Flume复制和多路复用

Flume安装：
1.解压apache-flume-1.7.0-bin.tar.gz到/opt/module/目录下
[root@JHB0 software]$ tar -zxf apache-flume-1.7.0-bin.tar.gz -C /opt/module/

2.改名
mv apache-flume-1.7.0-bin flume

3.修改配置文件
[root@JHB0 conf]$ mv flume-env.sh.template flume-env.sh
[root@JHB0 conf]$ vi flume-env.sh
export JAVA_HOME=/opt/module/jdk1.8.0_144

4.安装netcat工具
vi /etc/yum.repos.d/CentOS-Base.repo
[root@JHB0 software]$ sudo yum install -y nc

5.判断44444端口是否被占用
[root@JHB0 flume-telnet]$ sudo netstat -tunlp | grep 44444

6.创建Flume Agent配置文件flume-netcat-logger.conf
[root@JHB0 flume]$ mkdir job
[root@JHB0 flume]$ cd job/
[root@JHB0 job]$ vim flume-netcat-logger.conf
添加如下内容：
添加内容如下：
# Name the components on this agent
a1.sources = r1
a1.sinks = k1
a1.channels = c1

# Describe/configure the source
a1.sources.r1.type = netcat
a1.sources.r1.bind = localhost
a1.sources.r1.port = 44444

# Describe the sink
a1.sinks.k1.type = logger

# Use a channel which buffers events in memory
a1.channels.c1.type = memory
a1.channels.c1.capacity = 1000
a1.channels.c1.transactionCapacity = 100

# Bind the source and sink to the channel
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1

7. 开启Flume监听端口
两种写法
①
[root@JHB0 flume]$ bin/flume-ng agent --conf conf/ --name a1 --conf-file job/flume-netcat-logger.conf -Dflume.root.logger=INFO,console

②
[root@JHB0 flume]$ bin/flume-ng agent -c conf/ -n a1 –f job/flume-netcat-logger.conf -Dflume.root.logger=INFO,console

8.使用netcat工具向本机的44444端口发送内容
[root@JHB0 ~]$ nc localhost 44444
hello 
root

6.在Flume监控界面观察接收数据情况

==========================Flume复制和多路复用=====================================

1.在/opt/module/flume/job目录下创建group1文件夹
    [root@JHB0 job]$ cd group1/

2.在/opt/module/datas/目录下创建flume3文件夹
    [root@JHB0 datas]$ mkdir flume3

3.编辑配置文件
（1）flume-file-flume.conf
    [root@JHB0 group1]$ vim flume-file-flume.conf
    配置如下：

    # Name the components on this agent
    a1.sources = r1
    a1.sinks = k1 k2
    a1.channels = c1 c2
    # 将数据流复制给所有channel
    a1.sources.r1.selector.type = replicating

    # Describe/configure the source
    a1.sources.r1.type = exec
    a1.sources.r1.command = tail -F /opt/module/hive/logs/hive.log
    a1.sources.r1.shell = /bin/bash -c

    # Describe the sink
    # sink端的avro是一个数据发送者
    a1.sinks.k1.type = avro
    a1.sinks.k1.hostname = JHB0 
    a1.sinks.k1.port = 4141

    a1.sinks.k2.type = avro
    a1.sinks.k2.hostname = JHB0
    a1.sinks.k2.port = 4142

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

    a1.channels.c2.type = memory
    a1.channels.c2.capacity = 1000
    a1.channels.c2.transactionCapacity = 100

    # Bind the source and sink to the channel
    a1.sources.r1.channels = c1 c2
    a1.sinks.k1.channel = c1
    a1.sinks.k2.channel = c2

    （2）flume-flume-hdfs.conf
    [root@JHB0 group1]$ vim flume-flume-hdfs.conf
    配置如下：
    # Name the components on this agent
    a2.sources = r1
    a2.sinks = k1
    a2.channels = c1

    # Describe/configure the source
    # source端的avro是一个数据接收服务
    a2.sources.r1.type = avro
    a2.sources.r1.bind = JHB0
    a2.sources.r1.port = 4141

    # Describe the sink
    a2.sinks.k1.type = hdfs
    a2.sinks.k1.hdfs.path = hdfs://JHB0:9000/flume2/%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.roundValue = 1
    #重新定义时间单位
    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 = 100
    #设置文件类型，可支持压缩
    a2.sinks.k1.hdfs.fileType = DataStream
    #多久生成一个新的文件
    a2.sinks.k1.hdfs.rollInterval = 600
    #设置每个文件的滚动大小大概是128M
    a2.sinks.k1.hdfs.rollSize = 134217700
    #文件的滚动与Event数量无关
    a2.sinks.k1.hdfs.rollCount = 0

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

    # Bind the source and sink to the channel
    a2.sources.r1.channels = c1
    a2.sinks.k1.channel = c1

    （3）flume-flume-dir.conf
    [root@JHB0 group1]$ vim flume-flume-dir.conf
    配置信息：
    # Name the components on this agent
    a3.sources = r1
    a3.sinks = k1
    a3.channels = c2

    # Describe/configure the source
    a3.sources.r1.type = avro
    a3.sources.r1.bind = JHB0
    a3.sources.r1.port = 4142

    # Describe the sink
    a3.sinks.k1.type = file_roll
    a3.sinks.k1.sink.directory = /opt/module/data/flume3

    # Describe the channel
    a3.channels.c2.type = memory
    a3.channels.c2.capacity = 1000
    a3.channels.c2.transactionCapacity = 100

    # Bind the source and sink to the channel
    a3.sources.r1.channels = c2
    a3.sinks.k1.channel = c2

4.执行配置文件
[root@JHB0 flume]$ bin/flume-ng agent --conf conf/ --name a3 --conf-file job/group1/flume-flume-dir.conf

[root@JHB0 flume]$ bin/flume-ng agent --conf conf/ --name a2 --conf-file job/group1/flume-flume-hdfs.conf

[root@JHB0 flume]$ bin/flume-ng agent --conf conf/ --name a1 --conf-file job/group1/flume-file-flume.conf

5.启动Hadoop和Hive
[root@JHB0 hadoop-2.7.2]$ sbin/start-dfs.sh
[root@JHB1 hadoop-2.7.2]$ sbin/start-yarn.sh

[root@JHB0 hive]$ bin/hive
hive (default)>

6.检查HDFS上的数据
打开火狐浏览器，地址栏中输入JHB0://50070

7.检查/opt/module/datas/flume3目录中数据    flume3对应flume-1?
[root@JHB0 flume3]$ ll