黑猴子的家：Hadoop 伪分布式HDFS运行MapReduce

1、配置集群

和本地模式一样安装

(1) 配置hadoop-env.sh

获取jdk 的安装路径
[victor@node1 hadoop-2.8.2]$ echo $JAVA_HOME
/opt/module/jdk1.8.0_144

修改JAVA_HOME 路径
[victor@node1 hadoop-2.8.2]$ vim etc/hadoop/hadoop-env.sh
export JAVA_HOME=/opt/module/jdk1.8.0_144

(2) 配置：core-site.xml

<!-- 指定HDFS中NameNode的地址 -->
<property>
<name>fs.defaultFS</name>
<value>hdfs://node1:9000</value>
</property>

<!-- 指定hadoop运行时产生文件的存储目录 -->
<property>
<name>hadoop.tmp.dir</name>
<value>/opt/module/hadoop-2.8.2/data/tmp</value>
</property>

(3) 配置：hdfs-site.xml

<!-- 指定HDFS副本的数量 -->
<property>
    <name>dfs.replication</name>
    <value>1</value>
</property>

2、启动集群

(1) 格式化namenode（第一次启动时格式化，以后就不要格式化）

[victor@node1 hadoop-2.8.2]$ bin/hdfs namenode -format

(2) 启动namenode

[victor@node1 hadoop-2.8.2]$ sbin/hadoop-daemon.sh start namenode

(3) 启动datanode

[victor@node1 hadoop-2.8.2]$ sbin/hadoop-daemon.sh start datanode

3、 查看集群

(1) 查看是否启动成功

[victor@node1 hadoop-2.8.2]$ jps
13256 NameNode
13278 DataNode
13645 Jps

(2) 查看产生的log日志

[victor@node1 hadoop-2.8.2]$ pwd
/opt/module/hadoop-2.8.2
[victor@node1 hadoop-2.8.2]$ ls logs/
hadoop-victor-datanode-hadoop.victor.com.log
hadoop-victor-datanode-hadoop.victor.com.out
hadoop-victor-namenode-hadoop.victor.com.log
hadoop-victor-namenode-hadoop.victor.com.out

(3) web端浏览器查看HDFS文件系统

http://node1:50070/dfshealth.html#tab-overview

4、操作集群

(1) 在hdfs文件系统上创建一个input文件夹

[victor@node1 hadoop-2.8.2]$ bin/hdfs dfs -mkdir -p /user/victor/wordcount/input

(2) 将测试文件内容上传到文件系统上

[victor@node1 hadoop-2.8.2]$bin/hdfs dfs -put wcinput/wc.input /user/victor/wordcount/input/

(3) 查看上传的文件是否正确

[victor@node1 hadoop-2.8.2]$ bin/hdfs dfs -ls /user/victor/wordcount/input/
[victor@node1 hadoop-2.8.2]$ bin/hdfs dfs -cat /user/victor/wordcount/input/wc.input

(4) 运行mapreduce程序

[victor@node1 hadoop-2.8.2]$bin/hadoop jar share/hadoop/mapreduce/hadoop-mapreduce-examples-2.8.2.jar wordcount /user/victor/wordcount/input/ /user/victor/wordcount/output

(5) 查看输出结果

命令行查看
[victor@node1 hadoop-2.8.2]$ bin/hdfs dfs -cat /user/victor/wordcount/output/*

(6) 将测试文件内容下载到本地

[victor@node1 hadoop-2.8.2]$ hadoop fs -get /user/victor/wordcount/output/part-r-00000 ./wcoutput/

(7) 删除HDFS上的输出结果

[victor@node1 hadoop-2.8.2]$ hdfs dfs -rmr /user/victor/wordcount/output