准备3台机器（node1,node2,node3），每台机器上安装openjdk1.8

3台机器配置无密码访问

hadoop-3.3.3.tar.gz放到3台机器上，解压

(三台)编辑etc/hadoop/hadoop-env.sh
export JAVA_HOME=/usr/lib/jvm/java-1.8.0/....
export HADOOP_HOME=/root/package/hadoop-3.3.3
export HADOOP_LOG_DIR={HADOOP_HOME}/pid

(三台) 编辑etc/hadoop/core-site.xml
<configuration>
<property>
<name>fs.defaultFS</name>
<value>hdfs://node1:9000</value>
</property>
<property>
<name>io.file.buffer.size</name>
<value>131072</value>
</property>
</configuration>

编辑etc/hadoop/hdfs-site.xml
（node1）
<configuration>
<property>
<name>dfs.namenode.name.dir</name>
<value>/root/package/hadoop-3.3.3/data/namenode</value>
</property>
<property>
<name>dfs.blocksize</name>
<value>134217728</value>
</property>
<property>
<name>dfs.datanode.data.dir</name>
<value>/root/package/hadoop-3.3.3/data/datanode</value>
</property>
</configuration>
===============
（node2,node3）
<configuration>
<property>
<name>dfs.datanode.data.dir</name>
<value>/root/package/hadoop-3.3.3/data/datanode</value>
</property>
</configuration>

（node1）如果以root启动，需要编辑sbin/start-dfs.sh，bin/stop-dfs.sh
添加
HDFS_NAMENODE_USER=root
HDFS_DATANODE_USER=root
HDFS_SECURE_DN_USER=hdfs
HDFS_SECONDARYNAMENODE_USER=root

（node1）启动hdfs的namenode和datanode
bin/hdfs namenode -format
sbin/start-dfs.sh

开放9000和9870,9866,9864端口

（node2,node3）启动datanode
bin/hdfs --daemon start datanode

配置yarn
(node1,node2,node3)
<configuration>

<property>
<name>yarn.nodemanager.local-dirs</name>
<value>/root/package/hadoop-3.3.3/data/yarn/local</value>
</property>
<property>
<name>yarn.nodemanager.log-dirs</name>
<value>/root/package/hadoop-3.3.3/data/yarn/logs</value>
</property>
<property>
<name>yarn.resourcemanager.hostname</name>
<value>node2</value>
</property>
</configuration>

(node2)启动yarn resourcemanager，开放8030，8031，8032，8033端口
bin/yarn --daemon start resourcemanager

(三台)启动yarn nodemanager，开发8042端口
bin/yarn --daemon start nodemanager