Top

NSD ARCHITECTURE DAY05

1. 案例1：安装Hadoop
2. 案例2：安装配置Hadoop

1 案例1：安装Hadoop

1.1 问题

本案例要求安装单机模式Hadoop：

• 单机模式安装Hadoop
• 安装JAVA环境
• 设置环境变量，启动运行

1.2 步骤

实现此案例需要按照如下步骤进行。

步骤一：环境准备

1）配置主机名为nn01，ip为192.168.1.21，配置yum源（系统源）

备注：由于在之前的案例中这些都已经做过，这里不再重复，不会的学员可以参考之前的案例

2）安装java环境

1. [[email protected] ~]# yum -y install java-1.8.0-openjdk-devel
2. [[email protected] ~]# java -version
3. openjdk version "1.8.0_131"
4. OpenJDK Runtime Environment (build 1.8.0_131-b12)
5. OpenJDK 64-Bit Server VM (build 25.131-b12, mixed mode)
6. [[email protected] ~]# jps
7. 1235 Jps

3）安装hadoop

1. [[email protected] ~]# tar -xf hadoop-2.7.6.tar.gz
2. [[email protected] ~]# mv hadoop-2.7.6 /usr/local/hadoop
3. [[email protected] ~]# cd /usr/local/hadoop/
4. [[email protected] hadoop]# ls
5. bin include libexec     NOTICE.txt sbin
6. etc lib LICENSE.txt README.txt share
7. [[email protected] hadoop]# ./bin/hadoop //报错，JAVA_HOME没有找到
8. Error: JAVA_HOME is not set and could not be found.
9. [[email protected] hadoop]#

4）解决报错问题

1. [[email protected] hadoop]# rpm -ql java-1.8.0-openjdk
2. [[email protected] hadoop]# cd ./etc/hadoop/
3. [[email protected] hadoop]# vim hadoop-env.sh
4. 25 export \
5. JAVA_HOME="/usr/lib/jvm/java-1.8.0-openjdk-1.8.0.131-11.b12.el7.x86_64/jre"
6. 33 export HADOOP_CONF_DIR="/usr/local/hadoop/etc/hadoop"
7. [[email protected] ~]# cd /usr/local/hadoop/
8. [[email protected] hadoop]# ./bin/hadoop
9. Usage: hadoop [--config confdir] [COMMAND | CLASSNAME]
10. CLASSNAME run the class named CLASSNAME
11. or
12. where COMMAND is one of:
13. fs run a generic filesystem user client
14. version print the version
15. jar <jar> run a jar file
16. note: please use "yarn jar" to launch
17. YARN applications, not this command.
18. checknative [-a|-h] check native hadoop and compression libraries availability
19. distcp <srcurl> <desturl> copy file or directories recursively
20. archive -archiveName NAME -p <parent path> <src>* <dest> create a hadoop archive
21. classpath prints the class path needed to get the
22. credential interact with credential providers
23. Hadoop jar and the required libraries
24. daemonlog get/set the log level for each daemon
25. trace view and modify Hadoop tracing settings
26. Most commands print help when invoked w/o parameters.
27. [[email protected] hadoop]# mkdir /usr/local/hadoop/aa
28. [[email protected] hadoop]# ls
29. bin etc include lib libexec LICENSE.txt NOTICE.txt aa README.txt sbin share
30. [[email protected] hadoop]# cp *.txt /usr/local/hadoop/aa
31. [[email protected] hadoop]# ./bin/hadoop jar \
32. share/hadoop/mapreduce/hadoop-mapreduce-examples-2.7.6.jar wordcount aa bb     //wordcount为参数 统计aa这个文件夹，存到bb这个文件里面（这个文件不能存在，要是存在会报错，是为了防止数据覆盖）
33. [[email protected] hadoop]# cat bb/part-r-00000 //查看

2 案例2：安装配置Hadoop

2.1 问题

本案例要求：

• 另备三台虚拟机，安装Hadoop
• 使所有节点能够ping通，配置SSH信任关系
• 节点验证

2.2 方案

准备四台虚拟机，由于之前已经准备过一台，所以只需再准备三台新的虚拟机即可，安装hadoop，使所有节点可以ping通，配置SSH信任关系，如图-1所示：

图-1

2.3 步骤

实现此案例需要按照如下步骤进行。

步骤一：环境准备

1）三台机器配置主机名为node1、node2、node3，配置ip地址（ip如图-1所示），yum源（系统源）

2）编辑/etc/hosts（四台主机同样操作，以nn01为例）

1. [[email protected] ~]# vim /etc/hosts
2. 192.168.1.21 nn01
3. 192.168.1.22 node1
4. 192.168.1.23 node2
5. 192.168.1.24 node3

3）安装java环境，在node1，node2，node3上面操作（以node1为例）

1. [[email protected] ~]# yum -y install java-1.8.0-openjdk-devel

4）布置SSH信任关系

1. [[email protected] ~]# vim /etc/ssh/ssh_config //第一次登陆不需要输入yes
2. Host *
3. GSSAPIAuthentication yes
4. StrictHostKeyChecking no
5. [[email protected] .ssh]# ssh-keygen
6. Generating public/private rsa key pair.
7. Enter file in which to save the key (/root/.ssh/id_rsa):
8. Enter passphrase (empty for no passphrase):
9. Enter same passphrase again:
10. Your identification has been saved in /root/.ssh/id_rsa.
11. Your public key has been saved in /root/.ssh/id_rsa.pub.
12. The key fingerprint is:
13. SHA256:Ucl8OCezw92aArY5+zPtOrJ9ol1ojRE3EAZ1mgndYQM [email protected]
14. The key‘s randomart image is:
15. +---[RSA 2048]----+
16. | o*E*=. |
17. | +XB+. |
18. | ..=Oo. |
19. | o.+o... |
20. | .S+.. o |
21. | + .=o |
22. | o+oo |
23. | o+=.o |
24. | o==O. |
25. +----[SHA256]-----+
26. [[email protected] .ssh]# for i in 21 22 23 24 ; do ssh-copy-id 192.168.1.$i; done
27. //部署公钥给nn01，node1，node2，node3

5）测试信任关系

1. [[email protected] .ssh]# ssh node1
2. Last login: Fri Sep 7 16:52:00 2018 from 192.168.1.21
3. [[email protected] ~]# exit
4. logout
5. Connection to node1 closed.
6. [[email protected] .ssh]# ssh node2
7. Last login: Fri Sep 7 16:52:05 2018 from 192.168.1.21
8. [[email protected] ~]# exit
9. logout
10. Connection to node2 closed.
11. [[email protected] .ssh]# ssh node3

步骤二：配置hadoop

1）修改slaves文件

1. [[email protected] ~]# cd /usr/local/hadoop/etc/hadoop
2. [[email protected] hadoop]# vim slaves
3. node1
4. node2
5. node3

2）hadoop的核心配置文件core-site

1. [[email protected] hadoop]# vim core-site.xml
2. <configuration>
3. <property>
4. <name>fs.defaultFS</name>
5. <value>hdfs://nn01:9000</value>
6. </property>
7. <property>
8. <name>hadoop.tmp.dir</name>
9. <value>/var/hadoop</value>
10. </property>
11. </configuration>
12. [[email protected] hadoop]# mkdir /var/hadoop        //hadoop的数据根目录
13. [[email protected] hadoop]# ssh node1 mkdir /var/hadoop
14. [[email protected] hadoop]# ssh node2 mkdir /var/hadoop
15. [[email protected] hadoop]# ssh node3 mkdir /var/hadoop

3）配置hdfs-site文件

1. [[email protected] hadoop]# vim hdfs-site.xml
2. <configuration>
3. <property>
4. <name>dfs.namenode.http-address</name>
5. <value>nn01:50070</value>
6. </property>
7. <property>
8. <name>dfs.namenode.secondary.http-address</name>
9. <value>nn01:50090</value>
10. </property>
11. <property>
12. <name>dfs.replication</name>
13. <value>2</value>
14. </property>
15. </configuration>

4）同步配置到node1，node2，node3

1. [[email protected] hadoop]# yum –y install rsync //同步的主机都要安装rsync
2. [[email protected] hadoop]# for i in 22 23 24 ; do rsync -aSH --delete /usr/local/hadoop/
3. \ 192.168.1.$i:/usr/local/hadoop/ -e ‘ssh‘ & done
4. [1] 23260
5. [2] 23261
6. [3] 23262

5）查看是否同步成功

1. [[email protected] hadoop]# ssh node1 ls /usr/local/hadoop/
2. bin
3. etc
4. include
5. lib
6. libexec
7. LICENSE.txt
8. NOTICE.txt
9. bb
10. README.txt
11. sbin
12. share
13. aa
14. [[email protected] hadoop]# ssh node2 ls /usr/local/hadoop/
15. bin
16. etc
17. include
18. lib
19. libexec
20. LICENSE.txt
21. NOTICE.txt
22. bb
23. README.txt
24. sbin
25. share
26. aa
27. [[email protected] hadoop]# ssh node3 ls /usr/local/hadoop/
28. bin
29. etc
30. include
31. lib
32. libexec
33. LICENSE.txt
34. NOTICE.txt
35. bb
36. README.txt
37. sbin
38. share
39. aa

步骤三：格式化

1. [[email protected] hadoop]# cd /usr/local/hadoop/
2. [[email protected] hadoop]# ./bin/hdfs namenode -format         //格式化 namenode
3. [[email protected] hadoop]# ./sbin/start-dfs.sh        //启动
4. [[email protected] hadoop]# jps        //验证角色
5. 23408 NameNode
6. 23700 Jps
7. 23591 SecondaryNameNode
8. [[email protected] hadoop]# ./bin/hdfs dfsadmin -report        //查看集群是否组建成功
9. Live datanodes (3):        //有三个角色成功

原文地址：https://www.cnblogs.com/tiki/p/10785614.html