1.修改主机名
2.修改IP地址
3.添加主机名与ip的映射
dataNodeOneIP dataNodeOne
dataNodeTwoIP dataNodeTwo
nameNodeIP nameNode
slaveNodeIP slaveNode
4.防火墙:
查看:service iptables status
关闭:service iptables stop
查看开启状态:chkconfig iptables --list
关闭开机启动:chkconfig iptables off
验证:service iptables status
5.拷贝安装包:
6.关闭:selinux
验证:sestatus -v
没有修改成功就reboot
7.在隐藏文件中导入jdk环境变量:/home/用户名/.bash_profile (.bash_profile隐藏文件)
export JAVA_HOME=/home/用户名/setupPackage/jdk1.7.0_79
export JRE_HOME=/home/用户名/setupPackage/jdk1.7.0_79/jre
export CLASSPATH=$CLASSPATH:$JAVA_HOME/lib/tools.jar:$JAVA_HOME/lib/dt.jar:$JRE_HOME/lib
export PATH=$PATH:$JAVA_HOME/bin:$JRE_HOME/bin
生效环境变量:source /home/用户名/.bash_profile
检查jdk版本:java -version
分发/home/用户名/.bash_profile到集群其他机器,并逐个生效环境变量;
8.建立互信:
cd ~/.ssh
ssh-keygen -t rsa
#把公钥发送到authorized_keys (默认)
ssh-copy-id -i /home/用户名/.ssh/id_rsa.pub dataNodeOne
ssh-copy-id -i /home/用户名/.ssh/id_rsa.pub dataNodeTwo
ssh-copy-id -i /home/用户名/.ssh/id_rsa.pub nameNode
ssh-copy-id -i /home/用户名/.ssh/id_rsa.pub slaveNode
查看公钥:more authorized_keys
验证:ssh 集群主机名
9.在nameNode上安装Hadoop与zookeeper;
10.分发安装好的Hadoop与zookeeper 到 集群其他机器
scp -r /home/用户名/setupPackage/hadoop-2.7.3 slaveNodeIP:/home/用户名/setupPackage/
scp -r /home/用户名/setupPackage/hadoop-2.7.3 dataNodeOneIP:/home/用户名/setupPackage/
scp -r /home/用户名/setupPackage/hadoop-2.7.3 dataNodeTwoIP:/home/用户名/setupPackage/
scp -r /home/用户名/setupPackage/zookeeper-3.4.6 nameNodeIP:/home/用户名/setupPackage/
scp -r /home/用户名/setupPackage/zookeeper-3.4.6 dataNodeOneIP:/home/用户名/setupPackage/
scp -r /home/用户名/setupPackage/zookeeper-3.4.6 slaveNodeIP:/home/用户名/setupPackage/
11.启动zookeeper