五节点的Hadoop HA集群搭建

五节点的Hadoop HA集群搭建

1 集群规划

1.1 节点（虚拟机）

主机名	IP	操作系统	安装软件
vm1	10.211.55.11	Centos 7	JDK，Hadoop
vm2	10.211.55.12	Centos 7	JDK，Hadoop
vm3	10.211.55.13	Centos 7	JDK，Hadoop，Zookeeper
vm4	10.211.55.14	Centos 7	JDK，Hadoop，Zookeeper
vm5	10.211.55.15	Centos 7	JDK，Hadoop，Zookeeper

1.2 服务

	vm1	vm2	vm3	vm4	vm5
NameNode	✓	✓
DataNode			✓	✓	✓
ResourceManager	✓	✓
NodeManager			✓	✓	✓
Zookeeper			✓	✓	✓
JournalNode			✓	✓	✓
ZKFC	✓	✓

2 节点环境

root用户登录配置节点环境，以节点vm1为例，其他节点类似

2.1 修改主机名

hostnamectl # 查看主机名
hostnamectl set-hostname vm1 # 设置主机名
vim /etc/hostname # 设置主机名

2.2 修改IP地址

vim /etc/sysconfig/network-scripts/ifcfg-eth0
# 修改或者添加以下几项
# BOOTPROTO="static"
# ONBOOT="yes"
# IPADDR=10.211.55.11
# NETMASK=255.255.255.0
# GATEWAY=10.211.55.1

vim /etc/resolv.conf
# 添加以下内容
# nameserver 114.114.114.114
# nameserver 114.114.114.115

# 重启network服务
systemctl restart network.service

2.3 添加主机名与IP映射

vim /etc/hosts
# 添加如下内容
# 10.211.55.11 vm1
# 10.211.55.12 vm2
# 10.211.55.13 vm3
# 10.211.55.14 vm4
# 10.211.55.15 vm5

2.4 添加hadoop用户

useradd -m hadoop -s /bin/bash # 创建hadoop用户
passwd hadoop # 修改hadoop用户密码
visudo # 为hadoop用户添加sudo权限
# root   ALL=(ALL) ALL 下面添加一下内容
# hadoop ALL=(ALL) ALL

2.5 关闭防火墙

systemctl stop firewalld # 停止防火墙服务
systemctl disable firewalld # 取消开启启动防火墙服务

systemctl stop iptables.service # 停止防火墙服务
systemctl disable iptables.service # 取消开启启动防火墙服务

2.6 关闭Selinux

getenforce # 查看
vim /etc/selinux/config
# SELINUX=enforcing改为SELINUX=disabled
# 重启生效

2.7 安装 JDK

下载JDK安装包

jdk-8u251-linux-x64.rpm

# 本机命令
scp jdk-8u251-linux-x64.rpm root@10.211.55.11:/root #上传jdk安装包到vm1
scp jdk-8u251-linux-x64.rpm root@10.211.55.12:/root #上传jdk安装包到vm2
scp jdk-8u251-linux-x64.rpm root@10.211.55.13:/root #上传jdk安装包到vm3
scp jdk-8u251-linux-x64.rpm root@10.211.55.14:/root #上传jdk安装包到vm4
scp jdk-8u251-linux-x64.rpm root@10.211.55.15:/root #上传jdk安装包到vm5

# vm1-vm5命令
rpm -ivh jdk-8u251-linux-x64.rpm # 安装jdk
java -verbose # 查看JDK安装目录

vim /etc/profile # 配置JAVA_HOME环境变量
# 添加以下内容到文件末尾
# export JAVA_HOME=/usr/java/jdk1.8.0_251-amd64
# export PATH=$JAVA_HOME/bin:$PATH

source /etc/profile # 配置马上生效

echo $JAVA_HOME # 查看环境变

2.8 SSH免密登录

# vm1执行命令
su hadoop
ssh-keygen -t rsa # 生成公钥，一路回车
ssh-copy-id vm1
ssh-copy-id vm2
ssh-copy-id vm3
ssh-copy-id vm4
ssh-copy-id vm5

#测试免密登录
ssh vm1
ssh vm2
ssh vm3
ssh vm4
ssh vm5

2.9 NTP时间同步

# vm1-vm5命令
yum install -y ntp
systemctl start ntpd.service
systemctl enable ntpd.service

# vm1命令
vim /etc/ntp.conf
# server 0.centos.pool.ntp.org iburst
# server 1.centos.pool.ntp.org iburst
# server 2.centos.pool.ntp.org iburst
# server 3.centos.pool.ntp.org iburst
# 修改上面内容为下面内容
# server 127.127.1.0 # local clock
# fudge 127.127.1.0 stratum 10
systemctl restart ntpd.service # 重启ntp服务
ntpstat # 查看ntp状态

# vm2-vm5命令
vim /etc/ntp.conf
# server 0.centos.pool.ntp.org iburst
# server 1.centos.pool.ntp.org iburst
# server 2.centos.pool.ntp.org iburst
# server 3.centos.pool.ntp.org iburst
# 修改上面内容为下面内容
# server vm1

systemctl restart ntpd.service # 重启ntp服务
ntpdate -u vm1 # 同步时间
ntpstat # 查看ntp状态

2.9 准备目录

# vm1-vm5
mkdir /bdp
mkdir /bdp/data
mkdir /bdp/tmp

2.10 准备hadoop和zookeeper

下载hadoop

hadoop-2.10.0.tar.gz

下载zookeeper

apache-zookeeper-3.6.1-bin.tar.gz

# 本机命令
scp hadoop-2.10.0.tar.gz hadoop@10.211.55.11:/home/hadoop #上传hadoop安装包到vm1
scp hadoop-2.10.0.tar.gz hadoop@10.211.55.12:/home/hadoop #上传hadoop安装包到vm2
scp hadoop-2.10.0.tar.gz hadoop@10.211.55.13:/home/hadoop #上传hadoop安装包到vm3
scp hadoop-2.10.0.tar.gz hadoop@10.211.55.14:/home/hadoop #上传hadoop安装包到vm4
scp hadoop-2.10.0.tar.gz hadoop@10.211.55.15:/home/hadoop #上传hadoop安装包到vm5
scp apache-zookeeper-3.6.1-bin.tar.gz hadoop@10.211.55.11:/home/hadoop #上传zookeeper安装包到vm1
scp apache-zookeeper-3.6.1-bin.tar.gz hadoop@10.211.55.12:/home/hadoop #上传zookeeper安装包到vm2
scp apache-zookeeper-3.6.1-bin.tar.gz hadoop@10.211.55.13:/home/hadoop #上传zookeeper安装包到vm3
scp apache-zookeeper-3.6.1-bin.tar.gz hadoop@10.211.55.14:/home/hadoop #上传zookeeper安装包到vm4
scp apache-zookeeper-3.6.1-bin.tar.gz hadoop@10.211.55.15:/home/hadoop #上传zookeeper安装包到vm5

#vm1-vm5命令
mkdir /bdp
chown -R hadoop:hadoop /bdp
su hadoop
cd /home/hadoop
tar zxvf hadoop-2.10.0.tar.gz -C /bdp
tar zxvf apache-zookeeper-3.6.1-bin.tar.gz /bdp
cd /bdp
mv apache-zookeeper-3.6.1-bin zookeeper-3.6.1

vim ~/.bashrc # 配置环境变量
# 在文件末尾添加一下内容
# export HADOOP_HOME=/bdp/hadoop-2.10.0
# export HADOOP_USER_NAME=hadoop
# export HADOOP_CONF_DIR=$HADOOP_HOME/etc/hadoop
# export ZK_HOME=/bdp/zookeeper-3.6.1
# export PATH=$HADOOP_HOME/bin:$ZK_HOME/bin:$PATH
source ~/.bashrc # 配置立即生效

3 Zookeeper集群搭建

3.1 修改配置

# vm3-vm5命令
cd $ZK_HOME/conf
cp zoo_sample.cfg zoo.cfg
vim zoo.cfg
# 修改以下内容
dataDir=/bdp/data/zookeeper
# server.0=vm3:2888:3888
# server.1=vm4:2888:3888
# server.2=vm5:2888:3888

3.2 创建myid文件

# vm3-vm5命令
mkdir -p /bdp/data/zookeeper

# vm3命令
echo 0 > /bdp/data/zookeeper/myid
# vm4命令
echo 1 > /bdp/data/zookeeper/myid
# vm5命令
echo 2 > /bdp/data/zookeeper/myid

3.3 启动Zookeeper

# vm3-vm5命令
zkServer.sh start # 启动
zkServer.sh status # 查看状态

4 Hadoop集群搭建

ssh hadoop@vm1

4.1 hadoop-env.sh

vim $HADOOP_HOME/etc/hadoop/hadoop-env.sh
# 添加如下内容
# export JAVA_HOME=/usr/java/jdk1.8.0_251-amd64

4.2 core-site.xml

vim $HADOOP_HOME/etc/hadoop/core-site.xml #配置core-site.xml

core-site.xml

<configuration>
  <!-- 指定hdfs的nameservice为hdp -->
  <property>
    <name</