centos7下hadoop-3.1.0伪集群搭建

centos7下hadoop-3.1.0集群搭建

环境准备

1.服务器概览

hostname	ip	说明
node1.spark	192.168.2.140	node1节点(master)
node2.spark	192.168.2.141	node2节点
node3.spark	192.168.2.142	node3节点

分别在三台服务器上执行以下命令

#添加host
[root@node1 ~] vim /etc/hosts
192.168.2.140 node1.spark
192.168.2.141 node2.spark
192.168.2.142 node3.spark


#执行以下命令关闭防火墙
[root@node1 ~]systemctl stop firewalld && systemctl disable firewalld
[root@node1 ~]setenforce 0

#将SELINUX的值改成disabled
[root@node1 ~]vim /etc/selinux/config

SELINUX=disabled

#重启服务器
[root@node1 ~]reboot

2.配置免密码登录

#node1执行以下命令

#生成密钥Pair,输入之后一直选择enter即可。生成的秘钥位于 ~/.ssh文件夹下
[root@node1 ~]# ssh-keygen -t rsa 

[root@node1 .ssh]# scp /root/.ssh/id_rsa.pub root@spark.node2:~
[root@node1 .ssh]# scp /root/.ssh/id_rsa.pub root@spark.node3:~

##node2,node3 执行以下命令
[root@node2 ~]# mkdir -p ~/.ssh
[root@node2 ~]# cd .ssh/
[root@node2 .ssh]# cat ~/id_rsa.pub >> authorized_keys
[root@node2 .ssh]# vim /etc/ssh/sshd_config
#禁用root账户登录，如果是用root用户登录请开启
PermitRootLogin yes

3.JDK安装

#配置环境变量
[root@node1 ~]# vim /etc/profile

# 在最后下添加
# Java Environment Path
export JAVA_HOME=/opt/java/jdk1.8.0_172
export PATH=$JAVA_HOME/bin:$PATH
export CLASSPATH=.:$JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar

# 刷新配置文件
source /etc/profile

安装Hadoop

1.安装Hadoop

[root@node1 opt]# cd /opt/ & mkdir hadoop && cd hadoop
#解压hadoop-3.1.0.tar.gz
[root@node1 hadoop]# tar -zxvf hadoop-3.1.0.tar.gz

#修改环境变量
[root@node1 hadoop]# vim /etc/profile

# 在最后下添加
export HADOOP_HOME=/opt/hadoop/hadoop-3.1.0
export PATH=$PATH:$HADOOP_HOME/bin


# 刷新配置文件
[root@node1 hadoop]# source /etc/profile

2.修改配置文件

这些配置文件全部位于 /opt/hadoop/hadoop-3.1.0/etc/hadoop 文件夹下

hadoop-env.sh

#The java implementation to use. By default, this environment

# variable is REQUIRED on ALL platforms except OS X!
# export JAVA_HOME=
export JAVA_HOME=/opt/java/jdk1.8.0_172

core-site.xml

<configuration>
    <!-- 指定HDFS老大（namenode）的通信地址 -->
    <property>
        <name>fs.defaultFS</name>
        <value>hdfs://node1.spark:9000</value>
    </property>
    <!-- 指定hadoop运行时产生文件的存储路径 -->
    <property>
        <name>hadoop.tmp.dir</name>
        <value>/opt/hadoop/data/tmp</value>
    </property>
</configuration>

hdfs-site.xml

<configuration>

    <!-- 设置namenode的http通讯地址 -->
    <property>
        <name>dfs.namenode.http-address</name>
        <value>node1.spark:50070</value>
    </property>

    <!-- 设置secondarynamenode的http通讯地址 -->
    <property>
        <name>dfs.namenode.secondary.http-address</name>
        <value>node2.spark:50090</value>
    </property>

    <!-- 设置namenode存放的路径 -->
    <property>
        <name>dfs.namenode.name.dir</name>
        <value>/opt/hadoop/data/name</value>
    </property>

    <!-- 设置hdfs副本数量 -->
    <property>
        <name>dfs.replication</name>
        <value>2</value>
    </property>
    <!-- 设置datanode存放的路径 -->
    <property>
        <name>dfs.datanode.data.dir</name>
        <value>/opt/hadoop/data/datanode</value>
    </property>
    
    <property>
        <name>dfs.permissions</name>
        <value>false</value>
    </property>
</configuration>

mapred-site.xml

<configuration>
    <!-- 通知框架MR使用YARN -->
    <property>
        <name>mapreduce.framework.name</name>
        <value>yarn</value>
    </property>
    <property>  
        <name>mapreduce.application.classpath</name>  
        <value>  
        /opt/hadoop/hadoop-3.1.0/etc/hadoop,  
        /opt/hadoop/hadoop-3.1.0/share/hadoop/common/*,  
        /opt/hadoop/hadoop-3.1.0/share/hadoop/common/lib/*,  
        /opt/hadoop/hadoop-3.1.0/share/hadoop/hdfs/*,  
        /opt/hadoop/hadoop-3.1.0/share/hadoop/hdfs/lib/*,  
        /opt/hadoop/hadoop-3.1.0/share/hadoop/mapreduce/*,  
        /opt/hadoop/hadoop-3.1.0/share/hadoop/mapreduce/lib/*,  
        /opt/hadoop/hadoop-3.1.0/share/hadoop/yarn/*,  
        /opt/hadoop/hadoop-3.1.0/share/hadoop/yarn/lib/*  
        </value>  
    </property>
</configuration>

yarn-site.xml

<configuration>
    <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.ShuffleHandle</value>  
    </property>  
    <property>  
        <name>yarn.resourcemanager.resource-tracker.address</name>  
        <value>node1.spark:8025</value>  
    </property>  
    <property>  
        <name>yarn.resourcemanager.scheduler.address</name>  
        <value>node1.spark:8030</value>  
    </property>  
    <property>  
        <name>yarn.resourcemanager.address</name>  
        <value>node1.spark:8040</value>  
    </property>  
</configuration>

sbin/start-dfs.sh sbin/stop-dfs.sh

HDFS_DATANODE_USER=root 
HADOOP_SECURE_DN_USER=hdfs 
HDFS_NAMENODE_USER=root 
HDFS_SECONDARYNAMENODE_USER=root 

sbin/start-yarn.sh sbin/stop-yarn.sh

YARN_RESOURCEMANAGER_USER=root
HADOOP_SECURE_DN_USER=yarn 
YARN_NODEMANAGER_USER=root

masters

新建一个masters的文件,这里指定的是secondary namenode 的主机

[root@node1 hadoop]# touch /opt/hadoop/hadoop-3.1.0/etc/hadoop/masters
[root@node1 hadoop]# vim /opt/hadoop/hadoop-3.1.0/etc/hadoop/masters
#添加
node2.spark

workers

[root@node1 hadoop]# touch /opt/hadoop/hadoop-3.1.0/etc/hadoop/slaves
[root@node1 hadoop]# vim /opt/hadoop/hadoop-3.1.0/etc/hadoop/slaves
#添加
node2.spark
node3.spark

创建文件夹

[root@node1 hadoop]# mkdir -p /opt/hadoop/data/tmp
[root@node1 hadoop]# mkdir -p /opt/hadoop/data/name
[root@node1 hadoop]# mkdir -p /opt/hadoop/data/datanode

复制到其他主机

[root@node1 opt]# scp -r /opt/hadoop spark.node2:/opt/
[root@node1 opt]# scp -r /opt/hadoop spark.node3:/opt/

修改spark.node2 spark.node3 环境变量

#修改环境变量
[root@node2 hadoop]# vim /etc/profile

# 在最后下添加
export HADOOP_HOME=/opt/hadoop/hadoop-3.1.0
export PATH=$PATH:$HADOOP_HOME/bin


# 刷新配置文件
[root@node2 hadoop]# source /etc/profile

3.启动 第一次启动得格式化

[root@node1 opt]#  /opt/hadoop/hadoop-3.1.0/bin/hdfs namenode -format

启动

[root@node1 opt]#  /opt/hadoop/hadoop-3.1.0/sbin/start-all.sh

来源：oschina

链接：https://my.oschina.net/u/1028790/blog/1816023