一、jdk安装
1.准备工作
①.下载Hadoop-2.7.4和其对应的jdk版本
jdk-8u161:链接：https://pan.baidu.com/s/1Z1kDg_tgakA-Hx5VsakPQg 提取码：rel3
hadoop-2.7.4：链接：https://pan.baidu.com/s/1r5OwXFbuaByk45zADxfDNQ 提取码：nyt1
②.为系统设置静态IP
参考此文章：centos7如何设置静态IP_庸人自扰665的博客-CSDN博客_centos7配置静态ip
③.修改 /etc/hosts 文件
[root@hadoop01 ~]# vi /etc/hosts 127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4 ::1 localhost localhost.localdomain localhost6 localhost6.localdomain6 192.168.51.129 hadoop01 192.168.51.130 hadoop02 192.168.51.131 hadoop03
④.实现三台主机互信
[root@hadoop01 ~]# yum install openssh-server（三台主机上均进行安装） [root@hadoop01 ~]# ssh-keygen -t rsa(连续按四次回车键) [root@hadoop01 ~]# ssh-copy-id hadoop02 [root@hadoop01 ~]# ssh-copy-id hadoop03 [root@hadoop01 ~]# ssh-copy-id hadoop01 [root@hadoop01 ~]# scp -r /root/.ssh/ root@hadoop02:/root/.ssh [root@hadoop01 ~]# scp -r /root/.ssh/ root@hadoop03:/root/.ssh
2.开始安装jdk
①.新建/export/servers和/export/software文件夹
[root@hadoop01 ~]#mkdir -p /export/servers [root@hadoop01 ~]#mkdir -p /export/software [root@hadoop01 ~]#cd /export/software
②.通过工具上传到/export/software文件下
winscp下载地址：链接：https://pan.baidu.com/s/1cRxySrA84WDJm0vJW97t9A 提取码：7ovz
③.解压文件到/export/servers文件夹下
[root@hadoop01 ~]# tar -zxvf jdk-8u161-linux-x64.tar.gz -C /export/servers
④.对解压的文件进行重命名
[root@hadoop01 ~]# cd /export/servers [root@hadoop01 ~]#mv jdk1.8.0_161 jdk
⑤.在/etc/profile写入
[root@hadoop01 ~]# vi /etc/profile # JAVA_HOME export JAVA_HOME=/export/servers/jdk export PATH=$PATH:$JAVA_HOME/bin export CLASSPATH=.JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar 使命令生效 [root@hadoop01 ~]# source /etc/profile
⑥.查看jdk
[root@hadoop01 ~]# java -version
若出现jdk的版本即安装成功
二.hadoop-2.7.4的安装
1.解压文件
[root@hadoop01 ~]# cd /export/software [root@hadoop01 ~]# tar -zxvf hadoop-2.7.4.tar.gz -C /export/servers
2.在/etc/profile写入
[root@hadoop01 ~]# vi /etc/profile # HADOOP_HOME export HADOOP_HOME=/export/servers/hadoop-2.7.4 export PATH=$PATH:$HADOOP_HOME/bin:$HADOOP_HOME/sbin
3.查看Hadoop是否安装成功
[root@hadoop01 ~]# hadoop version
若出现Hadoop相关版本的信息说明安装成功
4.修改Hadoop文件
①.修改hadoop-env.sh文件
[root@hadoop01 ~]# cd /export/servers/hadoop-2.7.4/etc/hadoop/ [root@hadoop01 hadoop]# vi hadoop-env.sh # The java implementation to use. export JAVA_HOME=/export/servers/jdk
②.修改yarn-env.sh文件
[root@hadoop01 hadoop]# vi yarn-env.sh # some Java parameters export JAVA_HOME=/export/servers/jdk
③.修改core-site.xml文件
[root@hadoop01 hadoop]# vi core-site.xmlfs.defaultFS hdfs://hadoop01:9000 hadoop.tmp.dir /export/servers/hadoop-2.7.4/tmp
④。修改hdfs-site.xml文件
[root@hadoop01 hadoop]# vi hdfs-site.xmldfs.replication 3 dfs.namenode.secondary.http-address hadoop02:50090
⑤。修改mapred-site.xml文件
[root@hadoop01 hadoop]# cp mapred-site.xml.template mapred-site.xml [root@hadoop01 hadoop]# vi mapred-site.xmlmapreduce.framework.name yarn
⑥.修改yarn-site.xml文件
[root@hadoop01 hadoop]# vi yarn-site.xmlyarn.resourcemanager.hostname hadoop01 yarn.nodemanager.aux-services mapreduce_shuffle
⑦.修改slaves文件
[root@hadoop01 hadoop]# vi slaves hadoop01 hadoop02 hadoop03
5.将Hadoop01的配置分发给另外两台主机
[root@hadoop01 ~]# scp -r /etc/profile root@hadoop02:/etc/profile [root@hadoop01 ~]# scp -r /etc/profile root@hadoop03:/etc/profile 在hadoop02和hadoop03分别执行以下语句: [root@hadoop02 ~]# mkdir -p /export/servers 回到hadoop01: [root@hadoop01 ~]# scp -r /export/servers root@hadoop02:/export/ [root@hadoop01 ~]# scp -r /export/servers root@hadoop03:/export/ 在hadoop02和hadoop03中分别执行一下语句 [root@hadoop02 ~]# source /etc/profile
6.启动Hadoop集群
只在hadoop01上启动 [root@hadoop01 ~]# hdfs namenode -format [root@hadoop01 ~]# start-all.sh 查看启动情况hadoop01 [root@hadoop01 ~]# jps 14066 Jps 2389 DataNode 2759 NodeManager 2253 NameNode 2638 ResourceManager hadoop02: [root@hadoop02 ~]# jps 1584 SecondaryNameNode 1685 NodeManager 1527 DataNode 6011 Jps hadoop03:[root@hadoop03 ~]# jps 5016 Jps 1531 DataNode 1611 NodeManager
7.查看是否搭建完成（在hadoop01进行访问）
hadoop01:50070
hadoop02:8088
转到下一篇：
