Hadoop环境搭建及wordcount程序

    目的： 前期学习了一些机器学习基本算法，实际企业应用中算法是核心，运行的环境和数据处理的平台是基础。

    手段： 搭建简易hadoop集群（由于机器限制在自己的笔记本上通过虚拟机搭建）

一、基础环境介绍

---

win10

vmware15.0.0

3 ubuntu 虚拟机（1 台作为master ，另外2台作为 slave1、slave2）

hadoop2.8.5

jdk1.8

二、搭建步骤

---

1. 安装vmware ，安装ubuntu 先安装一台，后面配置完成后直接克隆 （此处不作详细介绍，可参考其它文档进行搭建）

2. linux基础环境配置

a) 创建用户 test 执行所有安装相关操作 :

          sudo useradd -m test -s /bin/bash

      sudo passwd hadoop

b）安装基础软件

1. 基础工具
      . sudo apt-get install vim    (edit tools)

      . sudo apt-get install openssh-client openssh-server  (openssh service for log in the server via ssh)

      . sudo apt-get install nfs-common  (for nfs mounting )

      . sudo apt-get install git (for git tool)

2.Setup nfs service on Ubuntu for mounting 
      . sudo apt-get install nfs-kernel-server       (install nfs server)
      
      . sudo mkdir /nfsroot; 
      
      . sudo chmod 777 /nfsroot         ( create /nfsroot fold as mounting directory)

      . sudo vim /etc/exports           (config the mount directory)

       add below line in /etc/exports: 
        
         /nfsroot *(rw,sync,no_root_squash)
    
     . sudo service nfs-kernel-server restart  (restart nfs service)
     
3. setup samba service for share folders with windows OS
     . sudo apt-get install samba smbclient     (install necessay tools)

     . sudo apt-get install samba smbclient      (config the samba server)

     . Add following lines in /etc/samba//smb.conf:
     
        [nfsroot]
        comment = nfsroot
        path = /nfsroot
        public = yes
        guest ok = yes
        browseable = yes
        writeable = yes
        
    . sudo service smbd restart  (restart the samba service)

c) 配置服务器之间免密互相访问（通过公钥私钥的方式）

       ssh-keygen -t rsa # 会有提示，都按回车就可以

       cat id_rsa.pub >> authorized_keys # 加入授权

      当所有节点都克隆完成后可以测试ssh登录：  ssh 192.168.xx.xxx@test   

3. 配置java和hadoop软件 

         下载jdk1.8                  解压文件放在 /opt/java 目录下，并配置环境变量 （java –version 进行测试）

         下载hadoop2.8.5         解压文件放在 /opt/hadoop 目录下，并配置环境变量 （hadoop version 进行测试）

4. 克隆当前版本的linux

      vmware有克隆虚拟机的功能，会将所有配置进行克隆

      配置每台机器的域名

sudo hostname master  (主节点)

sudo hostname slave1 （从节点）

sudo hostname slave2（从节点）

     配置每台机器的固定ip地址，并增加域名解析配置： vim /etc/hosts  文件增加如下配置：

127.0.0.1       localhost

192.168.61.100   master
192.168.61.101   slave1
192.168.61.102   slave2

  这里可以先配置一台，然后通过scp命令将配置复制到其他两台机器上去，后面的hdfs、yarn、MapReduce的配置同样如此。

5. 配置HDFS

       到hadoop安装目录下配置： ./etc/hadoop/core-site.xml

<configuration>
 <property>
                <name>hadoop.tmp.dir</name>
                <value>file:/home/test/hadoop-2.8.5/hdfs/tmp</value>
                <description>A base for other temporary directories.</description>
        </property>

        <property>
                <name>io.file.buffer.size</name>
                <value>131072</value>
        </property>
        <property>
                <name>fs.defaultFS</name>
                <value>hdfs://master:9000</value>
        </property>
</configuration>

配置hdfs： vim ./etc/hadoop/hdfs-site.xml

<configuration>
<property>
<name>dfs.replication</name>
  <value>2</value>
</property>
<property>
  <name>dfs.namenode.name.dir</name>
  <value>file:/opt/hadoop-2.8.5/hdfs/name</value>
  <final>true</final>
</property>
<property>
  <name>dfs.datanode.data.dir</name>
  <value>file:/opt/hadoop-2.8.5/hdfs/data</value>
  <final>true</final>
</property>
<property>
  <name>dfs.namenode.secondary.http-address</name>
  <value>master:9001</value>
</property>
<property>
  <name>dfs.webhdfs.enabled</name>
  <value>true</value>
</property>
<property>
  <name>dfs.permissions</name>
  <value>false</value>
</property>
</configuration>

6. 配置yarn

<configuration>

<!-- Site specific YARN configuration properties -->
<property>
<name>yarn.resourcemanager.address</name>
  <value>master:18040</value>
</property>
<property>
  <name>yarn.resourcemanager.scheduler.address</name>
  <value>master:18030</value>
</property>
<property>
  <name>yarn.resourcemanager.webapp.address</name>
  <value>master:18088</value>
</property>
<property>
  <name>yarn.resourcemanager.resource-tracker.address</name>
  <value>master:18025</value>
</property>
<property>
  <name>yarn.resourcemanager.admin.address</name>
  <value>master:18141</value>
</property>
<property>
  <name>yarn.nodemanager.aux-services</name>
  <value>mapreduce_shuffle</value>
</property>
<property>
  <name>yarn.nodemanager.aux-services.mapreduce.shuffle.class</name>
  <value>org.apache.hadoop.mapred.ShuffleHandler</value>
</property>
<property>
<name>yarn.nodemanager.resource.memory-mb</name>
<value>1024</value>
</property>


<property>
<name>yarn.nodemanager.vmem-check-enabled</name>
<value>false</value>
</property>

<property>
<name>yarn.nodemanager.vmem-pmem-ratio</name>
<value>3.0</value>
</property>


<property>
<name>yarn.nodemanager.resource.cpu-vcores</name>
<value>1</value>
</property>

<property>         

     <name>yarn.nodemanager.localizer.address</name>

     <value>0.0.0.0:8040</value>     

</property>     

<property>         

<description>The address of the container manager in the NM.</description>         

<name>yarn.nodemanager.address</name>         

<value>0.0.0.0:8041</value>     

</property>     

<property>         

<description>NM Webapp address.</description>         

<name>yarn.nodemanager.webapp.address</name>         

<value>0.0.0.0:8042</value>     

</property>
</configuration>

7.  配置mapreduce

<configuration>
<property>
  <name>mapreduce.framework.name</name>
  <value>yarn</value>
</property>

<property>
　　<name>yarn.app.mapreduce.am.resource.mb</name>
　　<value>1024</value>
</property>

<property>
　　<name>mapreduce.map.memory.mb</name>
　　<value>1024</value>
</property>

<property>
　　<name>mapreduce.reduce.memory.mb</name>
　　<value>1024</value>
</property>

</configuration>

8. 测试：

在master节点上运行 ./sbin/start-all.sh 

通过jps 可以查看 master上的namenode和slave上的datanode  (结果如下)

test@master:/opt/hadoop-2.8.5$ jps
8960 Jps
7940 NameNode
8373 ResourceManager
8206 SecondaryNameNode

slave2上运行结果如下：

test@slave2:/opt/hadoop-2.8.5/logs$ jps
7301 Jps
6938 NodeManager
6767 DataNode

三、wordcount程序

         在运行完start-all.sh脚本后。  就可以运行hadoop自带的wordcount程序了。

1. 上传文件到hdfs的wc_input中

2. 执行实例程序

./bin/yarn jar share/hadoop/mapreduce/hadoop-mapreduce-examples-2.8.5.jar wordcount /wc_input /wc_output.out7

3. 执行结果如下：

18/10/21 16:13:18 INFO client.RMProxy: Connecting to ResourceManager at master/192.168.61.100:18040
18/10/21 16:13:20 INFO input.FileInputFormat: Total input files to process : 2
18/10/21 16:13:20 INFO mapreduce.JobSubmitter: number of splits:2
18/10/21 16:13:20 INFO mapreduce.JobSubmitter: Submitting tokens for job: job_1540109557238_0001
18/10/21 16:13:21 INFO impl.YarnClientImpl: Submitted application application_1540109557238_0001
18/10/21 16:13:21 INFO mapreduce.Job: The url to track the job: http://master:18088/proxy/application_1540109557238_0001/
18/10/21 16:13:21 INFO mapreduce.Job: Running job: job_1540109557238_0001
18/10/21 16:13:35 INFO mapreduce.Job: Job job_1540109557238_0001 running in uber mode : false
18/10/21 16:13:35 INFO mapreduce.Job:  map 0% reduce 0%
18/10/21 16:13:42 INFO mapreduce.Job:  map 50% reduce 0%
18/10/21 16:13:46 INFO mapreduce.Job:  map 100% reduce 0%
18/10/21 16:13:51 INFO mapreduce.Job:  map 100% reduce 100%
18/10/21 16:13:52 INFO mapreduce.Job: Job job_1540109557238_0001 completed successfully
18/10/21 16:13:52 INFO mapreduce.Job: Counters: 49
        File System Counters
                FILE: Number of bytes read=93
                FILE: Number of bytes written=473483
                FILE: Number of read operations=0
                FILE: Number of large read operations=0
                FILE: Number of write operations=0
                HDFS: Number of bytes read=242
                HDFS: Number of bytes written=39
                HDFS: Number of read operations=9
                HDFS: Number of large read operations=0
                HDFS: Number of write operations=2
        Job Counters
                Launched map tasks=2
                Launched reduce tasks=1
                Data-local map tasks=2
                Total time spent by all maps in occupied slots (ms)=7691
                Total time spent by all reduces in occupied slots (ms)=3635
                Total time spent by all map tasks (ms)=7691
                Total time spent by all reduce tasks (ms)=3635
                Total vcore-milliseconds taken by all map tasks=7691
                Total vcore-milliseconds taken by all reduce tasks=3635
                Total megabyte-milliseconds taken by all map tasks=7875584
                Total megabyte-milliseconds taken by all reduce tasks=3722240
        Map-Reduce Framework
                Map input records=3
                Map output records=8
                Map output bytes=71
                Map output materialized bytes=99
                Input split bytes=203
                Combine input records=8
                Combine output records=8
                Reduce input groups=6
                Reduce shuffle bytes=99
                Reduce input records=8
                Reduce output records=6
                Spilled Records=16
                Shuffled Maps =2
                Failed Shuffles=0
                Merged Map outputs=2
                GC time elapsed (ms)=178
                CPU time spent (ms)=2180
                Physical memory (bytes) snapshot=721473536
                Virtual memory (bytes) snapshot=5936779264
                Total committed heap usage (bytes)=474480640
        Shuffle Errors
                BAD_ID=0
                CONNECTION=0
                IO_ERROR=0
                WRONG_LENGTH=0
                WRONG_MAP=0
                WRONG_REDUCE=0
        File Input Format Counters
                Bytes Read=39
        File Output Format Counters
                Bytes Written=39

注： 配置、安装、执行过程中不可避免遇到问题，需要学会看log解决问题。

参考： https://blog.csdn.net/xiao_bai_9527/article/details/79167562

https://blog.csdn.net/qinzhaokun/article/details/47804923