MapReduce and MachineLearning——笔记三(配置hadoop环境)
1.安装hadoop
用hadoop用户登录master节点,直接将hadoop压缩包解压到master节点的/home/hadoop目录下,我安装的的hadoop版本是hadoop-0.20.203.0,解压完毕后进入hadoop-0.20.203.0目录。
2.修改配置文件
2.1 进入conf目录,修改core-site.xml,加入以下内容:
<property>
<name>fs.default.name</name>
<value>hdfs://master:9000</value>
</property>
<property>
<name>hadoop.tmp.dir</name>
<value>hadoop_work/tmp/</value>
</property>
2.2 修改hdfs-site.xml,加入以下内容:
<property>
<name>dfs.name.dir</name>
<value>hadoop_work/name/</value>
</property>
<property>
<name>dfs.data.dir</name>
<value>hadoop_work/data/</value>
</property>
<property>
<name>dfs.replication</name>
<value>2</value>
</property>
<property>
<name>dfs.block.size</name>
<value>67108864</value>
</property>
<property>
<name>dfs.permissions</name>
<value>false</value>
</property>
<property>
<name>dfs.web.ugi</name>
<value>hadoop,supergroup</value>
</property>
2.3 修改mapred-site.xml
<property>
<name>mapred.job.tracker</name>
<value>hdfs://master:9001/</value>
</property>
<property>
<name>mapred.child.java.opts</name>
<value>-Xmx512M</value>
</property>
2.4 修改masters,slaves文件
在masters里面加入:
master
在salves文件中加入:
slave1
slave2
2.5 修改hadoop-env.sh
把JAVA_HOME改成本机的JAVA路径即可
3.将/home/hadoop/hadoop-0.20.203.0目录整个拷贝到另外两台机器
cd /home/hadoop
scp -r hadoop-0.20.203.0 hadoop@slave1:~/hadoop-0.20.203.0
scp -r hadoop-0.20.203.0 hadoop@slave2:~/hadoop-0.20.203.0
4.在master节点格式化DFS
cd /home/hadoop/hadoop-0.20.203.0
bin/hadoop namenode -format
5.关闭三台机子的防火墙
sudo /etc/init.d/iptables stop
6.测试
进入master的的hadoop-0.20.203.0目录,执行sh bin/start-all.sh