Hadoop -分布式环境搭建安装配置

jopen 11年前

集群环境：

1 NameNode(真实主机):

Linux yan-Server 3.4.36-gentoo #3 SMP Mon Apr 1 14:09:12 CST 2013 x86_64 AMD Athlon(tm) X4 750K Quad Core Processor AuthenticAMD GNU/Linux

2 DataNode1(虚拟机):

Linux node1 3.5.0-23-generic #35~precise1-Ubuntu SMP Fri Jan 25 17:13:26 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux

3 DataNode2(虚拟机):

Linux node2 3.5.0-23-generic #35~precise1-Ubuntu SMP Fri Jan 25 17:13:26 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux

4 DataNode3(虚拟机):

Linux node3 3.5.0-23-generic #35~precise1-Ubuntu SMP Fri Jan 25 17:13:26 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux


1.安装VirtualBox虚拟机

Gentoo下直接命令编译安装，或者官网下载二进制安装包直接安装：

emerge -av virtualbox

2.虚拟机下安装Ubuntu 12.04 LTS

使用Ubuntu镜像安装完成后，然后再克隆另外两台虚拟主机（这里会遇到克隆的主机启动的时候主机名和MAC地址会是一样的，局域网会造成冲突）

主机名修改文件

/etc/hostname

MAC地址修改需要先删除文件

/etc/udev/rules.d/70-persistent-net.rules

然后在启动之前设置VirtualBox虚拟机的MAC地址

f1.png

启动后会自动生成删除的文件，配置网卡的MAC地址。

为了更方便的在各主机之间共享文件，可以启动主机yan-Server的NFS，将命令加入/etc/rc.local中，让客户端自动挂载NFS目录。

删除各虚拟机的NetworkManager，手动设置静态的IP地址，例如node2主机的/etc/network/interfaces文件配置如下：

auto lo  iface lo inet loopback  auto eth0  iface eth0 inet static  address 192.168.137.202  gateway 192.168.137.1  netmask 255.255.255.0  network 192.168.137.0  broadcast 192.168.137.255

主机的基本环境设置完毕，下面是主机对应的IP地址

类型 主机名 IP
NameNode yan-Server 192.168.137.100
DataNode node1 192.168.137.201
DataNode node2 192.168.137.202
DataNode node3 192.168.137.203
为了节省资源，可以设置虚拟机默认启动字符界面，然后通过主机的TERMINAL ssh远程登录。（SSH已经启动服务，允许远程登录，安装方法不再赘述）

设置方式是修改/etc/default/grub文件将下面的一行解除注释

GRUB_TERMINAL=console

然后update-grub即可。

3.Hadoop环境的配置


3.1配置JDK环境（之前就做好了，这里不再赘述）

export JAVA_HOME=/opt/jdk1.7.0_21  export PATH=$JAVA_HOME/bin:$PATH  export CLASSPATH=.:$JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar

3.2在官网下载Hadoop，然后解压到/opt/目录下面（这里使用的是hadoop-2.0.4-alpha）

然后进入目录/opt/hadoop-2.0.4-alpha/etc/hadoop，配置hadoop文件

修改文件hadoop-env.sh

export HADOOP_FREFIX=/opt/hadoop-2.0.4-alpha  export HADOOP_COMMON_HOME=${HADOOP_FREFIX}  export HADOOP_HDFS_HOME=${HADOOP_FREFIX}  export PATH=$PATH:$HADOOP_FREFIX/bin  export PATH=$PATH:$HADOOP_FREFIX/sbin  export HADOOP_MAPRED_HOME=${HADOOP_FREFIX}  export YARN_HOME=${HADOOP_FREFIX}  export HADOOP_CONF_HOME=${HADOOP_FREFIX}/etc/hadoop  export YARN_CONF_DIR=${HADOOP_FREFIX}/etc/hadoop  export JAVA_HOME=/opt/jdk1.7.0_21

修改文件hdfs-site.xml

<configuration>   <property>    <name>dfs.namenode.name.dir</name>    <value>file:/opt/hadoop-2.0.4-alpha/workspace/name</value>    <description>Determines where on the local filesystem the DFS name node should store the name table.If this is a comma-delimited list of directories,then name table is replicated in all of the directories,for redundancy.</description>    <final>true</final>   </property>   <property>    <name>dfs.datanode.data.dir</name>    <value>file:/opt/hadoop-2.0.4-alpha/workspace/data</value>    <description>Determines where on the local filesystem an DFS data node should store its blocks.If this is a comma-delimited list of directories,then data will be stored in all named directories,typically on different devices.Directories that do not exist are ignored.    </description>    <final>true</final>   </property>   <property>    <name>dfs.replication</name>    <value>1</value>   </property>   <property>    <name>dfs.permission</name>    <value>false</value>   </property>  </configuration>


修改文件mapred-site.xml

<configuration>   <property>    <name>mapreduce.framework.name</name>    <value>yarn</value>   </property>   <property>    <name>mapreduce.job.tracker</name>    <value>hdfs://yan-Server:9001</value>    <final>true</final>   </property>   <property>    <name>mapreduce.map.memory.mb</name>    <value>1536</value>   </property>   <property>    <name>mapreduce.map.java.opts</name>    <value>-Xmx1024M</value>   </property>   <property>    <name>mapreduce.reduce.memory.mb</name>    <value>3072</value>   </property>   <property>    <name>mapreduce.reduce.java.opts</name>    <value>-Xmx2560M</value>   </property>   <property>    <name>mapreduce.task.io.sort.mb</name>    <value>512</value>   </property>   <property>    <name>mapreduce.task.io.sort.factor</name>    <value>100</value>   </property>   <property>    <name>mapreduce.reduce.shuffle.parallelcopies</name>    <value>50</value>   </property>   <property>    <name>mapred.system.dir</name>    <value>file:/opt/hadoop-2.0.4-alpha/workspace/systemdir</value>    <final>true</final>   </property>   <property>    <name>mapred.local.dir</name>    <value>file:/opt/hadoop-2.0.4-alpha/workspace/localdir</value>    <final>true</final>   </property>  </configuration>


修改文件yarn-env.xml

export HADOOP_FREFIX=/opt/hadoop-2.0.4-alpha  export HADOOP_COMMON_HOME=${HADOOP_FREFIX}  export HADOOP_HDFS_HOME=${HADOOP_FREFIX}  export PATH=$PATH:$HADOOP_FREFIX/bin  export PATH=$PATH:$HADOOP_FREFIX/sbin  export HADOOP_MAPRED_HOME=${HADOOP_FREFIX}  export YARN_HOME=${HADOOP_FREFIX}  export HADOOP_CONF_HOME=${HADOOP_FREFIX}/etc/hadoop  export YARN_CONF_DIR=${HADOOP_FREFIX}/etc/hadoop  export JAVA_HOME=/opt/jdk1.7.0_21

修改文件yarn-site.xml

<configuration>   <property>    <name>yarn.resourcemanager.address</name>    <value>yan-Server:8080</value>   </property>   <property>    <name>yarn.resourcemanager.scheduler.address</name>    <value>yan-Server:8081</value>   </property>   <property>    <name>yarn.resourcemanager.resource-tracker.address</name>    <value>yan-Server:8082</value>   </property>   <property>    <name>yarn.nodemanager.aux-services</name>    <value>mapreduce.shuffle</value>   </property>   <property>    <name>yarn.nodemanager.aux-services.mapreduce.shuffle.class</name>    <value>org.apache.hadoop.mapred.ShuffleHandler</value>   </property>  </configuration>

将配置好的Hadoop复制到各DataNode（这里DataNode的JDK配置和主机的配置是一致的，不需要再修改JDK的配置）


3.3 修改主机的/etc/hosts，将NameNode加入该文件

192.168.137.100 yan-Server    192.168.137.201 node1  192.168.137.202 node2  192.168.137.203 node3

3.4 修改各DataNode的/etc/hosts文件，也添加上述的内容

192.168.137.100 yan-Server  192.168.137.201 node1  192.168.137.202 node2  192.168.137.203 node3

3.5 配置SSH免密码登录（所有的主机都使用root用户登录）

主机上运行命令

ssh-kengen -t rsa
一路回车，然后复制.ssh/id_rsa.pub为各DataNode的root用户目录.ssh/authorized_keys文件

然后在主机上远程登录一次

ssh root@node1
首次登录可能会需要输入密码，之后就不再需要。（其他的DataNode也都远程登录一次确保可以免输入密码登录）

4.启动Hadoop

为了方便，在主机的/etc/profile配置hadoop的环境变量，如下：

export HADOOP_PREFIX="/opt/hadoop-2.0.4-alpha"  export PATH=$PATH:$HADOOP_PREFIX/bin  export PATH=$PATH:$HADOOP_PREFIX/sbin  export HADOOP_MAPRED_HOME=${HADOOP_PREFIX}  export HADOOP_COMMON_HOME=${HADOOP_PREFIX}  export HADOOP_HDFS_HOME=${HADOOP_PREFIX}  export YARN_HOME=${HADOOP_PREFIX}

4.1 格式化NameNode

hdfs namenode -format

4.2 启动全部进程

start-all.sh
f2.png
在浏览器查看，地址：

http://localhost:8088/

f3.png

所有数据节点DataNode正常启动。

4.3 关闭所有进程

stop-all.sh
f4.png

至此，Hadoop环境搭建基本结束。
上面使用的是alpha版本，不是稳定版，稳定版的配置文件会有所不同，如果照搬会可能导致jobtracker或tasktracker无法启动的问题。遇到问题查看日志文件。

下面是我配置的1.1.2稳定版本的配置文件：

conf/core-site.xml

<?xml version="1.0"?>  <?xml-stylesheet type="text/xsl" href="configuration.xsl"?>    <!-- Put site-specific property overrides in this file. -->    <configuration>   <property>       <name>fs.default.name</name>    <value>hdfs://yan-Server:49000</value>   </property>   <property>    <name>hadoop.tmp.dir</name>    <value>/tmp</value>    </property>  </configuration>

conf/hadoop-env.sh

export HADOOP_FREFIX=/opt/hadoop-1.1.2  export HADOOP_COMMON_HOME=${HADOOP_FREFIX}  export HADOOP_HDFS_HOME=${HADOOP_FREFIX}  export PATH=$PATH:$HADOOP_FREFIX/bin  export PATH=$PATH:$HADOOP_FREFIX/sbin  export HADOOP_MAPRED_HOME=${HADOOP_FREFIX}  export YARN_HOME=${HADOOP_FREFIX}  export HADOOP_CONF_HOME=${HADOOP_FREFIX}/conf  export YARN_CONF_DIR=${HADOOP_FREFIX}/conf  export JAVA_HOME=/opt/jdk1.7.0_21

conf/hdfs-site.xml

<?xml version="1.0"?>  <?xml-stylesheet type="text/xsl" href="configuration.xsl"?>    <!-- Put site-specific property overrides in this file. -->    <configuration>   <property>    <name>dfs.name.dir</name>    <value>/opt/hadoop-1.1.2/workspace/name</value>    <description>Determines where on the local filesystem the DFS name node should store the name table.If this is a comma-delimited list of directories,then name table is replicated in all of the directories,for redundancy.</description>    <final>true</final>   </property>   <property>    <name>dfs.data.dir</name>    <value>/opt/hadoop-1.1.2/workspace/data</value>    <description>Determines where on the local filesystem an DFS data node should store its blocks.If this is a comma-delimited list of directories,then data will be stored in all named directories,typically on different devices.Directories that do not exist are ignored.    </description>    <final>true</final>   </property>   <property>    <name>dfs.replication</name>    <value>1</value>   </property>   <property>    <name>dfs.permission</name>    <value>false</value>   </property>  </configuration>

 

conf/mapred-site.xml

<?xml version="1.0"?>  <?xml-stylesheet type="text/xsl" href="configuration.xsl"?>      <!-- Put site-specific property overrides in this file. -->    <configuration>   <property>    <name>mapred.job.tracker</name>    <value>hdfs://yan-Server:9001</value>    <final>true</final>   </property>  </configuration>


conf/masters中填写NameNode主机名

conf/slaves中填写DataNode主机名

 

参考：

1.实战Hadoop--开启通向云计算的捷径（刘鹏）

2.http://www.cnblogs.com/aniuer/archive/2012/07/16/2594448.html

来自：http://blog.csdn.net/yming0221/article/details/8989203