来自Flume的Kafka的EOFException

Question

I am trying to set up a simple data pipeline from a console Kafka producer to the Hadoop file system (HDFS). I am working on a 64bit Ubuntu Virtual Machine and have created separate users for both Hadoop and Kafka as was suggested by the guides that I have followed. Consuming the produced input in Kafka with a console consumer works and the HDFS seems to be up and running.

Now I want to use Flume to pipe the input into the HDFS. I am using the following configuration file:

tier1.sources  = source1
tier1.channels = channel1
tier1.sinks = sink1

tier1.sources.source1.type = org.apache.flume.source.kafka.KafkaSource
tier1.sources.source1.zookeeperConnect = 127.0.0.1:2181
tier1.sources.source1.topic = test
tier1.sources.source1.groupId = flume
tier1.sources.source1.channels = channel1
tier1.sources.source1.interceptors = i1
tier1.sources.source1.interceptors.i1.type = timestamp
tier1.sources.source1.kafka.consumer.timeout.ms = 2000

tier1.channels.channel1.type = memory
tier1.channels.channel1.capacity = 10000
tier1.channels.channel1.transactionCapacity = 1000

tier1.sinks.sink1.type = hdfs
tier1.sinks.sink1.hdfs.path = hdfs://flume/kafka/%{topic}/%y-%m-%d
tier1.sinks.sink1.hdfs.rollInterval = 5
tier1.sinks.sink1.hdfs.rollSize = 0
tier1.sinks.sink1.hdfs.rollCount = 0
tier1.sinks.sink1.hdfs.fileType = DataStream
tier1.sinks.sink1.channel = channel1

Now when I run Flume with the following command

bin/flume-ng agent --conf ./conf -f conf/flume.conf -Dflume.root.logger=DEBUG,console -n tier1

I get the same exception in the console output over and over again:

2017-10-19 12:17:04,279 (lifecycleSupervisor-1-2) [DEBUG - org.apache.kafka.clients.NetworkClient.handleConnections(NetworkClient.java:467)] Completed connection to node 2147483647
2017-10-19 12:17:04,279 (lifecycleSupervisor-1-2) [DEBUG - org.apache.kafka.common.network.Selector.poll(Selector.java:307)] Connection with Ubuntu-Sandbox/127.0.1.1 disconnected
java.io.EOFException
    at org.apache.kafka.common.network.NetworkReceive.readFromReadableChannel(NetworkReceive.java:83)
    at org.apache.kafka.common.network.NetworkReceive.readFrom(NetworkReceive.java:71)
    at org.apache.kafka.common.network.KafkaChannel.receive(KafkaChannel.java:153)
    at org.apache.kafka.common.network.KafkaChannel.read(KafkaChannel.java:134)
    at org.apache.kafka.common.network.Selector.poll(Selector.java:286)
    at org.apache.kafka.clients.NetworkClient.poll(NetworkClient.java:256)
    at org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient.clientPoll(ConsumerNetworkClient.java:320)
    at org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient.poll(ConsumerNetworkClient.java:213)
    at org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient.poll(ConsumerNetworkClient.java:193)
    at org.apache.kafka.clients.consumer.internals.ConsumerNetworkClient.poll(ConsumerNetworkClient.java:163)
    at org.apache.kafka.clients.consumer.internals.AbstractCoordinator.ensureActiveGroup(AbstractCoordinator.java:222)
    at org.apache.kafka.clients.consumer.internals.ConsumerCoordinator.ensurePartitionAssignment(ConsumerCoordinator.java:311)
    at org.apache.kafka.clients.consumer.KafkaConsumer.pollOnce(KafkaConsumer.java:890)
    at org.apache.kafka.clients.consumer.KafkaConsumer.poll(KafkaConsumer.java:853)
    at org.apache.flume.source.kafka.KafkaSource.doStart(KafkaSource.java:529)
    at org.apache.flume.source.BasicSourceSemantics.start(BasicSourceSemantics.java:83)
    at org.apache.flume.source.PollableSourceRunner.start(PollableSourceRunner.java:71)
    at org.apache.flume.lifecycle.LifecycleSupervisor$MonitorRunnable.run(LifecycleSupervisor.java:249)
    at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
    at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:308)
    at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:180)
    at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:294)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
    at java.lang.Thread.run(Thread.java:748)

The only way to stop Flume is to kill the Java process.

I thought that it might have something to do with the separate users for Hadoop and Kafka, but even when running everything with the Kafka user I get the same result. I haven't found anything concerning the EOFException method online either, which is strange considering that I have just followed the "Getting Started" guides and used pretty standard configurations for everything.

Maybe it has something to do with the preceding line ("Ubuntu-Sandbox/127.0.1.1 disconnected") and hence the configuration of my VM?

Any help is highly appreciated!

Answer 1

Have you considered using Kafka Connect (part of Apache Kafka) and the HDFS connector instead? This is generally seen to have superseded Flume. It is easy to use, with a similar file-based configuration as Flume.