zoukankan      html  css  js  c++  java
  • CentOS7搭建Flume1.6环境

    写在最前,后面测试会用到HDFS2,所以本次环境搭建是在Hadoop2.6.1,三节点的基础上完成的。

    (关于搭建Hadoop环境,可参考:https://www.cnblogs.com/SysoCjs/p/10835793.html

    说明:

    集群环境:

    Master 192.168.112.10

    Slave1 192.168.112.11

    Slave2 192.168.112.12

    #master,表示在master节点上操作;

    #master,#slave1,#slave2,表示在三个节点上都要操作;

    一、准备工作

    下载软件包,到flume官网下载tgz压缩包,并解压:

    #Master

    cd /usr/local/src/
    wget http://mirror.bit.edu.cn/apache/flume/1.6.0/apache-flume-1.6.0-bin.tar.gz
    
    tar zxvf apache-flume-1.6.0-bin.tar.gz

    二、配置文件、测试

    cd ./conf/

    # NetCat测试

    vim flume-netcat.conf

    输入内容:

    # Name the components on this agent
    agent.sources = r1
    agent.sinks = k1
    agent.channels = c1
    
    # Describe/configuration the source
    agent.sources.r1.type = netcat
    agent.sources.r1.bind = 192.168.112.10
    agent.sources.r1.port = 44444
    
    # Describe the sink
    agent.sinks.k1.type = logger
    
    # Use a channel which buffers events in memory
    agent.channels.c1.type = memory
    agent.channels.c1.capacity = 1000
    agent.channels.c1.transactionCapacity = 100
    
    # Bind the source and sink to the channel
    agent.sources.r1.channels = c1
    agent.sinks.k1.channel = c1 

    验证:

    开启动一个master窗口,作为#Server

    bin/flume-ng agent --conf conf --conf-file conf/flume-netcat.conf --name=agent -Dflume.root.logger=INFO,console

    再开启一个master窗口,作为#client

    telnet master 44444

    # Exec测试

    vim flume-exec.conf

    输入内容:

    # Name the components on this agent
    agent.sources = r1
    agent.sinks = k1
    agent.channels = c1
    
    # Describe/configuration the source
    agent.sources.r1.type = exec
    agent.sources.r1.command = tail -f /usr/local/src/apache-flume-1.6.0-bin/data/hadoop/flume/test.txt
    
    # Describe the sink
    agent.sinks.k1.type = logger
    
    # Use a channel which buffers events in memory
    agent.channels.c1.type = memory
    agent.channels.c1.capacity = 1000
    agent.channels.c1.transactionCapacity = 100
    
    # Bind the source and sink to the channel
    agent.sources.r1.channels = c1
    agent.sinks.k1.channel = c1 

    # Server

    ./bin/flume-ng agent --conf conf --conf-file conf/flume-exec.conf --name=agent -Dflume.root.logger=INFO,console

    # Client

    while true;do echo `date` >> /usr/local/src/apache-flume-1.6.0-bin/data/hadoop/flume/test.txt ; sleep 1; done

    # Avro测试

    vim flume-avro.conf

    输入内容:

    # Define a memory channel called c1 on agent
    agent.channels.c1.type = memory
    
    # Define an avro source alled r1 on agent and  tell it
    agent.sources.r1.channels = c1
    agent.sources.r1.type = avro
    agent.sources.r1.bind = 192.168.112.10
    agent.sources.r1.port = 44444
    
    # Describe/configuration the source
    agent.sinks.k1.type = hdfs
    agent.sinks.k1.channel = c1
    agent.sinks.k1.hdfs.path = hdfs://master:9000/flume_data_pool
    agent.sinks.k1.hdfs.filePrefix = events-
    agent.sinks.k1.hdfs.fileType = DataStream
    agent.sinks.k1.hdfs.writeFormat = Text
    agent.sinks.k1.hdfs.rollSize = 0
    agent.sinks.k1.hdfs.rollCount= 600000
    agent.sinks.k1.hdfs.rollInterval = 600
    
    agent.channels = c1
    agent.sources = r1
    agent.sinks = k1

    # Server

    ./bin/flume-ng agent --conf conf --conf-file conf/flume-avro.conf --name=agent -Dflume.root.logger=DEBUG,console

    # client

    ./bin/flume-ng avro-client --conf conf -H master -p 44444 -F /home/badou/flume_test/monitor_source/3.txt -Dflume.root.logger=DEBUG,consol

    # Netcat To Avro测试

    原理:

    #master

    1、创建flume-netcat2avro.conf文件

    添加内容:

    # 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 = 192.168.112.10
    a1.sources.r1.port = 44444
    a1.sources.r1.channels = c1
    
    #Use a channel which buffers events in memory
    a1.channels.c1.type = memory
    a1.channels.c1.keep-alive = 10
    a1.channels.c1.capacity = 100000
    a1.channels.c1.transactionCapacity = 100000
    
    # Describe /configure the sink/next source
    a1.sinks.k1.type = avro
    a1.sinks.k1.channel = c1
    a1.sinks.k1.hostname = 192.168.112.11
    a1.sinks.k1.port = 44444

    2、在master的一个终端启动flume:

    ./bin/flume-ng agent -c conf -f conf/flume-netcat2avro.conf -n a1 -Dflume.root.logger=INFO,console

    #slave1

    3、创建flume-netcat2avro.conf文件

    添加内容:

    # Name the components on this agent
    a2.sources = r1
    a2.sinks = k1
    a2.channels = c1
    
    #Describe/configure the source
    a2.sources.r1.type = avro
    a2.sources.r1.channels = c1
    a2.sources.r1.bind = 192.168.112.11
    a2.sources.r1.port = 44444
    
    # Describe the sink
    a2.sinks.k1.type = logger
    a2.sinks.k1.channel = c1
    
    # Use a channel which buffers events in memory
    a2.channels.c1.type = memory
    a2.channels.c1.keep-alive = 10
    a2.channels.c1.capacity = 100000
    a2.channels.c1.transactionCapacity = 100000

    4、在slave1的一个终端启动flume:

    ./bin/flume-ng agent -c conf -f conf/flume-netcat2avro.conf -n a2 -Dflume.root.logger=INFO,console

    5、再打开一个master终端,作为消息生产者,因为master的source源类型是netcat,所以连接用telnet方式:

    telnet master:4444

    查看刚才启动flume的slave1终端:

  • 相关阅读:
    Java连接Mysql数据库异常:Public Key Retrieval is not allowed
    java8的时间段比较处理工具类TimeUtils
    MAVEN最常用的远程仓库
    maven的settings.xml配置阿里云中央仓库
    idea如何将java程序打包成exe可执行文件
    FakerUtil
    Golang内存逃逸是什么?怎么避免内存逃逸?
    10个高效Linux技巧及Vim命令对比[转]
    进程间8种通信方式详解
    基于openresty的URL 断路器/熔断器 -- URL-fuse
  • 原文地址:https://www.cnblogs.com/SysoCjs/p/11206494.html
Copyright © 2011-2022 走看看