一、现象描述
使用Mac电脑本地启动spring-kakfa消费不到Kafka的消息，监控消费组的消息偏移量发现存在Lag的消息，但是本地客户端就是拉取不到，通过部署到公司k8s容器上消息却能正常消费！
本地启动的服务消费组监控
公司k8s容器服务消费组监控
二、环境信息
Spring Kafka版本： 2.1.13.RELEASE
Kafka Client版本： 1.0.2
Local JDK版本： Zulu 8.60.0.21-CA-macos-aarch64
K8s JDK版本： Oracle 1.8.0_202-b08
三、排查过程
-
猜测是JDK版本或者JDK 对 Apple Silicon芯片兼容问题
-
Debug跟踪了KafkaConsumer poll过程，并没有发现任何异常，轮询拉取的线程正常循环执行，只是每次都拉取到 records 为0条。
-
决定调整kafka 日志级别看下心跳是否正常，居然发现了有异常抛出，看到是snappy相关类NotClassFound
SLF4J: Failed toString() invocation on an object of type [org.apache.kafka.common.protocol.types.Struct]
Reported exception:
java.lang.NoClassDefFoundError: Could not initialize class org.xerial.snappy.Snappyat org.xerial.snappy.SnappyInputStream.hasNextChunk(SnappyInputStream.java:435)at org.xerial.snappy.SnappyInputStream.read(SnappyInputStream.java:466)at java.io.DataInputStream.readByte(DataInputStream.java:265)at org.apache.kafka.common.utils.ByteUtils.readVarint(ByteUtils.java:168)at org.apache.kafka.common.record.DefaultRecord.readFrom(DefaultRecord.java:292)at org.apache.kafka.common.record.DefaultRecordBatch$1.readNext(DefaultRecordBatch.java:264)at org.apache.kafka.common.record.DefaultRecordBatch$RecordIterator.next(DefaultRecordBatch.java:563)at org.apache.kafka.common.record.DefaultRecordBatch$RecordIterator.next(DefaultRecordBatch.java:532)at org.apache.kafka.common.record.MemoryRecords.toString(MemoryRecords.java:292)at java.lang.String.valueOf(String.java:2994)at java.lang.StringBuilder.append(StringBuilder.java:136)at org.apache.kafka.common.protocol.types.Struct.toString(Struct.java:390)at java.lang.String.valueOf(String.java:2994)at java.lang.StringBuilder.append(StringBuilder.java:136)at org.apache.kafka.common.protocol.types.Struct.toString(Struct.java:384)at java.lang.String.valueOf(String.java:2994)at java.lang.StringBuilder.append(StringBuilder.java:136)at org.apache.kafka.common.protocol.types.Struct.toString(Struct.java:384)at org.slf4j.helpers.MessageFormatter.safeObjectAppend(MessageFormatter.java:299)at org.slf4j.helpers.MessageFormatter.deeplyAppendParameter(MessageFormatter.java:271)at org.slf4j.helpers.MessageFormatter.arrayFormat(MessageFormatter.java:233)at org.slf4j.helpers.MessageFormatter.arrayFormat(MessageFormatter.java:173)at ch.qos.logback.classic.spi.LoggingEvent.getFormattedMessage(LoggingEvent.java:293)at ch.qos.logback.classic.spi.LoggingEvent.prepareForDeferredProcessing(LoggingEvent.java:206)at ch.qos.logback.core.OutputStreamAppender.subAppend(OutputStreamAppender.java:223)at ch.qos.logback.core.OutputStreamAppender.append(OutputStreamAppender.java:102)at ch.qos.logback.core.UnsynchronizedAppenderBase.doAppend(UnsynchronizedAppenderBase.java:84)at ch.qos.logback.core.spi.AppenderAttachableImpl.appendLoopOnAppenders(AppenderAttachableImpl.java:51)at ch.qos.logback.classic.Logger.appendLoopOnAppenders(Logger.java:270)at ch.qos.logback.classic.Logger.callAppenders(Logger.java:257)at ch.qos.logback.classic.Logger.buildLoggingEventAndAppend(Logger.java:421)at ch.qos.logback.classic.Logger.filterAndLog_0_Or3Plus(Logger.java:383)at ch.qos.logback.classic.Logger.trace(Logger.java:437)at org.apache.kafka.common.utils.LogContext$KafkaLogger.trace(LogContext.java:135)at org.apache.kafka.clients.NetworkClient.handleCompletedReceives(NetworkClient.java:689)at org.apache.kafka.clients.NetworkClient.poll(NetworkClient.java:469)at org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient.poll(ConsumerNetworkClient.java:258)at org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient.pollNoWakeup(ConsumerNetworkClient.java:297)at org.apache.kafka.clients.consumer.internals.AbstractCoordinator$HeartbeatThread.run(AbstractCoordinator.java:948)
[2023-09-15 14:02:27.248]^^A[TID: N/A]^^A[kafka-coordinator-heartbeat-thread | ingest-consume-group-follow-test-4]^^ATRACE^^Aorg.apache.kafka.clients.NetworkClient^^A[Consumer clientId=consumer-1, groupId=ingest-consume-group-follow-test-4] Completed receive from node 1 for FETCH with correlation id 15, received [FAILED toString()]
-
如果了解 snappy-java这个依赖包的话，到这里就对拉取不到消息原因猜测的八九不离十了，因为 Kafka 服务端使用 snappy对息做了压缩并序列化为二进制进行传输，如果客户端在对消息的解压与反序列化过程中抛出异常，那么自然就拉取不到消息。
-
接着，解决一下snappy-java包的兼容问题，通过验证升级版本可以解决此问题。
排除kafka-client包中 snappy-java v1.1.4版本依赖
<!-- spring-kafka --><dependency><groupId>org.springframework.kafka</groupId><artifactId>spring-kafka</artifactId><exclusions><!-- 排除 snappy-java 1.1.4 版本 --><exclusion><groupId>org.xerial.snappy</groupId><artifactId>snappy-java</artifactId></exclusion></exclusions></dependency>
- 再引入高版本v1.1.8.4的依赖包
<dependency><groupId>org.xerial.snappy</groupId><artifactId>snappy-java</artifactId><version>1.1.8.4</version><scope>compile</scope>
</dependency>
- 重新编译启动spring kafka客户端程序，消费问题解决~
四、疑问解答
- 为什么Kafka Consumer poll消息过程没有异常抛出且可以正常运行？
答：待补充 - 为什么调整日志级别为Trace才看到异常日志抛出？
答：待补充
