1) JDK -- Hadoop是用Java写的，不安装Java虚拟机怎么运行Hadoop的程序；
2）创建专门用于运行和执行hadoop任务（比如map和reduce任务）的linux用户，就像windows下的服务账户，并且授权给他可以访问运行JDK目录权限，让他可以执行java虚拟机。这个账户最后用来运行bin\.start_all.sh启动hadoop的所有服务，那运行账户肯定是他了，他肯定要有足够的权限。再者，需要配置这个账户的个人环境变量，把Java虚拟机的主目录地址环境变量之一，不然后面运行hadoop任务肯定报错，根本在当前的环境变量下找不到要运行的java命令。
2）修改/etc/hosts文件和/etc/hostname -- 集群模式下机器需要互相通信，靠的是IP地址。而通常我们利用的是主机名称，然后通过IP地址和主机名称的映射来和目标主机通信。那么我们自然就得改/etc/hosts文件和/etc/hostname，前者是主机和IP地址的映射关系，后者是主机的本地主机名；
4）SSH -- 集群机器间互相访问各自的资源需要建立连接通讯，使用的是SSH协议进行安全通道的数据交换，再者也需要利用SSH的授权证书办法来实现免密码登陆目标主机；
4）然后就是安装hadoop了。需要修改几个配置文件。hadoop的配置文件有很多种。
只读型配置文件：src/core/core-default.xml, src/hdfs/hdfs-default.xml, src/mapred/mapred-default.xml, conf/mapred-queues.xml。
定位设置：conf/core-site.xml, conf/hdfs-site.xml, conf/mapred-site.xml, conf/mapred-queues.xml。这种文件一般用于配置hadoop一些核心功能，比如hdfs和mapred的目录信息。
环境配置：conf/Hadoop-env.sh
说回正题，既然hadoop的核心是hdfs和mapreduce，那你至少要配置hdfs的namenode目录位置，datanode的目录位置，mapreduce的job tracker和task tracker通信的端口号，系统目录和本地目录等等。
这些在master和slave机子上都是相同的。
5）配置完后就是格式化hdfs了。在master机上格式化hdfs系统。
6）格式化完hdfs系统就启动所有的hdfs进程。
搭建环境用的是VMWare Workstation 12。这里用了三台Linux虚拟机：master，slave1，slave2。
具体配置：
master slave1 slave2OS Ubuntu 14.04 LTE x64 Ubuntu 14.04 LTE x64 Ubuntu 14.04 LTE x64
memory 1GB 1GB 1GB
hard drive space 20GB 20GB 20GB
processors 2 2 2
IP Address 192.168.2.110 192.168.2.111 192.168.2.112
Roles NameNode
DataNode
JobTracker
TaskTracker
SecondaryNameNode DataNode
TaskTracker DataNode
TaskTracker
hadoop directory /opt/hadoop /opt/hadoop /opt/hadoop
jdk version JDK 1.8 JDK 1.8 JDK 1.8
先在master机子上安装JDK和SSH
1. JDK安装
jerry@ubuntu:/run/network$ scp jerry@192.168.2.100:/home/jerry/Download/jdk-8u65-linux-x64.tar.gz ~
The authenticity of host '192.168.2.100 (192.168.2.100)' can't be established.
ECDSA key fingerprint is da:b7:c3:2a:ea:a2:76:4c:c3:c1:68:ca:0e:c2:ea:92.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added '192.168.2.100' (ECDSA) to the list of known hosts.
jerry@192.168.2.100's password:
scp: /home/jerry/Download/jdk-8u65-linux-x64.tar.gz: No such file or directory
jerry@ubuntu:/run/network$ scp jerry@192.168.2.100:/home/jerry/Downloads/jdk-8u65-linux-x64.tar.gz ~
jerry@192.168.2.100's password:
jdk-8u65-linux-x64.tar.gz 100% 173MB 21.6MB/s 00:08
jerry@ubuntu:/run/network$ cd ~
jerry@ubuntu:~$ ls
Desktop Downloads jdk-8u65-linux-x64.tar.gz Pictures Templates
Documents examples.desktop Music Public Videos
jerry@ubuntu:~$ cd .
jerry@ubuntu:~$ cd /
jerry@ubuntu:/$ ls
bin dev initrd.img lost+found opt run sys var
boot etc lib media proc sbin tmp vmlinuz
cdrom home lib64 mnt root srv usr
jerry@ubuntu:/$ sudo mkdir jvm
[sudo] password for jerry:
jerry@ubuntu:/$ rm jvm/
rm: cannot remove ‘jvm/’: Is a directory
jerry@ubuntu:/$ rm -d jvm/
rm: remove write-protected directory ‘jvm/’? y
rm: cannot remove ‘jvm/’: Permission denied
jerry@ubuntu:/$ ls
bin dev initrd.img lib64 mnt root srv usr
boot etc jvm lost+found opt run sys var
cdrom home lib media proc sbin tmp vmlinuz
jerry@ubuntu:/$ sudo mkdir /usr/lib/jvm
jerry@ubuntu:/$ cd ~
jerry@ubuntu:~$ sudo tar zxf ./jdk-8u65-linux-x64.tar.gz -C /usr/lib/jvm/
jerry@ubuntu:~$ cd /usr/lib/jvm/
jerry@ubuntu:/usr/lib/jvm$ sudo mv jdk1.8.0_65 java
jerry@ubuntu:/usr/lib/jvm$ cd java/
jerry@ubuntu:/usr/lib/jvm/java$ sudo vim ~/.bashrc
jerry@ubuntu:/usr/lib/jvm/java$ tail -n 4 ~/.bashrc
export JAVA_HOME=/usr/lib/jvm/java
export JRE_HOME=${JAVA_HOME}/jre
export CLASSPATH=.:${JAVA_HOME}/lib:${JRE_HOME}/lib
export PATH=${JAVA_HOME}/bin:$PATH
jerry@ubuntu:/usr/lib/jvm/java$ java -version
The program 'java' can be found in the following packages:
* default-jre
* gcj-4.8-jre-headless
* openjdk-7-jre-headless
* gcj-4.6-jre-headless
* openjdk-6-jre-headless
Try: sudo apt-get install <selected package>
jerry@ubuntu:/usr/lib/jvm/java$ sudo update-alternatives --install /usr/bin/java java /usr/lib/jvm/java/bin/java 300
update-alternatives: using /usr/lib/jvm/java/bin/java to provide /usr/bin/java (java) in auto mode
jerry@ubuntu:/usr/lib/jvm/java$ sudo update-alternatives --install /usr/bin/javac javac /usr/lib/jvm/java/bin/javac 300
update-alternatives: using /usr/lib/jvm/java/bin/javac to provide /usr/bin/javac (javac) in auto mode
jerry@ubuntu:/usr/lib/jvm/java$ sudo update-alternatives --install /usr/bin/jar jar /usr/lib/jvm/java/bin/jar 300
update-alternatives: using /usr/lib/jvm/java/bin/jar to provide /usr/bin/jar (jar) in auto mode
jerry@ubuntu:/usr/lib/jvm/java$ sudo update-alternatives --install /usr/bin/javah javah /usr/lib/jvm/java/bin/javah 300
update-alternatives: using /usr/lib/jvm/java/bin/javah to provide /usr/bin/javah (javah) in auto mode
jerry@ubuntu:/usr/lib/jvm/java$ sudo update-alternatives --install /usr/bin/javap javap /usr/lib/jvm/java/bin/javap 300
update-alternatives: using /usr/lib/jvm/java/bin/javap to provide /usr/bin/javap (javap) in auto mode
jerry@ubuntu:/usr/lib/jvm/java$ sudo update-alternatives --config java
There is only one alternative in link group java (providing /usr/bin/java): /usr/lib/jvm/java/bin/java
Nothing to configure.
jerry@ubuntu:/usr/lib/jvm/java$ java -version
java version "1.8.0_65"
Java(TM) SE Runtime Environment (build 1.8.0_65-b17)
Java HotSpot(TM) 64-Bit Server VM (build 25.65-b01, mixed mode)
jerry@ubuntu:/usr/lib/jvm/java$
2. 添加hadoop用户和用户组