一、概述
因为实验所需的5台机器配置大多相同，所以采用配置其中一台，然后使用克隆功能复制另外4份再做具体修改；
其中有些步骤以前配置过，此处就说明一下不再做具体配置，具体配置可翻阅以前的博文。
二、实验环境
1.关闭selinux和firewall
三、主机规划
| IP | Host | 进程 | |
| 192.168.100.11 | hadoop1 | NameNode ResourceManager DFSZKFailoverController |
|
| 192.168.100.12 | hadoop2 | NameNode ResourceManager DFSZKFailoverController |
|
| 192.168.100.13 | hadoop3 | DataNode NodeManager JournalNode QuorumPeerMain |
|
| 192.168.100.14 | hadoop4 | DataNode NodeManager JournalNode QuorumPeerMain |
|
| 192.168.100.15 | hadoop5 | DataNode NodeManager JournalNode QuorumPeerMain |
|
四、环境准备
1.设置IP地址：192.168.100.11
2.设置主机名：hadoop1
3.设置IP和主机名的映射
[root@hadoop1 ~]# cat /etc/hosts
127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4
::1 localhost localhost.localdomain localhost6 localhost6.localdomain6
192.168.100.11 hadoop1
192.168.100.12 hadoop2
192.168.100.13 hadoop3
192.168.100.14 hadoop4
192.168.100.15 hadoop5
4.配置ssh分发脚本
5.解压jdk
[root@hadoop1 ~]# tar -zxf jdk-8u131-linux-x64.tar.gz
[root@hadoop1 ~]# cp -r jdk1.8.0_131/ /usr/local/jdk
6.解压hadoop
[root@hadoop1 ~]# tar -zxf hadoop-2.7.4.tar.gz
[root@hadoop1 ~]# cp -r hadoop-2.7.4 /usr/local/hadoop
7.解压zookeeper
[root@hadoop1 ~]# tar -zxf zookeeper-3.4.10.tar.gz
[root@hadoop1 ~]# cp -r zookeeper-3.4.10 /usr/local/hadoop/zookeeper
[root@hadoop1 ~]# cd /usr/local/hadoop/zookeeper/conf/
[root@hadoop1 conf]# cp zoo_sample.cfg zoo.cfg
[root@hadoop1 conf]# vim zoo.cfg
#修改dataDir
dataDir=/usr/local/hadoop/zookeeper/data
#添加下面三行
server.1=hadoop3:2888:3888
server.2=hadoop4:2888:3888
server.3=hadoop5:2888:3888
[root@hadoop1 conf]# cd ..
[root@hadoop1 zookeeper]# mkdir data
#此处还有操作，但是hadoop1上不部署zookeeper模块所以后面再修改
8.配置环境变量
[root@hadoop1 ~]# tail -4 /etc/profile
export JAVA_HOME=/usr/local/jdk
export HADOOP_HOME=/usr/local/hadoop
export ZOOKEEPER_HOME=/usr/local/hadoop/zookeeper
export PATH=.:$HADOOP_HOME/bin:$HADOOP_HOME/sbin:$JAVA_HOME/bin:$ZOOKEEPER_HOME/bin:$PATH
[root@hadoop1 ~]# source /etc/profile
9.测试环境变量可用
[root@hadoop1 ~]# java -version
java version "1.8.0_131"
Java(TM) SE Runtime Environment (build 1.8.0_131-b11)
Java HotSpot(TM) 64-Bit Server VM (build 25.131-b11, mixed mode)
[root@hadoop1 ~]# hadoop version
Hadoop 2.7.4
Subversion Unknown -r Unknown
Compiled by root on 2017-08-28T09:30Z
Compiled with protoc 2.5.0
From source with checksum 50b0468318b4ce9bd24dc467b7ce1148
This command was run using /usr/local/hadoop/share/hadoop/common/hadoop-common-2.7.4.jar
五、配置hadoop
1.core-site.xml
<configuration>
<!-- 指定hdfs的nameservice为master -->
<property>
<name>fs.defaultFS</name>
<value>hdfs://master/</value>
</property>
<!-- 指定hadoop临时目录 -->
<property>
<name>hadoop.tmp.dir</name>
<value>/usr/local/hadoop/tmp</value>
</property>
<!-- 指定zookeeper地址 -->
<property>
<name>ha.zookeeper.quorum</name>
<value>hadoop3:2181,hadoop4:2181,hadoop5:2181</value>
</property>
</configuration>
2.hdfs-site.xml
<configuration>
<property>
<name>dfs.namenode.name.dir</name>
<value>/usr/local/hadoop/dfs/name</value>
</property>
<property>
<name>dfs.datanode.data.dir</name>
<value>/usr/local/hadoop/dfs/data</value>
</property>
<property>
<name>dfs.replication</name>
<value>2</value>
</property>
<!--HDFS高可用配置 -->
<!--指定hdfs的nameservice，需要和core-site.xml中的保持一致-->
<property>
<name>dfs.nameservices</name>
<value>master</value>
</property>
<!--指定master的两个namenode的名称 -->
<property>
<name>dfs.ha.namenodes.master</name>
<value>nn1,nn2</value>
</property>
<!-- nn1,nn2 rpc 通信地址 -->
<property>
<name>dfs.namenode.rpc-address.master.nn1</name>
<value>hadoop1:9000</value>
</property>
<property>
<name>dfs.namenode.rpc-address.master.nn2</name>
<value>hadoop2:9000</value>
</property>
<!-- nn1.nn2 http 通信地址 -->
<property>
<name>dfs.namenode.http-address.master.nn1</name>
<value>hadoop1:50070</value>
</property>
<property>
<name>dfs.namenode.http-address.master.nn2</name>
<value>hadoop2:50070</value>
</property>
<!--=========Namenode同步==========-->
<!--保证数据恢复 -->
<property>
<name>dfs.journalnode.http-address</name>
<value>0.0.0.0:8480</value>
</property>
<property>
<name>dfs.journalnode.rpc-address</name>
<value>0.0.0.0:8485</value>
</property>
<property>
<!--指定NameNode的元数据在JournalNode上的存放位置 -->
<name>dfs.namenode.shared.edits.dir</name>
<value>qjournal://hadoop3:8485;hadoop4:8485;hadoop5:8485/master</value>
</property>
<property>
<!--JournalNode存放数据地址 -->
<name>dfs.journalnode.edits.dir</name>
<value>/usr/local/hadoop/dfs/journal</value>
</property>
<property>
<!--NameNode失败自动切换实现方式 -->
<name>dfs.client.failover.proxy.provider.master</name>
<value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>
</property>
<!--=========Namenode fencing：======== -->
<!-- 配置隔离机制方法，多个机制用换行分割，即每个机制暂用一行 -->
<property>
<name>dfs.ha.fencing.methods</name>
<value>sshfence
shell(/bin/true)</value>
</property>
<!-- 使用sshfence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>
<!-- 配置sshfence隔离机制超时时间 -->
<property>
<name>dfs.ha.fencing.ssh.connect-timeout</name>
<value>30000</value>
</property>
<!--开启基于Zookeeper及ZKFC进程的自动备援设置,监视进程是否死掉 -->
<property>
<name>dfs.ha.automatic-failover.enabled</name>
<value>true</value>
</property>
<property>
<name>ha.zookeeper.quorum</name>
<value>hadoop3:2181,hadoop4:2181,hadoop5:2181</value>
</property>
<property>
<!--指定ZooKeeper超时间隔，单位毫秒 -->
<name>ha.zookeeper.session-timeout.ms</name>
<value>2000</value>
</property>
</configuration>
3.yarn-site.xml
<configuration>
<!--NodeManager上运行的附属服务。需配置成mapreduce_shuffle，才可运行MapReduce程序-->
<property>
<name>yarn.nodemanager.aux-services</name>
<value>mapreduce_shuffle</value>
</property>
<property>
<name>yarn.resourcemanager.connect.retry-interval.ms</name>
<value>2000</value>
</property>
<property>
<name>yarn.resourcemanager.ha.enabled</name>
<value>true</value>
</property>
<!-- 指定RM的cluster id -->
<property>
<name>yarn.resourcemanager.cluster-id</name>
<value>yrc</value>
</property>
<!--指定两台RM主机名标识符-->
<property>
<name>yarn.resourcemanager.ha.rm-ids</name>
<value>rm1,rm2</value>
</property>
<!--RM主机1-->
<property>
<name>yarn.resourcemanager.hostname.rm1</name>
<value>hadoop1</value>
</property>
<!--RM主机2-->
<property>
<name>yarn.resourcemanager.hostname.rm2</name>
<value>hadoop2</value>
</property>
<!--RM故障自动切换-->
<property>
<name>yarn.resourcemanager.ha.automatic-failover.enabled</name>
<value>true</value>
</property>
<!--RM故障自动恢复 -->
<property>
<name>yarn.resourcemanager.recovery.enabled</name>
<value>true</value>
</property>
<!--RM状态信息存储方式，一种基于内存(MemStore)，另一种基于ZK(ZKStore)-->
<property>
<name>yarn.resourcemanager.store.class</name>
<value>org.apache.hadoop.yarn.server.resourcemanager.recovery.ZKRMStateStore</value>
</property>
<!-- 指定zk集群地址 -->
<property>
<name>yarn.resourcemanager.zk-address</name>
<value>hadoop3:2181,hadoop4:2181,hadoop5:2181</value>
</property>
<!--向RM调度资源地址-->
<property>
<name>yarn.resourcemanager.scheduler.address.rm1</name>
<value>hadoop1:8030</value>
</property>
<property>
<name>yarn.resourcemanager.scheduler.address.rm2</name>
<value>hadoop2:8030</value>
</property>
<!--NodeManager通过该地址交换信息-->
<property>
<name>yarn.resourcemanager.resource-tracker.address.rm1</name>
<value>hadoop1:8031</value>
</property>
<property>
<name>yarn.resourcemanager.resource-tracker.address.rm2</name>
<value>hadoop2:8031</value>
</property>
<!--客户端通过该地址向RM提交对应用程序操作-->
<property>
<name>yarn.resourcemanager.address.rm1</name>
<value>hadoop1:8032</value>
</property>
<property>
<name>yarn.resourcemanager.address.rm2</name>
<value>hadoop2:8032</value>
</property>
<!--管理员通过该地址向RM发送管理命令-->
<property>
<name>yarn.resourcemanager.admin.address.rm1</name>
<value>hadoop1:8033</value>
</property>
<property>
<name>yarn.resourcemanager.admin.address.rm2</name>
<value>hadoop2:8033</value>
</property>
<!--RM HTTP访问地址,查看集群信息-->
<property>
<name>yarn.resourcemanager.webapp.address.rm1</name>
<value>hadoop1:8088</value>
</property>
<property>
<name>yarn.resourcemanager.webapp.address.rm2</name>
<value>hadoop2:8088</value>
</property>
</configuration>
4.mapred-site.xml
<configuration>
<!-- 指定mr框架为yarn方式 -->
<property>
<name>mapreduce.framework.name</name>
<value>yarn</value>
</property>
<!-- 配置 MapReduce JobHistory Server地址 ，默认端口10020 -->
<property>
<name>mapreduce.jobhistory.address</name>
<value>hadoop1:10020</value>
</property>
<!-- 配置 MapReduce JobHistory Server HTTP地址， 默认端口19888 -->
<property>
<name>mapreduce.jobhistory.webapp.address</name>
<value>hadoop1:19888</value>
</property>
</configuration>
5.slaves
[root@hadoop1 hadoop]# cat slaves
hadoop3
hadoop4
hadoop5
6.hadoop-env.sh
export JAVA_HOME=/usr/local/jdk #在后面添加
六、克隆虚拟机
1.使用hadoop1为模板克隆4台虚拟机，并将网卡的MAC地址重新生成
2.修改主机名为hadoop2-hadoop5
3.修改IP地址
4.配置所有机器之间的ssh免密登陆（ssh公钥分发）
七、配置zookeeper
[root@hadoop3 ~]# echo 1 > /usr/local/hadoop/zookeeper/data/myid #在hadoop3上
[root@hadoop4 ~]# echo 2 > /usr/local/hadoop/zookeeper/data/myid #在hadoop4上
[root@hadoop5 ~]# echo 3 > /usr/local/hadoop/zookeeper/data/myid #在hadoop5上
八、启动集群
1.在hadoop3-5上启动zookeeper
[root@hadoop3 ~]# zkServer.sh start
ZooKeeper JMX enabled by default
Using config: /usr/local/hadoop/zookeeper/bin/../conf/zoo.cfg
Starting zookeeper ... STARTED
[root@hadoop3 ~]# zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /usr/local/hadoop/zookeeper/bin/../conf/zoo.cfg
Mode: follower
[root@hadoop3 ~]# jps
2184 QuorumPeerMain
2237 Jps
#hadoop4和hadoop5相同操作
2.在hadoop1上格式化 ZooKeeper 集群
[root@hadoop1 ~]# hdfs zkfc -formatZK
3.在hadoop3-5上启动journalnode
[root@hadoop3 ~]# hadoop-daemon.sh start journalnode
starting journalnode, logging to /usr/local/hadoop/logs/hadoop-root-journalnode-hadoop3.out
[root@hadoop3 ~]# jps
2244 JournalNode
2293 Jps
2188 QuorumPeerMain
4.在hadoop1上格式化namenode
[root@hadoop1 ~]# hdfs namenode -format
...
17/08/29 22:53:30 INFO util.ExitUtil: Exiting with status 0
17/08/29 22:53:30 INFO namenode.NameNode: SHUTDOWN_MSG:
/************************************************************
SHUTDOWN_MSG: Shutting down NameNode at hadoop1/192.168.100.11
************************************************************/
5.在hadoop1上启动刚格式化的namenode
[root@hadoop1 ~]# hadoop-daemon.sh start namenode
starting namenode, logging to /usr/local/hadoop/logs/hadoop-root-namenode-hadoop1.out
[root@hadoop1 ~]# jps
2422 Jps
2349 NameNode
6.在hadoop2上同步nn1（hadoop1）数据到nn2（hadoop2）
12345678910 [root@hadoop2 ~]# hdfs namenode -bootstrapStandby
...
17/08/29 22:55:45 INFO namenode.TransferFsImage: Image Transfer timeout configured to 60000 milliseconds
17/08/29 22:55:45 INFO namenode.TransferFsImage: Transfer took 0.00s at 0.00 KB/s
17/08/29 22:55:45 INFO namenode.TransferFsImage: Downloaded file fsimage.ckpt_0000000000000000000 size 321 bytes.
17/08/29 22:55:45 INFO util.ExitUtil: Exiting with status 0
17/08/29 22:55:45 INFO namenode.NameNode: SHUTDOWN_MSG:
/************************************************************
SHUTDOWN_MSG: Shutting down NameNode at hadoop2/192.168.100.12
************************************************************/
7.启动hadoop2上的namenode
[root@hadoop2 ~]# hadoop-daemon.sh start namenode
8.启动集群中的所有服务
[root@hadoop1 ~]# start-all.sh
9.在hadoop2上启动yarn
[root@hadoop2 ~]# yarn-daemon.sh start resourcemanager
10.开启historyserver
[root@hadoop1 ~]# mr-jobhistory-daemon.sh start historyserver
starting historyserver, logging to /usr/local/hadoop/logs/mapred-root-historyserver-hadoop1.out
[root@hadoop1 ~]# jps
3026 DFSZKFailoverController
3110 ResourceManager
3894 JobHistoryServer
3927 Jps
2446 NameNode
11.查看进程
[root@hadoop3 ~]# jps
2480 DataNode
2722 Jps
2219 JournalNode
2174 QuorumPeerMain
2606 NodeManager
[root@hadoop4 ~]# jps
2608 NodeManager
2178 QuorumPeerMain
2482 DataNode
2724 Jps
2229 JournalNode
[root@hadoop5 ~]# jps
2178 QuorumPeerMain
2601 NodeManager
2475 DataNode
2717 Jps
2223 JournalNode
九、测试
1.连接
2.kill hadoop2上的namenode
[root@hadoop2 ~]# jps
2742 NameNode
3016 DFSZKFailoverController
4024 JobHistoryServer
4057 Jps
3133 ResourceManager
[root@hadoop2 ~]# kill -9 2742
[root@hadoop2 ~]# jps
3016 DFSZKFailoverController
3133 ResourceManager
4205 Jps
