeastOneRecord = fetchDataInfo.records.sizeInBytes() > 0; + + MemoryRecords memoryRecords = toReadableMemoryRecords(tp, fetchDataInfo.records, buffer); Review Comment: Thanks for pointing that out. I’m currently considering two ways to restore this behavior: 1. Introduce a simple wrapper class for the buffer, allowing it to be updated directly within the method. 2. Modify `toReadableMemoryRecords` to return a tuple containing both the MemoryRecords and the potentially updated buffer. Would you have a preference between the two approaches? -- 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
