hadoop之flume(三)

hadoop之flume(三)

四、单flume对应多个Channel

案例1：使用flume1监控文件变化，flume1将变动内容传递给flume2,flume2负责存储到HDFS，同时flume1将变动内容传递给flume3,flume3负责输出到local

1.创建flume1.conf配置文件，用于监控/home/f1文件的变化，同时产生2个channel和2个sink，2个sink分别输送给flume2,flume3

 

#定义agent
a1.sources = r1
a1.sinks = k1 k2
a1.channels = c1 c2

#将数据流复制给多个channel
a1.sources.r1.selector.type = replicating

#定义source r1
a1.sources.r1.type = exec
a1.sources.r1.command = tail -f /home/f1
a1.sources.r1.shell = /bin/bash -c

#定义sink k1
a1.sinks.k1.type = avro
a1.sinks.k1.hostname = hdnode1
a1.sinks.k1.port = 2000

#定义sink k2
a1.sinks.k2.type = avro
a1.sinks.k2.hostname = hdnode1
a1.sinks.k2.port = 2001

#定义channel c1
a1.channels.c1.type = memory
a1.channels.c1.capacity = 1000
a1.channels.c1.transactionCapacity = 100

#定义channel c2
a1.channels.c2.type = memory
a1.channels.c2.capacity = 1000
a1.channels.c2.transactionCapacity = 100

#绑定sources sinks channels
a1.sources.r1.channels = c1 c2
a1.sinks.k1.channel = c1
a1.sinks.k2.channel = c2

2.创建flume2.conf配置文件，用于接收flume1的event,同时产生1个channel和1个sink，将数据输送到hdfs中

#定义agent
a2.sources = r1
a2.sinks = k1
a2.channels = c1

#定义source r1
a2.sources.r1.type = avro
a2.sources.r1.bind = hdnode1
a2.sources.r1.port = 2000

#定义sink k1
a2.sinks.k1.type = hdfs
a2.sinks.k1.hdfs.path = hdfs://hdnode1:9000/flume/%Y%m%d%H
#上传文件的前缀
a2.sinks.k1.hdfs.filePrefix = flume2-
#是否按照时间滚动文件夹
a2.sinks.k1.hdfs.round = true
#多少时间单位创建一个新的文件夹
a2.sinks.k1.hdfs.roundVlaue = 1
#定义时间单位
a2.sinks.k1.hdfs.roundUnit = hour
#是否使用本地时间戳
a2.sinks.k1.hdfs.useLocalTimeStamp = true
#积攒多少个Event才flush到HDFS一次
a2.sinks.k1.hdfs.batchSize = 1000
#设置文件类型,可以支持压缩
a2.sinks.k1.hdfs.fileType = DataStream
#多久生成一个新文件
a2.sinks.k1.hdfs.rollInterval = 600
#设置每个文件的滚动大小
a2.sinks.k1.hdfs.rollSize = 134217700
#文件的滚动与Event数量无关
a2.sinks.k1.hdfs.roolCount = 0
#最小副本数
a2.sinks.k1.hdfs.minBlockReplicas = 1

#定义channel c1
a2.channels.c1.type = memory
a2.channels.c1.capacity = 1000
a2.channels.c1.transactionCapacity = 100

#绑定sources sinks channels
a2.sources.r1.channels = c1
a2.sinks.k1.channel = c1

3.创建flume3.conf配置文件，用于接收flume1的event,同时产生1个channel和1个sink，将数据输送到本地目录

#定义agent
a3.sources = r1
a3.sinks = k1
a3.channels = c1

#定义source r1
a3.sources.r1.type = avro
a3.sources.r1.bind = hdnode1
a3.sources.r1.port = 2001

#3 sink k1
a3.sinks.k1.type = file_roll
#文件夹需要提前先创建好
a3.sinks.k1.sink.directory = /home/tmpdir

#定义channel c1
a3.channels.c1.type = memory
a3.channels.c1.capacity = 1000
a3.channels.c1.transactionCapacity = 100

#绑定sources sinks channels
a3.sources.r1.channels = c1
a3.sinks.k1.channel = c1

4.启动flume

5.运行结果 

           

 

五、多flume对应单个Channel

案例：flume1监控文件/home/f1变化，flume2监控端口数据流，flume1和flume2将数据发送给flume3，flume3将最终数据写入hdfs

1.创建flume1,监控文件/home/f1变化，并且结果发送到flume2

#定义agent
a1.sources = r1
a1.sinks = k1
a1.channels = c1

#定义source r1
a1.sources.r1.type = exec
a1.sources.r1.command = tail -f /home/f1
a1.sources.r1.shell = /bin/bash -c

#定义sink k1
a1.sinks.k1.type = avro
a1.sinks.k1.hostname = hdnode1
a1.sinks.k1.port = 2000

#定义channel c1
a1.channels.c1.type = memory
a1.channels.c1.capacity = 1000
a1.channels.c1.transactionCapacity = 100

#绑定sources sinks channels
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1

2.创建flume2，监听端口数据流，发送给flume3

#定义agent
a2.sources = r1
a2.sinks = k1
a2.channels = c1

#定义source r1
a2.sources.r1.type = netcat
a2.sources.r1.bind = hdnode1
a2.sources.r1.port = 44445

#定义sink k1
a2.sinks.k1.type = avro
a2.sinks.k1.hostname = hdnode1
a2.sinks.k1.port = 2000

#定义channel c1
a2.channels.c1.type = memory
a2.channels.c1.capacity = 1000
a2.channels.c1.transactionCapacity = 100

#绑定sources sinks channels
a2.sources.r1.channels = c1
a2.sinks.k1.channel = c1

3.创建flume3，上报数据到hdfs

#定义Agent
a3.sources = r1
a3.sinks = k1
a3.channels = c1

#定义source
a3.sources.r1.type = avro
a3.sources.r1.bind = hdnode1
a3.sources.r1.port = 2000

# 定义sink
a3.sinks.k1.type = hdfs
a3.sinks.k1.hdfs.path = hdfs://hdnode1:9000/flume/%Y%m%d%H
#上传文件的前缀
a3.sinks.k1.hdfs.filePrefix = flume-
#是否按照时间滚动文件夹
a3.sinks.k1.hdfs.round = true
#多少时间单位创建一个新的文件夹
a3.sinks.k1.hdfs.roundVlaue = 1
#定义时间单位
a3.sinks.k1.hdfs.roundUnit = hour
#是否使用本地时间戳
a3.sinks.k1.hdfs.useLocalTimeStamp = true
#积攒多少个Event才flush到HDFS一次
a3.sinks.k1.hdfs.batchSize = 1000
#设置文件类型,可以支持压缩
a3.sinks.k1.hdfs.fileType = DataStream
#多久生成一个新文件
a3.sinks.k1.hdfs.rollInterval = 600
#设置每个文件的滚动大小
a3.sinks.k1.hdfs.rollSize = 134217700
#文件的滚动与Event数量无关
a3.sinks.k1.hdfs.roolCount = 0
#最小副本数
a3.sinks.k1.hdfs.minBlockReplicas = 1

# 定义memory
a3.channels.c1.type = memory
a3.channels.c1.capacity = 1000
a3.channels.c1.transactionCapacity = 100

# 双向链接
a3.sources.r1.channels = c1
a3.sinks.k1.channel = c1

4.启动flume

 启动flume3

/opt/module/flume-1.8.0/bin/flume-ng agent --conf /opt/module/flume1.8.0/conf/ --name a3 --conf-file /opt/module/flume-1.8.0/jobconf/flume3.conf -Dflume.root.logger==INFO,console

 启动flume2

/opt/module/flume-1.8.0/bin/flume-ng agent --conf /opt/module/flume1.8.0/conf/ --name a2 --conf-file /opt/module/flume-1.8.0/jobconf/flume2.conf -Dflume.root.logger==INFO,console

 启动flume1

/opt/module/flume-1.8.0/bin/flume-ng agent --conf /opt/module/flume1.8.0/conf/ --name a1 --conf-file /opt/module/flume-1.8.0/jobconf/flume1.conf -Dflume.root.logger==INFO,console

5.运行结果