一、网络及主机名配置
| ip地址 | 主机名 | 安装软件 | j进程 |
| 192.168.1.2 | master.hadoop | jdk,hadoop,zookeeper | namenode,datanode,ZKFC,ResourceManager,nodemanager |
| 192.168.1.3 | slave1.hadoop | jdk.hadoop,zookeeper | namenode,datanode,ZKFC,ResouceManager,nodemanager |
| 192.168.1.4 | slave1.hadoop | jdk.hadoop.zookeeper | datanode,nodemanager |
1.修改主机名
[root@master /]# vi /etc/hostname #主机名
master.hadoop[root@master ~]# hostname master.hadoop
[root@master ~]# hostname
master.hadoop其余结点一样的操作
[root@master ~]# cd /etc/hosts
192.168.1.2 master.hadoop
192.168.1.3 slave1.hadoop
192.168.1.4 slave2.hadoop
[root@master ~]# scp /etc/hosts 192.168.1.3:/etc
[root@master ~]# scp /etc/hosts 192.168.1.4:/etc2.修改网卡配置
硬件配置：取消动态ip地址，使用固定的
[root@master network-scripts]# cd /etc/sysconfig/network-scripts/
[root@master network-scripts]# ll
总用量 244
-rw-r--r--. 1 root root 363 7月 6 08:09 ifcfg-ens33
-rw-r--r--. 1 root root 254 5月 3 2017 ifcfg-lo
lrwxrwxrwx. 1 root root 24 4月 27 06:30 ifdown -> ../../../usr/sbin/ifdown
-rwxr-xr-x. 1 root root 654 5月 3 2017 ifdown-bnep
-rwxr-xr-x. 1 root root 6571 5月 3 2017 ifdown-eth
-rwxr-xr-x. 1 root root 6190 8月 4 2017 ifdown-ib
[root@master network-scripts]# vi ifcfg-ens33
[root@master network-scripts]# TYPE=Ethernet
PROXY_METHOD=none
BROWSER_ONLY=no
BOOTPROTO=none
DEFROUTE=yes
IPV4_FAILURE_FATAL=yes
IPV6INIT=yes
IPV6_AUTOCONF=yes
IPV6_DEFROUTE=yes
IPV6_FAILURE_FATAL=no
IPV6_ADDR_GEN_MODE=stable-privacy
NAME=ens33
UUID=f090a391-d137-4d93-8594-03baeada0d1f
DEVICE=ens33
ONBOOT=yes
IPADDR=192.168.1.2
PREFIX=24
GATEWAY=192.168.1.1
IPV6_PRIVACY=no
DNS1=192.168.1.13.重启网卡
[root@master network-scripts]# service network restart
#查看网络配置
[root@master network-scripts]# ifconfig
ens33: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
inet 192.168.1.2 netmask 255.255.255.0 broadcast 192.168.1.255
inet6 fe80::5d8a:5d86:d69a:1d54 prefixlen 64 scopeid 0x20<link>
ether 00:0c:29:be:6c:d6 txqueuelen 1000 (Ethernet)
RX packets 2787 bytes 199847 (195.1 KiB)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 1488 bytes 96410 (94.1 KiB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
lo: flags=73<UP,LOOPBACK,RUNNING> mtu 65536
inet 127.0.0.1 netmask 255.0.0.0
inet6 ::1 prefixlen 128 scopeid 0x10<host>
loop txqueuelen 1 (Local Loopback)
RX packets 1477 bytes 143901 (140.5 KiB)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 1477 bytes 143901 (140.5 KiB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
virbr0: flags=4099<UP,BROADCAST,MULTICAST> mtu 1500
inet 192.168.122.1 netmask 255.255.255.0 broadcast 192.168.122.255
ether 52:54:00:a0:aa:ad txqueuelen 1000 (Ethernet)
RX packets 0 bytes 0 (0.0 B)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 0 bytes 0 (0.0 B)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 04、centos7关闭防火墙
//临时关闭
systemctl stop firewalld
//禁止开机启动
systemctl disable firewalld
Removed symlink /etc/systemd/system/multi-user.target.wants/firewalld.service.
Removed symlink /etc/systemd/system/dbus-org.fedoraproject.FirewallD1.service.二、安装JDK
1、下载linux版本的jdk
jdk1.8下载地址
2、解压配置并环境变量
[root@master apps]# tar -zxvf jdk-8u171-linux-x64.tar.gz
[root@master apps]# vi /etc/profile
在其中添加一下内容：
export JAVA_HOME=/apps/jdk1.8.0_171
export JRE_HOME=/apps/jdk1.8.0_171/jre
export CLASS_PATH=.:$JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar:$JRE_HOME/lib
export PATH=$PATH:$JAVA_HOME/bin:$JRE_HOME/bin然后更新配置文件
[root@master apps]# source /etc/profile[root@master apps]# java -version
java version "1.8.0_171"
Java(TM) SE Runtime Environment (build 1.8.0_171-b11)
Java HotSpot(TM) 64-Bit Server VM (build 25.171-b11, mixed mode)三、配置SSH免密登录
1、每台机器生成自己的私钥和公钥
[root@master apps]# cd /root/.ssh/
[root@master .ssh]# ssh-keygen -t rsa
然后一直回车什么也不输入，直到结束。
[root@master .ssh]# touch authorized_keys
[root@master .ssh]# ll
总用量 16
-rw-r--r--. 1 root root 2000 7月 4 20:48 authorized_keys
-rw-------. 1 root root 1679 6月 19 15:17 id_rsa
-rw-r--r--. 1 root root 400 6月 19 15:17 id_rsa.pub
-rw-r--r--. 1 root root 935 7月 4 20:54 known_hosts
[root@master .ssh]# cat id_rsa.pub >> authorized_keys在另外两台机器同样的执行以下操作：
[root@slave apps]# cd /root/.ssh/
[root@salve .ssh]# ssh-keygen -t rsa
[root@slave .ssh]# ll
-rw-------. 1 root root 1679 6月 19 15:17 id_rsa
-rw-r--r--. 1 root root 400 6月 19 15:17 id_rsa.pub
-rw-r--r--. 1 root root 935 7月 4 20:54 known_hosts
将自己机器上的id_rsa.pub追加到主节点的authorized_keys文件中。
[root@slave .ssh]# scp id_rsa.pub 192.168.1.2:/
去主节点追加
[root@master .ssh]# cd /
[root@master .ssh]# cat id_rsa.pub >> /root/.ssh/authorized_keys3、将主节点上的authorized_keys分发给每一个从节点，测试登录
[root@master .ssh]# scp authorized_keys 192.168.1.3:$PWD
[root@master .ssh]# scp authorized_keys 192.168.1.4:$PWD
[root@master .ssh]# ssh 192.168.1.3
Last login: Fri Jul 6 16:02:23 2018
[root@slave1 ~]# 四、hadoop2.8.0+zookeeper-3.4.10搭建HA
1、安装zookeeper
下载zookeeper下载链接
在三台机器上都安装zookeeper
解压到/apps
[root@master apps]# tar -zxvf zookeeper-3.4.10.tar.gz #删除没用的文件，只留下一下部分
[root@master apps]# cd zookeeper-3.4.10/
[root@master zookeeper-3.4.10]# ll
总用量 1456
drwxr-xr-x. 2 root root 149 3月 23 2017 bin
drwxr-xr-x. 2 root root 91 7月 5 11:28 conf
drwxr-xr-x. 4 root root 267 3月 23 2017 lib
-rw-rw-r--. 1 root root 1456729 3月 23 2017 zookeeper-3.4.10.jar配置环境变量
[root@master /]# vi /etc/profile#set java environment
export JAVA_HOME=/apps/jdk1.8.0_171
export JRE_HOME=/apps/jdk1.8.0_171/jre
export CLASS_PATH=.:$JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar:$JRE_HOME/lib
export ZOOKEEPER_HOME=/apps/zookeeper-3.4.10
export PATH=$PATH:$JAVA_HOME/bin:$ZOOKEEPER_HOME/bin[root@master /]# source /etc/profile切换到conf目录，修改zookeepe配置文件：
将zoo_sample.cfg改名为zoo.cfg
[root@master conf]# mv zoo_sample.cfg zoo.cfg [root@master zookeeper-3.4.10]# cd conf/
[root@master conf]# ll
总用量 40
-rw-rw-r--. 1 root root 535 3月 23 2017 configuration.xsl
-rw-rw-r--. 1 root root 2161 3月 23 2017 log4j.properties
-rw-rw-r--. 1 root root 1019 7月 5 11:19 zoo.cfg提前建好这两个目录（/apps/zkdata）（/apps/zookeeper-3.4.10/datalog）
dataDir=/apps/zkdata
dataLogDir=/apps/zookeeper-3.4.10/datalog末尾添加：
server.1=master.hadoop:2888:3888
server.2=slave1.hadoop:2888:3888
server.3=slave2.hadoop:2888:3888
# The number of milliseconds of each tick
tickTime=2000
# The number of ticks that the initial
# synchronization phase can take
initLimit=10
# The number of ticks that can pass between
# sending a request and getting an acknowledgement
syncLimit=5
# the directory where the snapshot is stored.
# do not use /tmp for storage, /tmp here is just
# example sakes.
dataDir=/apps/zkdata
dataLogDir=/apps/zookeeper-3.4.10/datalog
# the port at which the clients will connect
clientPort=2181
# the maximum number of client connections.
# increase this if you need to handle more clients
#maxClientCnxns=60
#
# Be sure to read the maintenance section of the
# administrator guide before turning on autopurge.
#
# http://zookeeper.apache.org/doc/current/zookeeperAdmin.html#sc_maintenance
#
# The number of snapshots to retain in dataDir
#autopurge.snapRetainCount=3
# Purge task interval in hours
# Set to "0" to disable auto purge feature
#autopurge.purgeInterval=1
server.1=master.hadoop:2888:3888
server.2=slave1.hadoop:2888:3888
server.3=slave2.hadoop:2888:3888[root@master conf]# cd /apps/zkdata
[root@master zkdata]# echo 1 > myidslave1.hadoop: 在zkdata目录下创建myid文件并添加内容为2(以此类推)
[root@slave1 conf]# cd /apps/zkdata
[root@slave1 zkdata]# echo 2 > myid[root@master /]# zkServer.sh start
ZooKeeper JMX enabled by default
Using config: /apps/zookeeper-3.4.10/bin/../conf/zoo.cfg
Starting zookeeper ... STARTED[root@slave1 ~]# zkServer.sh start
ZooKeeper JMX enabled by default
Using config: /apps/zookeeper-3.4.10/bin/../conf/zoo.cfg
Starting zookeeper ... STARTED[root@slave2 ~]# zkServer.sh start
ZooKeeper JMX enabled by default
Using config: /apps/zookeeper-3.4.10/bin/../conf/zoo.cfg
Starting zookeeper ... STARTED[root@master /]# jps
43587 QuorumPeerMain
43871 Jps
[root@master /]# zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /apps/zookeeper-3.4.10/bin/../conf/zoo.cfg
Mode: follower
[root@master /]# [root@slave1 ~]# jps
2709 QuorumPeerMain
2757 Jps
[root@slave1 ~]# zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /apps/zookeeper-3.4.10/bin/../conf/zoo.cfg
Mode: leader
[root@slave1 ~]# [root@slave2 ~]# jps
1784 Jps
1759 QuorumPeerMain
[root@slave2 ~]# zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /apps/zookeeper-3.4.10/bin/../conf/zoo.cfg
Mode: follower
[root@slave2 ~]#
[root@master /]# zkServer.sh stop[root@slave1 ~]# zkServer.sh stop[root@slave2 ~]# zkServer.sh stop2、安装hadoop
下载hadoop安装包
解压安装包，并修改配置文件
[root@master apps]# tar -zxvf hadoop-2.8.0.tar.gz
[root@master apps]# mkdir hdptmp
在两台从节点上建立相同的目录(hdfs初始化的时候会用到)1）配置hadoop-env.sh
# set java environment（添加jdk环境变量）
export JAVA_HOME=/apps/jdk1.8.0_1712）配置core-site.xml文件
修改Hadoop核心配置文件core-site.xml，这里配置的是HDFS的地址和端口号。
core-site.xml
<configuration>
<!-- 指定hdfs的nameservice为ns -->
<property>
<name>fs.defaultFS</name>
<value>hdfs://ns</value>
</property>
<!--指定hadoop数据临时存放目录-->
<property>
<name>hadoop.tmp.dir</name>
<value>/apps/hdptmp</value>
</property>
<!--指定zookeeper地址-->
<property>
<name>ha.zookeeper.quorum</name>
<value>master.hadoop:21810,slave1.hadoop:21810,slave2.hadoop:21810</value>
</property>
</configuration>3）配置hdfs-site.xml文件
修改Hadoop中HDFS的配置，配置的备份方式默认为3。
<configuration>
<!--指定hdfs的nameservice为ns，需要和core-site.xml中的保持一致 -->
<property>
<name>dfs.nameservices</name>
<value>ns</value>
</property>
<!-- ns下面有两个NameNode，分别是nn1，nn2 -->
<property>
<name>dfs.ha.namenodes.ns</name>
<value>nn1,nn2</value>
</property>
<!-- nn1的RPC通信地址 -->
<property>
<name>dfs.namenode.rpc-address.ns.nn1</name>
<value>master.hadoop:9000</value>
</property>
<!-- nn1的http通信地址 -->
<property>
<name>dfs.namenode.http-address.ns.nn1</name>
<value>master.hadoop:50070</value>
</property>
<!-- nn2的RPC通信地址 -->
<property>
<name>dfs.namenode.rpc-address.ns.nn2</name>
<value>slave1.hadoop:9000</value>
</property>
<!-- nn2的http通信地址 -->
<property>
<name>dfs.namenode.http-address.ns.nn2</name>
<value>slave1.hadoop:50070</value>
</property>
<!-- 指定NameNode的元数据在JournalNode上的存放位置 -->
<property>
<name>dfs.namenode.shared.edits.dir</name>
<value>qjournal://master.hadoop:8485;slave1.hadoop:8485;slave2.hadoop:8485/ns</value>
</property>
<!-- 指定JournalNode在本地磁盘存放数据的位置 -->
<property>
<name>dfs.journalnode.edits.dir</name>
<value>/apps/journal</value>
</property>
<!-- 开启NameNode故障时自动切换 -->
<property>
<name>dfs.ha.automatic-failover.enabled</name>
<value>true</value>
</property>
<!-- 配置失败自动切换实现方式 -->
<property>
<name>dfs.client.failover.proxy.provider.ns</name>
<value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>
</property>
<!-- 配置隔离机制 -->
<property>
<name>dfs.ha.fencing.methods</name>
<value>
sshfence
shell(/bin/true)
</value>
</property>
<!-- 使用隔离机制时需要ssh免登陆 -->
<property>
<name>dfs.ha.fencing.ssh.private-key-files</name>
<value>/root/.ssh/id_rsa</value>
</property>
<property>
<name>dfs.replication</name>
<value>2</value>
</property>
<!-- 在NN和DN上开启WebHDFS (REST API)功能,不是必须 -->
<property>
<name>dfs.webhdfs.enabled</name>
<value>true</value>
</property>
<property>
<name>dfs.ha.fencing.ssh.connect-timeout</name>
<value>10000</value>
</property>
<property>
<name>dfs.permissions</name>
<value>false</value>
</property>
</configuration>4）配置mapred-site.xml文件
修改Hadoop中MapReduce的配置文件，配置的是JobTracker的地址和端口。
<configuration>
<property>
<name>mapreduce.framework.name</name>
<value>yarn</value>
</property>
</configuration><configuration>
<!-- //////////////以下为YARN HA的配置////////////// -->
<!-- 开启YARN HA -->
<property>
<name>yarn.resourcemanager.ha.enabled</name>
<value>true</value>
</property>
<!-- 启用自动故障转移 -->
<property>
<name>yarn.resourcemanager.ha.automatic-failover.enabled</name>
<value>true</value>
</property>
<!-- 指定YARN HA的名称 -->
<property>
<name>yarn.resourcemanager.cluster-id</name>
<value>yarncluster</value>
</property>
<!-- 指定两个resourcemanager的名称 -->
<property>
<name>yarn.resourcemanager.ha.rm-ids</name>
<value>rm1,rm2</value>
</property>
<!-- 配置rm1，rm2的主机 -->
<property>
<name>yarn.resourcemanager.hostname.rm1</name>
<value>master.hadoop1</value>
</property>
<property>
<name>yarn.resourcemanager.hostname.rm2</name>
<value>slave1.hadoop</value>
</property>
<!-- 配置zookeeper的地址 -->
<property>
<name>yarn.resourcemanager.zk-address</name>
<value>master.hadoop:2181,slave1.hadoop:2181,slave2.hadoop:2181</value>
</property>
<property>
<name>yarn.nodemanager.aux-services</name>
<value>mapreduce_shuffle</value>
</property>
</configuration>6)配置slaves文件
[root@master /]# vi slaves
master.hadoop
slave1.hadoop
slave2.hadoop7) 配置hadoop环境变量
[root@master /]# vi /etc/profile
#set hadoop enviroment
export HADOOP_HOME=/apps/hadoop-2.8.0/
export PATH=$PATH:$JAVA_HOME/bin:$JRE_HOME/bin:$HADOOP_HOME/bin:$HADOOP_HOME/sbin#生效
[root@master /]# source /etc/profile启动集群（按照顺序启动）
1.启动zookeeper（三台机器一台一台的启动）
[root@master /]# zkServer.sh start
ZooKeeper JMX enabled by default
Using config: /apps/zookeeper-3.4.10/bin/../conf/zoo.cfg
Starting zookeeper ... STARTED[root@slave1 ~]# zkServer.sh start
ZooKeeper JMX enabled by default
Using config: /apps/zookeeper-3.4.10/bin/../conf/zoo.cfg
Starting zookeeper ... STARTED[root@slave2 ~]# zkServer.sh start
ZooKeeper JMX enabled by default
Using config: /apps/zookeeper-3.4.10/bin/../conf/zoo.cfg
Starting zookeeper ... STARTED2.启动journalnode（三台一台一台启动）
[root@master /]# hadoop-daemon.sh start journalnode
starting journalnode, logging to /apps/hadoop-2.8.0/logs/hadoop-root-journalnode-master.hadoop.out
[root@master /]#
[root@slave1 ~]# hadoop-daemon.sh start journalnode
starting journalnode, logging to /apps/hadoop-2.8.0/logs/hadoop-root-journalnode-slave1.hadoop.out
[root@slave1 ~]#
[root@slave2 ~]# hadoop-daemon.sh start journalnode
starting journalnode, logging to /apps/hadoop-2.8.0/logs/hadoop-root-journalnode-slave2.hadoop.out
[root@slave2 ~]# 3.启动hdfs
[root@master /]# start-dfs.sh
Starting namenodes on [master.hadoop slave1.hadoop]
slave1.hadoop: starting namenode, logging to /apps/hadoop-2.8.0/logs/hadoop-root-namenode-slave1.hadoop.out
master.hadoop: starting namenode, logging to /apps/hadoop-2.8.0/logs/hadoop-root-namenode-master.hadoop.out
slave2.hadoop: starting datanode, logging to /apps/hadoop-2.8.0/logs/hadoop-root-datanode-slave2.hadoop.out
master.hadoop: starting datanode, logging to /apps/hadoop-2.8.0/logs/hadoop-root-datanode-master.hadoop.out
slave1.hadoop: starting datanode, logging to /apps/hadoop-2.8.0/logs/hadoop-root-datanode-slave1.hadoop.out
Starting journal nodes [master.hadoop slave1.hadoop slave2.hadoop]
slave2.hadoop: journalnode running as process 1870. Stop it first.
slave1.hadoop: journalnode running as process 2842. Stop it first.
master.hadoop: journalnode running as process 45029. Stop it first.
Starting ZK Failover Controllers on NN hosts [master.hadoop slave1.hadoop]
slave1.hadoop: starting zkfc, logging to /apps/hadoop-2.8.0/logs/hadoop-root-zkfc-slave1.hadoop.out
master.hadoop: starting zkfc, logging to /apps/hadoop-2.8.0/logs/hadoop-root-zkfc-master.hadoop.out
[root@master /]# 4.启动yarn
[root@master /]# start-yarn.sh
starting yarn daemons
starting resourcemanager, logging to /apps/hadoop-2.8.0/logs/yarn-root-resourcemanager-master.hadoop.out
slave1.hadoop: starting nodemanager, logging to /apps/hadoop-2.8.0/logs/yarn-root-nodemanager-slave1.hadoop.out
slave2.hadoop: starting nodemanager, logging to /apps/hadoop-2.8.0/logs/yarn-root-nodemanager-slave2.hadoop.out
master.hadoop: starting nodemanager, logging to /apps/hadoop-2.8.0/logs/yarn-root-nodemanager-master.hadoop.out
[root@master /]# 5.启动单个结点的yarn进程（因为yarn的两个resourcemanager不会同时启动，必须手动启动另一台上的resourcemanager）
去slave1.hadoop上启动
[root@slave1 ~]# yarn-daemon.sh start resourcemanager
starting resourcemanager, logging to /apps/hadoop-2.8.0/logs/yarn-root-resourcemanager-slave1.hadoop.out
[root@slave1 ~]# jps
3489 Jps
2931 NameNode
2709 QuorumPeerMain
2842 JournalNode
3131 DFSZKFailoverController
3451 ResourceManager
3276 NodeManager
3005 DataNode
[root@slave1 ~]# 到此集群搭建完成！！！！！！！！！！！！！！！！！
============================================================================测试高可用
访问 http://slave1.hadoop:50070/
接下来是一台宕机（关闭slave1.hadoop上的namenode进程）杀掉2931进程
[root@slave1 ~]# jps
2931 NameNode
2709 QuorumPeerMain
3733 Jps
2842 JournalNode
3131 DFSZKFailoverController
3451 ResourceManager
3276 NodeManager
3005 DataNode
[root@slave1 ~]# kill -9 2931
[root@slave1 ~]# jps
2709 QuorumPeerMain
3753 Jps
2842 JournalNode
3131 DFSZKFailoverController
3451 ResourceManager
3276 NodeManager
3005 DataNode
[root@slave1 ~]# 然后访问 http://slave1.hadoop:50070/（无响应）
访问http://master.hadoop:50070/ （变成active）
============================================================================