1. 编写集群分发脚本xsync
- 在/home/yanlzh目录下创建bin目录,并在bin目录下xsync创建文件,文件内容如下:
[yanlzh@node11 ~]$ mkdir bin
[yanlzh@node11 ~]$ cd bin/
[yanlzh@node11 bin]$ touch xsync
[yanlzh@node11 bin]$ vi xsync
- 在该文件中编写如下代码
#!/bin/bash
#1 获取输入参数个数,如果没有参数,直接退出
pcount=$#
if((pcount==0)); then
echo no args;
exit;
fi
#2 获取文件名称
p1=$1
fname=`basename $p1`
echo fname=$fname
#3 获取上级目录到绝对路径
pdir=`cd -P $(dirname $p1); pwd`
echo pdir=$pdir
#4 获取当前用户名称
user=`whoami`
#5 循环
for((host=11; host<14; host++)); do
echo ------------------- hadoop$host --------------
rsync -rvl $pdir/$fname $user@node$host:$pdir
done
- 修改脚本 xsync 具有执行权限
[yanlzh@node11 bin]$ chmod 777 xsync
2. SSH 免密登录配置
- 生成公钥私钥
[yanlzh@node11 .ssh]$ ssh-keygen -t rsa
- 将公钥私钥拷贝到要免密登录的目标机器上
[yanlzh@node11 .ssh]$ ssh-copy-id node11
[yanlzh@node11 .ssh]$ ssh-copy-id node12
[yanlzh@node11 .ssh]$ ssh-copy-id node13
- 还需要在node12上采用yanlzh账号配置一下无密登录到node11 、node12 、node13 服务器上。
[yanlzh@node12 .ssh]$ ssh-copy-id node11
[yanlzh@node12 .ssh]$ ssh-copy-id node12
[yanlzh@node12 .ssh]$ ssh-copy-id node13
- 还需要在node11 上采用root账号,配置一下无密登录到hadoop102、node12 、node13 ;
[root@node11 .ssh]$ ssh-copy-id node11
[root@node11 .ssh]$ ssh-copy-id node12
[root@node11 .ssh]$ ssh-copy-id node13
3. 集群配置
- 集群部署规划
node11 | node12 | node13 |
---|---|---|
NameNode | ||
DataNode | DataNode | DataNode |
SecondaryNameNode | ||
NodeManager | NodeManager | NodeManager |
ResourceManager |
- 配置集群
core-site.xml
<!-- 指定HDFS中NameNode的地址 -->
<property>
<name>fs.defaultFS</name>
<value>hdfs://node11:9000</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-2.7.2/data/tmp</value>
</property>
hadoop-env.sh
export JAVA_HOME=/opt/module/jdk1.8.0_144
hdfs-site.xml
<property>
<name>dfs.replication</name>
<value>3</value>
</property>
<!-- 指定Hadoop辅助名称节点主机配置 -->
<property>
<name>dfs.namenode.secondary.http-address</name>
<value>node13:50090</value>
</property>
yarn-env.sh
export JAVA_HOME=/opt/module/jdk1.8.0_144
yarn-site.xml
<!-- Reducer获取数据的方式 -->
<property>
<name>yarn.nodemanager.aux-services</name>
<value>mapreduce_shuffle</value>
</property>
<!-- 指定YARN的ResourceManager的地址 -->
<property>
<name>yarn.resourcemanager.hostname</name>
<value>node12</value>
</property>
mapred-env.sh
export JAVA_HOME=/opt/module/jdk1.8.0_144
mapred-site.xml
<!-- 指定MR运行在Yarn上 -->
<property>
<name>mapreduce.framework.name</name>
<value>yarn</value>
</property>
slaves
node11
node12
node13
- 在集群上分发配置好的Hadoop配置文件
[yanlzh@node11 hadoop]$ xsync /opt/module/hadoop-2.7.2/
- 查看文件分发情况
[yanlzh@node12 hadoop]$ cat /opt/module/hadoop-2.7.2/etc/hadoop/core-site.xml
4. 启动集群
- 如果集群是第一次启动,需要格式化NameNode(注意格式化之前,一定要先停止上次启动的所有namenode和datanode进程,然后再删除data和log数据)
[yanlzh@node11 hadoop-2.7.2]$ bin/hdfs namenode -format
- 启动HDFS
[yanlzh@node11 hadoop-2.7.2]$ sbin/start-dfs.sh
[yanlzh@node11 hadoop-2.7.2]$ jps
4166 NameNode
4482 Jps
4263 DataNode
[yanlzh@node12 hadoop-2.7.2]$ jps
3218 DataNode
3288 Jps
[yanlzh@node13 hadoop-2.7.2]$ jps
3221 DataNode
3283 SecondaryNameNode
3364 Jps
- 启动YARN
[yanlzh@node13 hadoop-2.7.2]$ sbin/start-yarn.sh
注意:NameNode和ResourceManger如果不是同一台机器,不能在NameNode上启动 YARN,应该在ResouceManager所在的机器上启动YARN。