数据采集到kafka
flume:
a1.sources = r1
a1.channels = c1
a1.sources.r1.type = TAILDIR
a1.sources.r1.filegroups = f1
a1.sources.r1.filegroups.f1 = /opt/installs/flume1.9/job/a.log
a1.sources.r1.positionFile = /opt/installs/flume1.9/job/taildir-kafka.json
a1.channels.c1.type = org.apache.flume.channel.kafka.KafkaChannel
a1.channels.c1.kafka.bootstrap.servers =hadoop11:9092,hadoop12:9092,hadoop13:9092
a1.channels.c1.kafka.topic = topica
a1.channels.c1.parseAsFlumeEvent = false
a1.sources.r1.channels = c1
执行命令：
flume-ng agent --conf conf --name a1 --conf-file job/taildir-kafka.conf -Dflume.root.logger=INFO,console
向a.log添加测试数据： 
消费者：
数据从kafka到hdfs
flume：
a1.sources = r1
a1.channels = c1
a1.sinks = k1
a1.sources.r1.type = org.apache.flume.source.kafka.KafkaSource
a1.sources.r1.batchSize=5000
a1.sources.r1.batchDurationMillis=2000
a1.sources.r1.kafka.bootstrap.servers =hadoop11:9092,hadoop12:9092,hadoop13:9092
a1.sources.r1.kafka.topics = topica
a1.sources.r1.kafka.consumer.group.id = g1
a1.channels.c1.type = memory
a1.channels.c1.capacity=5000
a1.channels.c1.transactionCapacity=5000
a1.sinks.k1.type = hdfs
a1.sinks.k1.batchSize = 5000
a1.sinks.k1.hdfs.path = hdfs://hadoop11:8020/flume/date=%Y-%m-%d
a1.sinks.k1.hdfs.useLocalTimeStamp = true
a1.sinks.k1.hdfs.fileType = DataStream
a1.sinks.k1.hdfs.round = true
a1.sinks.k1.hdfs.rollInterval =0
a1.sinks.k1.hdfs.rollSize = 1048576
a1.sinks.k1.hdfs.rollCount = 0
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1
执行命令：
flume-ng agent --conf conf --name a1 --conf-file job/kafka-hdfs.conf -Dflume.root.logger=INFO,console
向a.log添加测试数据：
消费者：
hdfs：
