快速搭建 Hadoop 环境
对于Hadoop来说，最主要的是两个方面，一个是分布式文件系统HDFS，另一个是MapReduce计算模型，下面讲解下我在搭建Hadoop 环境过程。Hadoop 测试环境
共4台测试机，1台namenode 3台datanode OS版本：RHEL 5.5 X86_64 Hadoop：0.20.203.0 Jdk：jdk1.7.0 角色 ip地址 namenode 192.168.57.75 datanode1 192.168.57.76 datanode2 192.168.57.78 datanode3 192.168.57.79
一 部署 Hadoop 前的准备工作
1 需要知道hadoop依赖Java和SSH Java 1.5.x (以上)，必须安装。 ssh 必须安装并且保证 sshd 一直运行，以便用Hadoop 脚本管理远端Hadoop守护进程。 2 建立 Hadoop 公共帐号 所有的节点应该具有相同的用户名，可以使用如下命令添加： useradd hadoop passwd hadoop 3 配置 host 主机名 tail -n 3 /etc/hosts 192.168.57.75 namenode 192.168.57.76 datanode1 192.168.57.78 datanode2 192.168.57.79 datanode3 4 以上几点要求所有节点(namenode|datanode)配置全部相同
二 ssh 配置
1 生成私匙 id_rsa 与 公匙 id_rsa.pub 配置文件
[hadoop@hadoop1 ~]$ ssh-keygen -t rsa
Generating public/private rsa key pair.
Enter file in which to save the key (/home/hadoop/.ssh/id_rsa):
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /home/hadoop/.ssh/id_rsa.
Your public key has been saved in /home/hadoop/.ssh/id_rsa.pub.
The key fingerprint is:
d6:63:76:43:e2:5b:8e:85:ab:67:a2:7c:a6:8f:23:f9 hadoop@hadoop1.test.com
2 私匙 id_rsa 与 公匙 id_rsa.pub 配置文件
[hadoop@hadoop1 ~]$ ls .ssh/
authorized_keys id_rsa id_rsa.pub known_hosts
3 把公匙文件上传到datanode服务器
[hadoop@hadoop1 ~]$ ssh-copy-id -i ~/.ssh/id_rsa.pub hadoop@datanode1
28
hadoop@datanode1's password:
Now try logging into the machine, with "ssh 'hadoop@datanode1'", and check in:
.ssh/authorized_keys
to make sure we haven't added extra keys that you weren't expecting.
[hadoop@hadoop1 ~]$ ssh-copy-id -i ~/.ssh/id_rsa.pub hadoop@datanode2
28
hadoop@datanode2's password:
Now try logging into the machine, with "ssh 'hadoop@datanode2'", and check in:
.ssh/authorized_keys
to make sure we haven't added extra keys that you weren't expecting.
[hadoop@hadoop1 ~]$ ssh-copy-id -i ~/.ssh/id_rsa.pub hadoop@datanode3
28
hadoop@datanode3's password:
Now try logging into the machine, with "ssh 'hadoop@datanode3'", and check in:
.ssh/authorized_keys
to make sure we haven't added extra keys that you weren't expecting.
[hadoop@hadoop1 ~]$ ssh-copy-id -i ~/.ssh/id_rsa.pub hadoop@localhost
28
hadoop@localhost's password:
Now try logging into the machine, with "ssh 'hadoop@localhost'", and check in:
.ssh/authorized_keys
to make sure we haven't added extra keys that you weren't expecting.
4 验证
[hadoop@hadoop1 ~]$ ssh datanode1
Last login: Thu Feb 2 09:01:16 2012 from 192.168.57.71
[hadoop@hadoop2 ~]$ exit
logout
[hadoop@hadoop1 ~]$ ssh datanode2
Last login: Thu Feb 2 09:01:18 2012 from 192.168.57.71
[hadoop@hadoop3 ~]$ exit
logout
[hadoop@hadoop1 ~]$ ssh datanode3
Last login: Thu Feb 2 09:01:20 2012 from 192.168.57.71
[hadoop@hadoop4 ~]$ exit
logout
[hadoop@hadoop1 ~]$ ssh localhost
Last login: Thu Feb 2 09:01:24 2012 from 192.168.57.71
[hadoop@hadoop1 ~]$ exit
logout
三 java环境配置
1 下载合适的jdk
//此文件为64Linux 系统使用的 RPM包
wget http://download.oracle.com/otn-pub/java/jdk/7/jdk-7-linux-x64.rpm
2 安装jdk
rpm -ivh jdk-7-linux-x64.rpm
3 验证java
[root@hadoop1 ~]# java -version
java version "1.7.0"
Java(TM) SE Runtime Environment (build 1.7.0-b147)
Java HotSpot(TM) 64-Bit Server VM (build 21.0-b17, mixed mode)
[root@hadoop1 ~]# ls /usr/java/
default jdk1.7.0 latest
4 配置java环境变量
vim /etc/profile //在profile文件中加入如下信息：
add for hadoop
export JAVA_HOME=/usr/java/jdk1.7.0
export CLASSPATH=.:JAVA_HOME/lib/dt.jar:$JAVA_HOME/
export PATH=JAVA_HOME/bin
//使环境变量生效
source /etc/profile
5 拷贝 /etc/profile 到 datanode
[root@hadoop1 src]# scp /etc/profile root@datanode1:/etc/
The authenticity of host 'datanode1 (192.168.57.86)' can't be established.
RSA key fingerprint is b5:00:d1:df:73:4c:94:f1:ea:1f:b5:cd:ed:3a:cc:e1.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added 'datanode1,192.168.57.86' (RSA) to the list of known hosts.
root@datanode1's password:
profile 100% 1624 1.6KB/s 00:00
[root@hadoop1 src]# scp /etc/profile root@datanode2:/etc/
The authenticity of host 'datanode2 (192.168.57.87)' can't be established.
RSA key fingerprint is 57:cf:96:15:78:a3:94:93:30:16:8e:66:47:cd:f9:cd.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added 'datanode2,192.168.57.87' (RSA) to the list of known hosts.
root@datanode2's password:
profile 100% 1624 1.6KB/s 00:00
[root@hadoop1 src]# scp /etc/profile root@datanode3:/etc/
The authenticity of host 'datanode3 (192.168.57.88)' can't be established.
RSA key fingerprint is 31:73:e8:3c:20:0c:1e:b2:59:5c:d1:01:4b:26:41:70.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added 'datanode3,192.168.57.88' (RSA) to the list of known hosts.
root@datanode3's password:
profile 100% 1624 1.6KB/s 00:00
6 拷贝 jdk 安装包，并在每个datanode 节点安装 jdk 包
[root@hadoop1 ~]# scp -r /home/hadoop/src/ hadoop@datanode1:/home/hadoop/
hadoop@datanode1's password:
hadoop-0.20.203.0rc1.tar.gz 100% 58MB 57.8MB/s 00:01
jdk-7-linux-x64.rpm 100% 78MB 77.9MB/s 00:01
[root@hadoop1 ~]# scp -r /home/hadoop/src/ hadoop@datanode2:/home/hadoop/
hadoop@datanode2's password:
hadoop-0.20.203.0rc1.tar.gz 100% 58MB 57.8MB/s 00:01
jdk-7-linux-x64.rpm 100% 78MB 77.9MB/s 00:01
[root@hadoop1 ~]# scp -r /home/hadoop/src/ hadoop@datanode3:/home/hadoop/
hadoop@datanode3's password:
hadoop-0.20.203.0rc1.tar.gz 100% 58MB 57.8MB/s 00:01
jdk-7-linux-x64.rpm 100% 78MB 77.9MB/s 00:01
四 hadoop 配置 //注意使用hadoop 用户 操作
1 配置目录
[hadoop@hadoop1 ~]$ pwd
/home/hadoop
[hadoop@hadoop1 ~]$ ll
total 59220
lrwxrwxrwx 1 hadoop hadoop 17 Feb 1 16:59 hadoop -> hadoop-0.20.203.0
drwxr-xr-x 12 hadoop hadoop 4096 Feb 1 17:31 hadoop-0.20.203.0
-rw-r--r-- 1 hadoop hadoop 60569605 Feb 1 14:24 hadoop-0.20.203.0rc1.tar.gz
2 配置hadoop-env.sh，指定java位置
vim hadoop/conf/hadoop-env.sh
export JAVA_HOME=/usr/java/jdk1.7.0
3 配置core-site.xml //定位文件系统的 namenode
[hadoop@hadoop1 ~]$ cat hadoop/conf/core-site.xml
<?xml version="1.0"?>
<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>
<configuration>
<property>
<name>fs.default.name</name>
<value>hdfs://namenode:9000</value>
</property>
</configuration>
4 配置mapred-site.xml //定位jobtracker 所在的主节点
[hadoop@hadoop1 ~]$ cat hadoop/conf/mapred-site.xml
<?xml version="1.0"?>
<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>
<configuration>
<property>
<name>mapred.job.tracker</name>
<value>namenode:9001</value>
</property>
</configuration>
5 配置hdfs-site.xml //配置HDFS副本数量
[hadoop@hadoop1 ~]$ cat hadoop/conf/hdfs-site.xml
<?xml version="1.0"?>
<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>
<configuration>
<property>
<name>dfs.replication</name>
<value>3</value>
</property>
</configuration>
6 配置 master 与 slave 配置文档
[hadoop@hadoop1 ~]$ cat hadoop/conf/masters
namenode
[hadoop@hadoop1 ~]$ cat hadoop/conf/slaves
datanode1
datanode2
7 拷贝hadoop 目录到所有节点(datanode)
[hadoop@hadoop1 ~]$ scp -r hadoop hadoop@datanode1:/home/hadoop/
[hadoop@hadoop1 ~]$ scp -r hadoop hadoop@datanode2:/home/hadoop/
[hadoop@hadoop1 ~]$ scp -r hadoop hadoop@datanode3:/home/hadoop
8 格式化 HDFS
[hadoop@hadoop1 hadoop]$ bin/hadoop namenode -format
12/02/02 11:31:15 INFO namenode.NameNode: STARTUP_MSG:
/************************************************************
STARTUP_MSG: Starting NameNode
STARTUP_MSG: host = hadoop1.test.com/127.0.0.1
STARTUP_MSG: args = [-format]
STARTUP_MSG: version = 0.20.203.0
STARTUP_MSG: build = http://svn.apache.org/repos/asf/hadoop/common/branches/branch-0.20-security-203 -r 1099333; compiled by 'oom' on Wed May 4 07:57:50 PDT 2011
************************************************************/
Re-format filesystem in /tmp/hadoop-hadoop/dfs/name ? (Y or N) Y //这里输入Y
12/02/02 11:31:17 INFO util.GSet: VM type = 64-bit
12/02/02 11:31:17 INFO util.GSet: 2% max memory = 19.33375 MB
12/02/02 11:31:17 INFO util.GSet: capacity = 2^21 = 2097152 entries
12/02/02 11:31:17 INFO util.GSet: recommended=2097152, actual=2097152
12/02/02 11:31:17 INFO namenode.FSNamesystem: fsOwner=hadoop
12/02/02 11:31:18 INFO namenode.FSNamesystem: supergroupsupergroup=supergroup
12/02/02 11:31:18 INFO namenode.FSNamesystem: isPermissionEnabled=true
12/02/02 11:31:18 INFO namenode.FSNamesystem: dfs.block.invalidate.limit=100
12/02/02 11:31:18 INFO namenode.FSNamesystem: isAccessTokenEnabled=false accessKeyUpdateInterval=0 min(s), accessTokenLifetime=0 min(s)
12/02/02 11:31:18 INFO namenode.NameNode: Caching file names occuring more than 10 times
12/02/02 11:31:18 INFO common.Storage: Image file of size 112 saved in 0 seconds.
12/02/02 11:31:18 INFO common.Storage: Storage directory /tmp/hadoop-hadoop/dfs/name has been successfully formatted.
12/02/02 11:31:18 INFO namenode.NameNode: SHUTDOWN_MSG:
/************************************************************
SHUTDOWN_MSG: Shutting down NameNode at hadoop1.test.com/127.0.0.1
************************************************************/
[hadoop@hadoop1 hadoop]$
9 启动hadoop 守护进程
[hadoop@hadoop1 hadoop]$ bin/start-all.sh
starting namenode, logging to /home/hadoop/hadoop/bin/../logs/hadoop-hadoop-namenode-hadoop1.test.com.out
datanode1: starting datanode, logging to /home/hadoop/hadoop/bin/../logs/hadoop-hadoop-datanode-hadoop2.test.com.out
datanode2: starting datanode, logging to /home/hadoop/hadoop/bin/../logs/hadoop-hadoop-datanode-hadoop3.test.com.out
datanode3: starting datanode, logging to /home/hadoop/hadoop/bin/../logs/hadoop-hadoop-datanode-hadoop4.test.com.out
starting jobtracker, logging to /home/hadoop/hadoop/bin/../logs/hadoop-hadoop-jobtracker-hadoop1.test.com.out
datanode1: starting tasktracker, logging to /home/hadoop/hadoop/bin/../logs/hadoop-hadoop-tasktracker-hadoop2.test.com.out
datanode2: starting tasktracker, logging to /home/hadoop/hadoop/bin/../logs/hadoop-hadoop-tasktracker-hadoop3.test.com.out
datanode3: starting tasktracker, logging to /home/hadoop/hadoop/bin/../logs/hadoop-hadoop-tasktracker-hadoop4.test.com.out
10 验证
//namenode
[hadoop@hadoop1 logs]$ jps
2883 JobTracker
3002 Jps
2769 NameNode
//datanode
[hadoop@hadoop2 ~]$ jps
2743 TaskTracker
2670 DataNode
2857 Jps
[hadoop@hadoop3 ~]$ jps
2742 TaskTracker
2856 Jps
2669 DataNode
[hadoop@hadoop4 ~]$ jps
2742 TaskTracker
2852 Jps
2659 DataNode
Hadoop 监控web页面
http://192.168.57.75:50070/dfshealth.jsp
五 简单验证HDFS
hadoop 的文件命令格式如下:
hadoop fs -cmd <args>
//建立目录
[hadoop@hadoop1 hadoop]$ bin/hadoop fs -mkdir /test-hadoop
//査看目录
[hadoop@hadoop1 hadoop]$ bin/hadoop fs -ls /
Found 2 items
drwxr-xr-x - hadoop supergroup 0 2012-02-02 13:32 /test-hadoop
drwxr-xr-x - hadoop supergroup 0 2012-02-02 11:32 /tmp
//査看目录包括子目录
[hadoop@hadoop1 hadoop]$ bin/hadoop fs -lsr /
drwxr-xr-x - hadoop supergroup 0 2012-02-02 13:32 /test-hadoop
drwxr-xr-x - hadoop supergroup 0 2012-02-02 11:32 /tmp
drwxr-xr-x - hadoop supergroup 0 2012-02-02 11:32 /tmp/hadoop-hadoop
drwxr-xr-x - hadoop supergroup 0 2012-02-02 11:32 /tmp/hadoop-hadoop/mapred
drwx------ - hadoop supergroup 0 2012-02-02 11:32 /tmp/hadoop-hadoop/mapred/system
-rw------- 2 hadoop supergroup 4 2012-02-02 11:32 /tmp/hadoop-hadoop/mapred/system/jobtracker.info
//添加文件
[hadoop@hadoop1 hadoop]$ bin/hadoop fs -put /home/hadoop/hadoop-0.20.203.0rc1.tar.gz /test-hadoop
[hadoop@hadoop1 hadoop]$ bin/hadoop fs -lsr /
drwxr-xr-x - hadoop supergroup 0 2012-02-02 13:34 /test-hadoop
-rw-r--r-- 2 hadoop supergroup 60569605 2012-02-02 13:34 /test-hadoop/hadoop-0.20.203.0rc1.tar.gz
drwxr-xr-x - hadoop supergroup 0 2012-02-02 11:32 /tmp
drwxr-xr-x - hadoop supergroup 0 2012-02-02 11:32 /tmp/hadoop-hadoop
drwxr-xr-x - hadoop supergroup 0 2012-02-02 11:32 /tmp/hadoop-hadoop/mapred
drwx------ - hadoop supergroup 0 2012-02-02 11:32 /tmp/hadoop-hadoop/mapred/system
-rw------- 2 hadoop supergroup 4 2012-02-02 11:32 /tmp/hadoop-hadoop/mapred/system/jobtracker.info
//获取文件
[hadoop@hadoop1 hadoop]$ bin/hadoop fs -get /test-hadoop/hadoop-0.20.203.0rc1.tar.gz /tmp/
[hadoop@hadoop1 hadoop]$ ls /tmp/*.tar.gz
/tmp/1.tar.gz /tmp/hadoop-0.20.203.0rc1.tar.gz
//删除文件
[hadoop@hadoop1 hadoop]$ bin/hadoop fs -rm /test-hadoop/hadoop-0.20.203.0rc1.tar.gz
Deleted hdfs://namenode:9000/test-hadoop/hadoop-0.20.203.0rc1.tar.gz
[hadoop@hadoop1 hadoop]$ bin/hadoop fs -lsr /
drwxr-xr-x - hadoop supergroup 0 2012-02-02 13:57 /test-hadoop
drwxr-xr-x - hadoop supergroup 0 2012-02-02 11:32 /tmp
drwxr-xr-x - hadoop supergroup 0 2012-02-02 11:32 /tmp/hadoop-hadoop
drwxr-xr-x - hadoop supergroup 0 2012-02-02 11:32 /tmp/hadoop-hadoop/mapred
drwx------ - hadoop supergroup 0 2012-02-02 11:32 /tmp/hadoop-hadoop/mapred/system
-rw------- 2 hadoop supergroup 4 2012-02-02 11:32 /tmp/hadoop-hadoop/mapred/system/jobtracker.info
drwxr-xr-x - hadoop supergroup 0 2012-02-02 13:36 /user
-rw-r--r-- 2 hadoop supergroup 321 2012-02-02 13:36 /user/hadoop
//删除目录
[hadoop@hadoop1 hadoop]$ bin/hadoop fs -rmr /test-hadoop
Deleted hdfs://namenode:9000/test-hadoop
[hadoop@hadoop1 hadoop]$ bin/hadoop fs -lsr /
drwxr-xr-x - hadoop supergroup 0 2012-02-02 11:32 /tmp
drwxr-xr-x - hadoop supergroup 0 2012-02-02 11:32 /tmp/hadoop-hadoop
drwxr-xr-x - hadoop supergroup 0 2012-02-02 11:32 /tmp/hadoop-hadoop/mapred
drwx------ - hadoop supergroup 0 2012-02-02 11:32 /tmp/hadoop-hadoop/mapred/system
-rw------- 2 hadoop supergroup 4 2012-02-02 11:32 /tmp/hadoop-hadoop/mapred/system/jobtracker.info
drwxr-xr-x - hadoop supergroup 0 2012-02-02 13:36 /user
-rw-r--r-- 2 hadoop supergroup 321 2012-02-02 13:36 /user/hadoop
//hadoop fs 帮助（部分）
[hadoop@hadoop1 hadoop]$ bin/hadoop fs -help
hadoop fs is the command to execute fs commands. The full syntax is:
hadoop fs [-fs <local | file system URI>] [-conf <configuration file>]
[-D <propertyproperty=value>] [-ls <path>] [-lsr <path>] [-du <path>]
[-dus <path>] [-mv <src> <dst>] [-cp <src> <dst>] [-rm [-skipTrash] <src>]
[-rmr [-skipTrash] <src>] [-put <localsrc> ... <dst>] [-copyFromLocal <localsrc> ... <dst>]
[-moveFromLocal <localsrc> ... <dst>] [-get [-ignoreCrc] [-crc] <src> <localdst>
[-getmerge <src> <localdst> [addnl]] [-cat <src>]
[-copyToLocal [-ignoreCrc] [-crc] <src> <localdst>] [-moveToLocal <src> <localdst>]
[-mkdir <path>] [-report] [-setrep [-R] [-w] <rep> <path/file>]
[-touchz <path>] [-test -[ezd] <path>] [-stat [format] <path>]
[-tail [-f] <path>] [-text <path>]
[-chmod [-R] <MODE[,MODE]... | OCTALMODE> PATH...]
[-chown [-R] [OWNER][:[GROUP]] PATH...]
[-chgrp [-R] GROUP PATH...]
[-count[-q] <path>]
[-help [cmd]]
结束Hadoop 环境搭建步骤繁琐，需要具备一定的Linux 系统知识，需要注意的是，通过以上步骤搭建的Hadoop 环境只能让你大体了解的hadoop ，如果想将HDFS 用于线上服务，还需对hadoop 配置文档做进一步配置 ，后续文档将继续以博文的形式发布，敬请期待。
