flume常见异常汇总以及解决方案

　　　　　　　　　　　　　　　　　　　　　　flume常见异常汇总以及解决方案

　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　作者：尹正杰

版权声明：原创作品，谢绝转载！否则将追究法律责任。

　　实际生产环境中，我用flume将kafka的数据定期的往hdfs集群中上传数据，也遇到过一系列的坑，我在这里做个记录，如果你也遇到同样的错误，可以参考一下我的解决方案。

1>.服务器在接收到响应之前断开连接。

报错信息如下：

　　Caused by: org.apache.kafka.common.errors.NetworkException: The server disconnected before a response was received.

-- ::, (PollableSourceRunner-KafkaSource-kafkaSource) [INFO - org.apache.kafka.clients.consumer.internals.AbstractCoordinator.coordinatorDead(AbstractCoordinator.java:)] Marking the coordinator  dead.
Caused by: org.apache.kafka.common.errors.NetworkException: The server disconnected before a response was received.
        ...  more
        at org.apache.flume.channel.kafka.KafkaChannel$KafkaTransaction.doCommit(KafkaChannel.java:)
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.get(FutureRecordMetadata.java:)
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.get(FutureRecordMetadata.java:)
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.valueOrError(FutureRecordMetadata.java:)
Caused by: java.util.concurrent.ExecutionException: org.apache.kafka.common.errors.NetworkException: The server disconnected before a response was received.
        at java.lang.Thread.run(Thread.java:)
        at org.apache.flume.source.PollableSourceRunner$PollingRunner.run(PollableSourceRunner.java:)
        at org.apache.flume.source.AbstractPollableSource.process(AbstractPollableSource.java:)
        at org.apache.flume.source.kafka.KafkaSource.doProcess(KafkaSource.java:)
        at org.apache.flume.channel.ChannelProcessor.processEventBatch(ChannelProcessor.java:)
        at org.apache.flume.channel.BasicTransactionSemantics.commit(BasicTransactionSemantics.java:)
        at org.apache.flume.channel.kafka.KafkaChannel$KafkaTransaction.doCommit(KafkaChannel.java:)
org.apache.flume.ChannelException: Commit failed as send to Kafka failed
-- ::, (PollableSourceRunner-KafkaSource-kafkaSource) [ERROR - org.apache.flume.source.kafka.KafkaSource.doProcess(KafkaSource.java:)] KafkaSource EXCEPTION, {}
Caused by: org.apache.kafka.common.errors.NetworkException: The server disconnected before a response was received.
        at java.lang.Thread.run(Thread.java:)
        at org.apache.flume.source.PollableSourceRunner$PollingRunner.run(PollableSourceRunner.java:)
        at org.apache.flume.source.AbstractPollableSource.process(AbstractPollableSource.java:)
        at org.apache.flume.source.kafka.KafkaSource.doProcess(KafkaSource.java:)
        at org.apache.flume.channel.ChannelProcessor.processEventBatch(ChannelProcessor.java:)
        at org.apache.flume.channel.BasicTransactionSemantics.commit(BasicTransactionSemantics.java:)
        at org.apache.flume.channel.kafka.KafkaChannel$KafkaTransaction.doCommit(KafkaChannel.java:)
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.get(FutureRecordMetadata.java:)
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.get(FutureRecordMetadata.java:)
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.valueOrError(FutureRecordMetadata.java:)
java.util.concurrent.ExecutionException: org.apache.kafka.common.errors.NetworkException: The server disconnected before a response was received.
-- ::, (PollableSourceRunner-KafkaSource-kafkaSource) [WARN - org.apache.flume.channel.kafka.KafkaChannel$KafkaTransaction.doCommit(KafkaChannel.java:)] Sending events to Kafka failed
-- ::, (SinkRunner-PollingRunner-DefaultSinkProcessor) [INFO - org.apache.kafka.clients.consumer.internals.AbstractCoordinator.coordinatorDead(AbstractCoordinator.java:)] Marking the coordinator  dead.
-- ::, (SinkRunner-PollingRunner-DefaultSinkProcessor) [INFO - org.apache.kafka.clients.consumer.internals.ConsumerCoordinator$OffsetCommitResponseHandler.handle(ConsumerCoordinator.java:)] Offset commit for group flume-consumer-against_cheating_02 failed due to REQUEST_TIMED_OUT, will find new coordinator and retry
-- ::, (SinkRunner-PollingRunner-DefaultSinkProcessor) [INFO - org.apache.kafka.clients.consumer.internals.AbstractCoordinator.coordinatorDead(AbstractCoordinator.java:)] Marking the coordinator  dead.
-- ::, (SinkRunner-PollingRunner-DefaultSinkProcessor) [INFO - org.apache.kafka.clients.consumer.internals.ConsumerCoordinator$OffsetCommitResponseHandler.handle(ConsumerCoordinator.java:)] Offset commit for group flume-consumer-against_cheating_02 failed due to REQUEST_TIMED_OUT, will find new coordinator and retry

Offset commit for group flume-consumer-against_cheating_02 failed due to REQUEST_TIMED_OUT, will find new coordinator and retry

 　　根据报错分析，是由于时间过长导致重新均衡的，参考：https://kafka.apache.org/090/documentation/#configuration，经查阅资料应该调大以下几个参数：

注意，这个*表示的是channels的名称，这些参数不仅仅是可以给kafka channel设置，还可以给kafka source配置哟！

#配置控制服务器等待追随者确认以满足生产者用acks配置指定的确认要求的最大时间量。如果超时后所请求的确认数不满足，将返回一个错误。此超时在服务器端进行测量，不包括请求的网络延迟。
agent.channels.*.kafka.consumer.timeout.ms = 

#配置控制客户端等待请求响应的最大时间量。如果在超时之前没有接收到响应，则客户端将在必要时重新发送请求，或者如果重试用尽，则请求失败。
agent.channels.*.kafka.consumer.request.timeout.ms = 

#如果没有足够的数据立即满足fetch.min.bytes给出的要求，服务器在回答获取请求之前将阻塞的最大时间。
agent.channels.*.kafka.consumer.fetch.max.wait.ms=

#在取消处理和恢复要提交的偏移数据之前，等待记录刷新和分区偏移数据提交到偏移存储的最大毫秒数。
agent.channels.*.kafka.consumer.offset.flush.interval.ms = 

#用于在使用kafka组管理设施时检测故障的超时时间。
agent.channels.*.kafka.consumer.session.timeout.ms = 

#使用kafka组管理设施时，消费者协调器心跳的预期时间。心跳用于确保消费者的会话保持活跃，并在新消费者加入或离开组时促进重新平衡。该值必须设置为低于session.timeout.ms，但通常应设置为不高于该值的1/。它可以调整得更低，以控制正常再平衡的预期时间。
agent.channels.*.kafka.consumer.heartbeat.interval.ms = 

#如果是true，消费者的偏移将在后台周期性地提交。如果auto.commit.enable=true，当consumer fetch了一些数据但还没有完全处理掉的时候，刚好到commit interval出发了提交offset操作，接着consumer crash掉了。这时已经fetch的数据还没有处理完成但已经被commit掉，因此没有机会再次被处理，数据丢失。
agent.channels.*.kafka.consumer.enable.auto.commit = false

2>.producer在向kafka broker写的时候，刚好发生选举，本来是向broker0上写的，选举之后broker1成为leader，所以无法写成功，就抛异常了。

报错信息如下：

　　java.util.concurrent.ExecutionException: org.apache.kafka.common.errors.NotLeaderForPartitionException: This server is not the leader for that topic-partition. 

　　以上报错是我在重启kafka集群中发现的报错，百度了一下说是：producer在向kafka broker写的时候，刚好发生选举，本来是向broker0上写的，选举之后broker1成为leader，所以无法写成功，就抛异常了。

-- ::, (SinkRunner-PollingRunner-DefaultSinkProcessor) [INFO - org.apache.flume.sink.hdfs.BucketWriter.open(BucketWriter.java:)] Creating hdfs://hdfs-ha/user/against_cheating/20181115/10-1-2-120_02_20181115_16.1542271273895.txt.tmp
-- ::, (SinkRunner-PollingRunner-DefaultSinkProcessor) [INFO - org.apache.kafka.clients.consumer.internals.ConsumerCoordinator$OffsetCommitResponseHandler.handle(ConsumerCoordinator.java:)] Offset commit for group flume-consumer-against_cheating_02 failed due to REQUEST_TIMED_OUT, will find new coordinator and retry
-- ::, (SinkRunner-PollingRunner-DefaultSinkProcessor) [INFO - org.apache.kafka.clients.consumer.internals.AbstractCoordinator.coordinatorDead(AbstractCoordinator.java:)] Marking the coordinator  dead.
-- ::, (PollableSourceRunner-KafkaSource-kafkaSource) [WARN - org.apache.flume.channel.kafka.KafkaChannel$KafkaTransaction.doCommit(KafkaChannel.java:)] Sending events to Kafka failed
java.util.concurrent.ExecutionException: org.apache.kafka.common.errors.NotLeaderForPartitionException: This server is not the leader for that topic-partition.
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.valueOrError(FutureRecordMetadata.java:)
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.get(FutureRecordMetadata.java:)
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.get(FutureRecordMetadata.java:)
        at org.apache.flume.channel.kafka.KafkaChannel$KafkaTransaction.doCommit(KafkaChannel.java:)
        at org.apache.flume.channel.BasicTransactionSemantics.commit(BasicTransactionSemantics.java:)
        at org.apache.flume.channel.ChannelProcessor.processEventBatch(ChannelProcessor.java:)
        at org.apache.flume.source.kafka.KafkaSource.doProcess(KafkaSource.java:)
        at org.apache.flume.source.AbstractPollableSource.process(AbstractPollableSource.java:)
        at org.apache.flume.source.PollableSourceRunner$PollingRunner.run(PollableSourceRunner.java:)
        at java.lang.Thread.run(Thread.java:)
Caused by: org.apache.kafka.common.errors.NotLeaderForPartitionException: This server is not the leader for that topic-partition.
-- ::, (PollableSourceRunner-KafkaSource-kafkaSource) [ERROR - org.apache.flume.source.kafka.KafkaSource.doProcess(KafkaSource.java:)] KafkaSource EXCEPTION, {}
org.apache.flume.ChannelException: Commit failed as send to Kafka failed
        at org.apache.flume.channel.kafka.KafkaChannel$KafkaTransaction.doCommit(KafkaChannel.java:)
        at org.apache.flume.channel.BasicTransactionSemantics.commit(BasicTransactionSemantics.java:)
        at org.apache.flume.channel.ChannelProcessor.processEventBatch(ChannelProcessor.java:)
        at org.apache.flume.source.kafka.KafkaSource.doProcess(KafkaSource.java:)
        at org.apache.flume.source.AbstractPollableSource.process(AbstractPollableSource.java:)
        at org.apache.flume.source.PollableSourceRunner$PollingRunner.run(PollableSourceRunner.java:)
        at java.lang.Thread.run(Thread.java:)
Caused by: java.util.concurrent.ExecutionException: org.apache.kafka.common.errors.NotLeaderForPartitionException: This server is not the leader for that topic-partition.
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.valueOrError(FutureRecordMetadata.java:)
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.get(FutureRecordMetadata.java:)
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.get(FutureRecordMetadata.java:)
        at org.apache.flume.channel.kafka.KafkaChannel$KafkaTransaction.doCommit(KafkaChannel.java:)
        ...  more
Caused by: org.apache.kafka.common.errors.NotLeaderForPartitionException: This server is not the leader for that topic-partition.
-- ::, (PollableSourceRunner-KafkaSource-kafkaSource) [WARN - org.apache.flume.channel.kafka.KafkaChannel$KafkaTransaction.doCommit(KafkaChannel.java:)] Sending events to Kafka failed
java.util.concurrent.ExecutionException: org.apache.kafka.common.errors.NotLeaderForPartitionException: This server is not the leader for that topic-partition.
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.valueOrError(FutureRecordMetadata.java:)
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.get(FutureRecordMetadata.java:)
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.get(FutureRecordMetadata.java:)
        at org.apache.flume.channel.kafka.KafkaChannel$KafkaTransaction.doCommit(KafkaChannel.java:)
        at org.apache.flume.channel.BasicTransactionSemantics.commit(BasicTransactionSemantics.java:)
        at org.apache.flume.channel.ChannelProcessor.processEventBatch(ChannelProcessor.java:)
        at org.apache.flume.source.kafka.KafkaSource.doProcess(KafkaSource.java:)
        at org.apache.flume.source.AbstractPollableSource.process(AbstractPollableSource.java:)
        at org.apache.flume.source.PollableSourceRunner$PollingRunner.run(PollableSourceRunner.java:)
        at java.lang.Thread.run(Thread.java:)
Caused by: org.apache.kafka.common.errors.NotLeaderForPartitionException: This server is not the leader for that topic-partition.
-- ::, (PollableSourceRunner-KafkaSource-kafkaSource) [ERROR - org.apache.flume.source.kafka.KafkaSource.doProcess(KafkaSource.java:)] KafkaSource EXCEPTION, {}
org.apache.flume.ChannelException: Commit failed as send to Kafka failed
        at org.apache.flume.channel.kafka.KafkaChannel$KafkaTransaction.doCommit(KafkaChannel.java:)
        at org.apache.flume.channel.BasicTransactionSemantics.commit(BasicTransactionSemantics.java:)
        at org.apache.flume.channel.ChannelProcessor.processEventBatch(ChannelProcessor.java:)
        at org.apache.flume.source.kafka.KafkaSource.doProcess(KafkaSource.java:)
        at org.apache.flume.source.AbstractPollableSource.process(AbstractPollableSource.java:)
        at org.apache.flume.source.PollableSourceRunner$PollingRunner.run(PollableSourceRunner.java:)
        at java.lang.Thread.run(Thread.java:)
Caused by: java.util.concurrent.ExecutionException: org.apache.kafka.common.errors.NotLeaderForPartitionException: This server is not the leader for that topic-partition.
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.valueOrError(FutureRecordMetadata.java:)
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.get(FutureRecordMetadata.java:)
        at org.apache.kafka.clients.producer.internals.FutureRecordMetadata.get(FutureRecordMetadata.java:)
        at org.apache.flume.channel.kafka.KafkaChannel$KafkaTransaction.doCommit(KafkaChannel.java:)
        ...  more
Caused by: org.apache.kafka.common.errors.NotLeaderForPartitionException: This server is not the leader for that topic-partition.
-- ::, (hdfs-hdfsSink-roll-timer-) [INFO - org.apache.flume.sink.hdfs.BucketWriter.close(BucketWriter.java:)] Closing hdfs://hdfs-ha/user/against_cheating/20181115/10-1-2-120_02_20181115_16.1542271273895.txt.tmp

java.util.concurrent.ExecutionException: org.apache.kafka.common.errors.NotLeaderForPartitionException: This server is not the leader for that topic-partition.

　　解决方案就是：

　　　　1>.先确认kafka集群是否在稳定运行，如果kafka集群异常的话，这个报错会一致不断的发出来；

　　　　2>.如果刚刚重启集群的话，暂时先不高管它，flume会自动去重试，但是你也别闲着，查看kafka监控界面，观察是否有异常的现象，如果时间超过了2分钟还没有恢复，那你就得考虑是否是你的kafka集群出现问题了。

3>.指定在 DataNode 内外传输数据使用的最大线程数偏小。

报错信息如下：

　　java.io.IOException: Bad connect ack with firstBadLink as 10.1.1.120:50010

　　百度了一下原因：　　

　　　　Datanode往hdfs上写时，实际上是通过使用xcievers这个中间服务往linux上的文件系统上写文件的。其实这个xcievers就是一些负责在DataNode和本地磁盘上读，写文件的线程。DataNode上Block越多，这个线程的数量就应该越多。然后问题来了，这个线程数有个上线（默认是配置的4096）。所以，当Datenode上的Block数量过多时，就会有些Block文件找不到。线程来负责他的读和写工作了。所以就出现了上面的错误（写块失败）。

　　解决方案：

　　　　将DataNode 内外传输数据使用的最大线程数增大，比如：65535。

 

4>.java.io.EOFException: Premature EOF: no length prefix available

　　根据上图到提示，我们可以依稀看到DN节点，于是我们去CDH（如果你用到时HDP就去相应到平台即可）找相应到日志，发现的确有报错信息如下：

我遇到了上述的问题后我做了3给操作，最终问题得以解决：

　　第一步：调优hdfs集群，详细参数请参考我的笔记：https://www.cnblogs.com/yinzhengjie/p/10006880.html。

　　第二步：编辑了以下2个配置文件。

[root@calculation101 ~]# cat /etc/security/limits.d/-nproc.conf
# Default limit for number of user's processes to prevent
# accidental fork bombs.
# See rhbz # for reasoning.

*          soft    nproc
root       soft    nproc     unlimited
[root@calculation101 ~]# 

[root@calculation101 ~]# cat /etc/security/limits.d/20-nproc.conf

[root@calculation101 ~]# cat /etc/security/limits.conf  | grep -v ^#  | grep -v ^$
*        soft    nofile
*         hard    nofile
*        soft    nproc
*        hard    nproc        unlimited
*        soft    memlock        unlimited
*        hard    memlock        unlimited
[root@calculation101 ~]# 

[root@calculation101 ~]# cat /etc/security/limits.conf | grep -v ^# | grep -v ^$

　　第三步：重启操作系统，重启前确保所有的服务关闭，重启成功后，确保所有的hdfs集群启动成功，200G的数据只需要3分钟左右就跑完了，2天过去了，上述的报错依旧没有复现过，如果大家遇到跟我相同的问题，也可以试试我的这个方法。

5>.