官网：hadoop.apache.org
下载地址：
Apache Hadoophttps://hadoop.apache.org/releases.html
1.下载安装包
2.上传到Linux中/tmp目录中
[root@localhost ~]# cd software [root@localhost software]# rz
3.添加一个维护的用户和文件夹
[root@localhost software]# useradd ssn [root@localhost software]# id ssn uid=1001(ssn) gid=1001(ssn) groups=1001(ssn) [root@localhost software]# pwd /root/software [root@localhost software]# su - ssn
[ssn@localhost ~]$ mkdir source software app log data lib tmp [ssn@localhost ~]$ ls app data lib log software source tmp
创建tmp目录的原因：系统自带的tmp目录，定期不访问，根据当前版本特性，自动清除，默认是30天
4.移动安装包，并赋予权限
[ssn@localhost ~]$ exit [root@localhost software]# mv ~/software/hadoop-3.2.2.tar.gz /home/ssn/software/ [root@localhost software]# chown ssn:ssn /home/ssn/software
[ssn@localhost hadoop]$
e.NameNode daemon and DataNode daemon启动：（名称节点和数据节点启动）
[ssn@localhost hadoop]$ sbin/start-dfs.sh[ssn@localhost hadoop]$ sbin/start-dfs.sh
Starting namenodes on [localhost]
Starting datanodes
Starting secondary namenodes [localhost.localdomain]
localhost.localdomain: Warning: Permanently added 'localhost.localdomain' (ECDSA) to the list of known hosts.[ssn@localhost hadoop]$
jps查看服务有没有成功启动，但是不保险，建议还是使用ps -ef | grep hadoop
f.访问http://localhost:9870/
如果使用的云主机，需要配置安全组才能访问，id地址：9870，访问成功如图所示
g.
注意：
先用root用户执行
rm -rf /tmp/hadoop然后使用ssn用户执行
[ssn@localhost hadoop]$bin/hdfs dfs -mkdir /user [ssn@localhost hadoop]$bin/hdfs dfs -mkdir /user/ssn [ssn@localhost hadoop]$bin/hdfs dfs -mkdir input [ssn@localhost hadoop]$bin/hdfs dfs -put etc/hadoop/*.xml input [ssn@localhost hadoop]$bin/hadoop jar share/hadoop/mapreduce/hadoop-mapreduce-examples-3.2.2.jar grep input output 'dfs[a-z.]+' [ssn@localhost hadoop]$bin/hdfs dfs -get output output [ssn@localhost hadoop]$bin/hdfs dfs -cat output/* [ssn@localhost hadoop]$$ sbin/stop-dfs.sh
