VMware 搭建 hadoop 完全分布式集群

VMware 虚拟机搭建 hadoop 完全分布式集群，集成 Spark、HBASE、Hive 等

风老魔

1888人浏览 · 2023-01-17 20:03:19

风老魔 · 2023-01-17 20:03:19 发布

1. 虚拟机安装、克隆

使用VMware安装CentOS6.8详细教程

2. 网络配置

# 修改 hostname
hostname hadoop1

hostnamectl set-hostname hadoop1

# 最靠谱方式
Vim /etc/sysconfig/network

打开网卡：

centos 安装后默认是关闭网卡的，要进 /etc/sysconfig/network-scripts/ 下修改 ifcfg-ens33，将ONBOOT 的值改成 yes。ONBOOT是指明在系统启动时是否激活网卡，只有在激活状态的网卡才能连接网络，进行网络通信

修改完毕后，service network restart 重启网络，检查 ping baidu.com 是否能够通

克隆虚拟机修改 onboot 后重启网络失败

错误类型：Bringing up interface eth0:Device eth0 does not seem to be present,delaying
解决方法：删除 /etc/udev/rules.d/70-persistent-net.rules，然后将网卡配置文件 /etc/sysconfig/network-scripts/ifcfg-eth0 的 uuid 和 hwaddr 这两行删除，重启系统即可！
参考：https://jingyan.baidu.com/article/e75aca85006645142edac6df.html

注意：有些没有 ifcfg-ens33，只有 ifcfg-eth0，修改 ifcfg-eth0 即可！

2.1 设置静态 IP、主机映射

设置静态 IP

以下操作每台服务器 hadoop1/2/3 都需要设置：

1、Vim /etc/sysconfig/network-scripts/ifcfg-ens33，有些可能会是 ifcfg-eth0：

TYPE=Ethernet
PROXY_METHOD=none
BROWSER_ONLY=no
BOOTPROTO=static   	# 设置为 静态
DEFROUTE=yes
IPV4_FAILURE_FATAL=no
IPV6INIT=yes
IPV6_AUTOCONF=yes
IPV6_DEFROUTE=yes
IPV6_FAILURE_FATAL=no
IPV6_ADDR_GEN_MODE=stable-privacy
NAME=ens33
UUID=1c71c1dc-e4f8-4594-b77e-5e04f6906a31
DEVICE=ens33
ONBOOT=yes			# 必须为 yes
IPADDR=192.168.131.137	# 静态 IP	
GATEWAY=192.168.2.2		# 网关
NETMAK=255.255.255.0	# 子网掩码
DNS1=8.8.8.8
DNS2=114.114.114.114

2、每一台虚拟机配置 ip 和域名映射：

# vim /etc/hosts
192.168.131.137 hadoop1
192.168.131.138 hadoop2
192.168.131.139 hadoop3

3、关闭防火墙：

# 关闭防火墙
Service iptables stop 

# 禁止防火墙开机自启
Chkconfig iptables off

# 关闭selinux，设置 SELINUX 参数为 disabled
Vim /etc/selinux/config

4、重启网络、检查：

# 重启网络
service network restart

# hadoop1/2/3 三台服务器之间相互 ping 看是否能够通
ping hadoop1	

# 重启
reboot

2.2 添加hadoop账户并分配sudo权限

useradd -m hadoop
passwd hadoop

# 分配 sudo 权限，在该行root ALL=(ALL) ALL下添加hadoop ALL=(ALL) ALL保存后退出，并切换回hadoop用户
visudo

# 如下配置
## The COMMANDS section may have other options added to it.
##
## Allow root to run any commands anywhere
root    ALL=(ALL)       ALL
hadoop  ALL=(ALL)       ALL

# 切换用户
su hadoop

三台服务器IP、用户名、密码对应

# 分别有两个用户：root（110139）、hadoop（hadoop）
192.168.131.137 hadoop1
192.168.131.138 hadoop2
192.168.131.139 hadoop3

2.3 实现三台服务器之间免密登录

1、安装 ssh 服务：

yum install -y openssl openssh-server
yum -y install openssh-clients			# 一定要安装客户端，不然使用不了 ssh

# 用vim打开配置文件/etc/ssh/sshd_config
# 将 PermitRootLogin，RSAAuthentication，PubkeyAuthentication 的注释设置打开

# 常用命令
service sshd restart # 重启SSH服务
service sshd start # 启动服务
service sshd stop  # 停止服务
netstat -antp | grep sshd 	# 查看是否启动22端口

chkconfig sshd on		# 设置开机启动
chkconfig sshd off	# 设置禁止开机启动

yum install 时出现：

[root@hadoop3 .ssh]# yum install -y openssl openssh-server
Loaded plugins: fastestmirror
Setting up Install Process
Loading mirror speeds from cached hostfile
YumRepo Error: All mirror URLs are not using ftp, http[s] or file.
 Eg. Invalid release/repo/arch combination/
removing mirrorlist with no valid mirrors: /var/cache/yum/x86_64/6/base/mirrorlist.txt
Error: Cannot find a valid baseurl for repo: base

解决方法：

sed -i "s|enabled=1|enabled=0|g" /etc/yum/pluginconf.d/fastestmirror.conf
mv /etc/yum.repos.d/CentOS-Base.repo /etc/yum.repos.d/CentOS-Base.repo.backup
curl -o /etc/yum.repos.d/CentOS-Base.repo https://www.xmpan.com/Centos-6-Vault-Aliyun.repo 
    
yum clean all
yum makecache
yum repolist		# 查看当前源

参考：YumRepo Error: All mirror URLs are not using ftp, http[s] or file.centos6 yum失败！

2、生成公钥私钥（一定要在 hadoop 用户下执行！！！，否则后续 start-dfs.sh 没有权限）

# 一路回车即可 路径：/home/hadoop/.ssh/id_rsa、id_rsa.pub
ssh-keygen -t rsa

# 在 hadoop1  服务器
cd /home/hadoop/.ssh/
cat id_rsa.pub > authorized_keys

# 再将 Hadoop2、Hadoop3 的 id_rsa.pub 的内容拷贝到 Hadoop1 的 authorized_keys，最后将 authorized_keys 分别传到 Hadoop2、Hadoop3 的 /home/hadoop/.ssh/
scp authorized_keys root@hadoop2:/home/hadoop/.ssh/
scp authorized_keys root@hadoop3:/home/hadoop/.ssh/
    
# 分别给三个节点 修改authorized_keys权限
chmod 600 authorized_keys
chmod 700 .ssh

# 测试，ssh + 主机名
ssh hadoop2

# hbase 集群
https://www.cnblogs.com/qingyunzong/p/8668880.html

# hadoop 集群
https://www.cnblogs.com/qingyunzong/p/8496127.html

3. jdk 安装

1、如果你使用root用户进行安装。 vi /etc/profile 即可系统变量

2、如果你使用普通用户进行安装。 vi ~/.bashrc 用户变量

# .bashrc

# Source global definitions
if [ -f /etc/bashrc ]; then
	. /etc/bashrc
fi

# User specific aliases and functions
export JAVA_HOME=/home/hadoop/apps/jdk1.8.0_261
export PATH=$JAVA_HOME/bin:$PATH

检查：

[hadoop@hadoop3 apps]$ source ~/.bashrc
[hadoop@hadoop3 apps]$ echo $JAVA_HOME
/home/hadoop/apps/jdk1.8.0_261
[hadoop@hadoop3 apps]$ java -version
java version "1.8.0_261"
Java(TM) SE Runtime Environment (build 1.8.0_261-b12)
Java HotSpot(TM) 64-Bit Server VM (build 25.261-b12, mixed mode)

4. hadoop 安装

集群规划：

服务/主机名	IP	用户	HDFS	YARN
hadoop1	192.168.131.137	hadoop	NameNode,DataNode	NodeManager,ResourceManager
hadoop2	192.168.131.138	hadoop	DataNode	NodeManager
hadoop3	192.168.131.139	hadoop	DataNode	NodeManager

服务	主节点	从节点
HDFS	NameNode	DataNode
YARN	ResourceManager	NodeManager

注意：hadoop 用户最好给 apps、data、hadoop-2.7.5/ 等都添加权限 sudo chown hadoop:hadoop apps/

1、解压：

[hadoop@hadoop1 apps]$ tar -zxvf hadoop-2.7.5.tar.gz

2、hadoop-env.sh：

配置文件目录：/home/hadoop/apps/hadoop-2.7.5/etc/hadoop

cd /home/hadoop/apps/hadoop-2.7.5/etc/hadoop
vim hadoop-env.sh

# 只需修改 jdk 路径
export JAVA_HOME=/usr/local/jdk1.8.0_73

3、core-site.xml：

<configuration>
        <property>
                <name>fs.defaultFS</name>
                <value>hdfs://hadoop1:9000</value>
        </property>
        <property>
                <name>hadoop.tmp.dir</name>
                <value>/home/hadoop/data/hadoopdata</value>
        </property>
</configuration>

4、hdfs-site.xml：

<configuration>
        <property>
                <name>dfs.namenode.name.dir</name>
                <value>/home/hadoop/data/hadoopdata/name</value>
                <description>为了保证元数据的安全一般配置多个不同目录</description>
        </property>

        <property>
                <name>dfs.datanode.data.dir</name>
                <value>/home/hadoop/data/hadoopdata/data</value>
                <description>datanode 的数据存储目录</description>
        </property>

        <property>
                <name>dfs.replication</name>
                <value>2</value>
                <description>HDFS 的数据块的副本存储个数, 默认是3</description>
        </property>

        <property>
                <name>dfs.secondary.http.address</name>
                <value>hadoop3:50090</value>
                <description>secondarynamenode 运行节点的信息，和 namenode 不同节点</description>
        </property>
</configuration>

5、mapred-site.xml：

<configuration>
        <property>
                <name>mapreduce.framework.name</name>
                <value>yarn</value>
        </property>
</configuration>

6、yarn-site.xml：

<configuration>

<!-- Site specific YARN configuration properties -->

        <property>
                <name>yarn.resourcemanager.hostname</name>
                <value>hadoop1</value>			// 这个必须与 resourcemanager 服务器一致，我这里是 hadoop1
        </property>
        
        <property>
                <name>yarn.nodemanager.aux-services</name>
                <value>mapreduce_shuffle</value>
                <description>YARN 集群为 MapReduce 程序提供的 shuffle 服务</description>
        </property>

</configuration>

7、slaves：

vim slaves 

hadoop1
hadoop2
hadoop3

8、将 hadoop-2.7.5/ 分发到其他集群：

scp -r hadoop-2.7.5/ hadoop2:/home/hadoop/apps
scp -r hadoop-2.7.5/ hadoop3:/home/hadoop/apps

9、添加环境变量（每台服务器都需要添加）：

[hadoop@hadoop1 apps]$ vim ~/.bashrc 
export HADOOP_HOME=/home/hadoop/apps/hadoop-2.7.5
export PATH=$PATH:$HADOOP_HOME/bin:$HADOOP_HOME/sbin:
            
[hadoop@hadoop1 apps]$ source ~/.bashrc 
[hadoop@hadoop1 apps]$ hadoop version
Hadoop 2.7.5
Subversion https://shv@git-wip-us.apache.org/repos/asf/hadoop.git -r 18065c2b6806ed4aa6a3187d77cbe21bb3dba075
Compiled by kshvachk on 2017-12-16T01:06Z
Compiled with protoc 2.5.0
From source with checksum 9f118f95f47043332d51891e37f736e9
This command was run using /home/hadoop/apps/hadoop-2.7.5/share/hadoop/common/hadoop-common-2.7.5.jar

10、格式化：

[hadoop@hadoop1 bin]$ hadoop namenode -format

21/06/13 14:34:26 INFO util.GSet: Computing capacity for map NameNodeRetryCache
21/06/13 14:34:26 INFO util.GSet: VM type       = 64-bit
21/06/13 14:34:26 INFO util.GSet: 0.029999999329447746% max memory 966.7 MB = 297.0 KB
21/06/13 14:34:26 INFO util.GSet: capacity      = 2^15 = 32768 entries
21/06/13 14:34:27 INFO namenode.FSImage: Allocated new BlockPoolId: BP-57205935-192.168.131.137-1623566067094
21/06/13 14:34:27 INFO common.Storage: Storage directory /home/hadoop/data/hadoopdata/name has been successfully formatted.
21/06/13 14:34:27 INFO namenode.FSImageFormatProtobuf: Saving image file /home/hadoop/data/hadoopdata/name/current/fsimage.ckpt_0000000000000000000 using no compression
21/06/13 14:34:27 INFO namenode.FSImageFormatProtobuf: Image file /home/hadoop/data/hadoopdata/name/current/fsimage.ckpt_0000000000000000000 of size 322 bytes saved in 0 seconds.
21/06/13 14:34:28 INFO namenode.NNStorageRetentionManager: Going to retain 1 images with txid >= 0
21/06/13 14:34:28 INFO util.ExitUtil: Exiting with status 0
21/06/13 14:34:28 INFO namenode.NameNode: SHUTDOWN_MSG: 
/************************************************************
SHUTDOWN_MSG: Shutting down NameNode at hadoop1/192.168.131.137
************************************************************/

格式化 Namenode 时不能创建目录

java.io.IOException: Cannot create directory /home/hadoop/data/hadoopdata/na
    
# 解决方法
https://blog.csdn.net/qq_40414738/article/details/99544777

start-dfs.sh 时没有权限

[hadoop@hadoop1 sbin]$ sudo start-dfs.sh
[sudo] password for hadoop: 
sudo: start-dfs.sh: command not found
[hadoop@hadoop1 sbin]$ start-dfs.sh
Starting namenodes on [hadoop1]
hadoop@hadoop1's password: 
hadoop1: namenode running as process 4090. Stop it first.
hadoop@hadoop2's password: hadoop@hadoop1's password: hadoop@hadoop3's password: 
hadoop2: mkdir: cannot create directory `/home/hadoop/apps/hadoop-2.7.5/logs': Permission denied
hadoop2: chown: cannot access `/home/hadoop/apps/hadoop-2.7.5/logs': No such file or directory
hadoop2: starting datanode, logging to /home/hadoop/apps/hadoop-2.7.5/logs/hadoop-hadoop-datanode-hadoop2.out
hadoop2: /home/hadoop/apps/hadoop-2.7.5/sbin/hadoop-daemon.sh: line 159: /home/hadoop/apps/hadoop-2.7.5/logs/hadoop-hadoop-datanode-hadoop2.out: No such file or directory
hadoop2: head: cannot open `/home/hadoop/apps/hadoop-2.7.5/logs/hadoop-hadoop-datanode-hadoop2.out' for reading: No such file or directory
hadoop2: /home/hadoop/apps/hadoop-2.7.5/sbin/hadoop-daemon.sh: line 177: /home/hadoop/apps/hadoop-2.7.5/logs/hadoop-hadoop-datanode-hadoop2.out: No such file or directory
hadoop2: /home/hadoop/apps/hadoop-2.7.5/sbin/hadoop-daemon.sh: line 178: /home/hadoop/apps/hadoop-2.7.5/logs/hadoop-hadoop-datanode-hadoop2.out: No such file or directory

解决办法：https://www.cnblogs.com/zknublx/p/8066693.html

开启 hdfs

[hadoop@hadoop1 sbin]$ start-dfs.sh 
Starting namenodes on [hadoop1]
hadoop1: starting namenode, logging to /home/hadoop/apps/hadoop-2.7.5/logs/hadoop-hadoop-namenode-hadoop1.out
hadoop3: starting datanode, logging to /home/hadoop/apps/hadoop-2.7.5/logs/hadoop-hadoop-datanode-hadoop3.out
hadoop2: starting datanode, logging to /home/hadoop/apps/hadoop-2.7.5/logs/hadoop-hadoop-datanode-hadoop2.out
hadoop1: datanode running as process 4822. Stop it first.
Starting secondary namenodes [hadoop3]
hadoop3: starting secondarynamenode, logging to /home/hadoop/apps/hadoop-2.7.5/logs/hadoop-hadoop-secondarynamenode-hadoop3.out
[hadoop@hadoop1 sbin]$ jps
4822 DataNode
5465 NameNode
5724 Jps

开启 yarn

[hadoop@hadoop1 sbin]$ start-yarn.sh
starting yarn daemons
starting resourcemanager, logging to /home/hadoop/apps/hadoop-2.7.5/logs/yarn-hadoop-resourcemanager-hadoop1.out
hadoop3: starting nodemanager, logging to /home/hadoop/apps/hadoop-2.7.5/logs/yarn-hadoop-nodemanager-hadoop3.out
hadoop2: starting nodemanager, logging to /home/hadoop/apps/hadoop-2.7.5/logs/yarn-hadoop-nodemanager-hadoop2.out
hadoop1: starting nodemanager, logging to /home/hadoop/apps/hadoop-2.7.5/logs/yarn-hadoop-nodemanager-hadoop1.out

查看集群 jps 进程

[hadoop@hadoop1 sbin]$ jps
5891 Jps
4822 DataNode
5465 NameNode
5770 ResourceManager
5870 NodeManager

[hadoop@hadoop2 ~]$ jps
2643 NodeManager
2532 DataNode
2717 Jps

[hadoop@hadoop3 ~]$ jps
2816 DataNode
2994 NodeManager
3091 Jps

web 管理界面

HDFS : http://192.168.131.137:50070
YARN ： http://192.168.131.137:8088/cluster

https://www.cnblogs.com/qingyunzong/p/8496127.html
https://www.cnblogs.com/qingyunzong/p/8668880.html
https://blog.csdn.net/Beans___Lee/article/details/105265244
https://blog.csdn.net/qq_43106863/article/details/101530233
https://blog.csdn.net/ASN_forever/article/details/80701003
https://blog.csdn.net/weixin_41552767/article/details/107221454
https://segmentfault.com/a/1190000038229319
VMware 搭建 hadoop 完全分布式集群

5. zookeeper 集群安装

这里规划三台服务器：hadoop1,hadoop2,hadoop3

下载地址：http://mirrors.hust.edu.cn/apache/ZooKeeper/

此处使用的是3.4.10版本

1、解压：

[hadoop@hadoop1 ~]$ cd /home/hadoop/apps
[hadoop@hadoop1 ~]$ tar -zxvf zookeeper-3.4.10.tar.gz -C apps/

2、修改配置文件：

[hadoop@hadoop1 zookeeper-3.4.10]$ cd conf/
[hadoop@hadoop1 conf]$ mv zoo_sample.cfg zoo.cfg
[hadoop@hadoop1 conf]$ vim zoo.cfg 

# dataDir  内存数据库快照存放地址
dataDir=/home/hadoop/data/zkdata/

# 末尾添加，配置ZK监听客户端连接的端口
dataLogDir=/home/hadoop/log/zklog		# 事务日志

server.1=hadoop1:2888:3888
server.2=hadoop2:2888:3888
server.3=hadoop3:2888:3888
        
# server.serverid=host:tickpot:electionport   
# server：固定写法
# serverid：每个服务器的指定ID（必须处于1-255之间，必须每一台机器不能重复）
# host：主机名
# tickpot：心跳通信端口
# electionport：选举端口

3、将配置文件分发到集群其他机器中：

[hadoop@hadoop1 bin]$ scp -r zookeeper-3.4.10/ hadoop2:/home/hadoop/apps
[hadoop@hadoop1 bin]$ scp -r zookeeper-3.4.10/ hadoop3:/home/hadoop/apps

注意：以下操作 4、5、6 每台服务器都需要这样操作！

4、创建日志目录、配置服务器 ID：

[hadoop@hadoop1 bin]$ mkdir -p /home/hadoop/log
[hadoop@hadoop1 bin]$ sudo chown hadoop:hadoop log/			# 给 hadoop 用户添加权限
[hadoop@hadoop1 bin]$ mkdir -p /home/hadoop/data/zkdata		
[hadoop@hadoop1 ~]$ cd /home/hadoop/data/zkdata
[hadoop@hadoop1 ~]$ echo 1 > myid		# myid,里面存放的内容就是服务器的 id,就是 server.1=hadoop01:2888:3888 当中的 id, 就是 1

5、配置环境变量：

vim ~/.bashrc

export ZOOKEEPER_HOME=/home/hadoop/apps/zookeeper-3.4.10
export PATH=$PATH:$ZOOKEEPER_HOME/bin

source ~/.bashrc

6、启动 zkserver 服务，检查：

启动：zkServer.sh start
停止：zkServer.sh stop
查看状态：zkServer.sh status

ps -aux | grep 'zookeeper'

[hadoop@hadoop1 bin]$ zkServer.sh start
ZooKeeper JMX enabled by default
Using config: /home/hadoop/apps/zookeeper-3.4.10/bin/../conf/zoo.cfg
Starting zookeeper ... STARTED
[hadoop@hadoop1 bin]$ zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /home/hadoop/apps/zookeeper-3.4.10/bin/../conf/zoo.cfg
Error contacting service. It is probably not running.
[hadoop@hadoop1 bin]$ jps
2769 Jps
2723 QuorumPeerMain

[hadoop@hadoop2 bin]$ zkServer.sh start
ZooKeeper JMX enabled by default
Using config: /home/hadoop/apps/zookeeper-3.4.10/bin/../conf/zoo.cfg
Starting zookeeper ... STARTED
[hadoop@hadoop2 bin]$ zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /home/hadoop/apps/zookeeper-3.4.10/bin/../conf/zoo.cfg
Mode: leader
[hadoop@hadoop2 bin]$ jps

[hadoop@hadoop3 bin]$ zkServer.sh start
ZooKeeper JMX enabled by default
Using config: /home/hadoop/apps/zookeeper-3.4.10/bin/../conf/zoo.cfg
Starting zookeeper ... STARTED
[hadoop@hadoop3 bin]$ zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /home/hadoop/apps/zookeeper-3.4.10/bin/../conf/zoo.cfg
Mode: follower
[hadoop@hadoop3 bin]$ jps
1898 QuorumPeerMain
1951 Jps

参考：https://www.cnblogs.com/qingyunzong/p/8619184.html

6. HBase 集群搭建

[hadoop@hadoop1 apps]$ tar -zxvf hbase-1.2.6-bin.tar.gz
[hadoop@hadoop1 apps]$ cd hbase-1.2.6/conf/				# 配置文件目录

[hadoop@hadoop1 conf]$ vim hbase-env.sh 

# 修改 jdk 路径 和 zookeeper 
export JAVA_HOME=/home/hadoop/apps/jdk1.8.0_261

# Tell HBase whether it should manage it's own instance of Zookeeper or not.
export HBASE_MANAGES_ZK=false

1、修改 hbase-site.xml：

<configuration>
        <property>
                <!-- 指定 hbase 在 HDFS 上存储的路径 -->
                <name>hbase.rootdir</name>
                <value>hdfs://hadoop1:9000/hbase</value>
        </property>
        <property>
                <!-- 指定 hbase 是分布式的 -->
                <name>hbase.cluster.distributed</name>
                <value>true</value>
        </property>
        <property>
                <!-- 指定 zk 的地址，多个用“,”分割 -->
                <name>hbase.zookeeper.quorum</name>
                <value>hadoop1:2181,hadoop2:2181,hadoop3:2181</value>
        </property>
        <property>
            <name>hbase.master.maxclockskew</name>
            <value>180000</value>
            <description>Time difference of regionserver from master</description>
       </property>
</configuration>

2、修改 regionservers ：

vim regionservers

hadoop1
hadoop2
hadoop3

3、修改 backup-masters（该文件是不存在的，先自行创建）：

# 备用 master
vim backup-masters

hadoop3

4、要把 hadoop 的 hdfs-site.xml 和 core-site.xml 放到 hbase-1.2.6/conf 下

5、删除 docs（删除 HBase 目录下的 docs 文件夹）：

rm -rf docs

# 分发
scp -r hbase-1.2.6/ hadoop2:/home/hadoop/apps/
scp -r hbase-1.2.6/ hadoop3:/home/hadoop/apps/

6、时间同步：

HBase 集群对于时间的同步要求的比 HDFS 严格，所以，集群启动之前千万记住要进行时间同步，要求相差不要超过 30s：

方法一：

# 配置 hbase-site.xml，三个服务器之间时间差不超过 30s

<property>
    <name>hbase.master.maxclockskew</name>
    <value>180000</value>
    <description>Time difference of regionserver from master</description>
</property>

方法二：

从节点服务器与主节点服务器时间同步
所有节点与网络服务器时间同步

[hadoop@hadoop2 hbase-1.2.6]$ sudo yum install -y ntp

# 与 time.nist.gov 网络服务器同步
[hadoop@hadoop2 hbase-1.2.6]$ sudo /usr/sbin/ntpdate time.nist.gov
15 Jun 20:54:27 ntpdate[1780]: step time server 132.163.97.3 offset -28785.531070 sec
            
# 设置定时任务
[hadoop@hadoop2 hbase-1.2.6]$ crontab -l
# 服务器时间同步
0 */1 * * *  /usr/sbin/ntpdate time.nist.gov

参考：http://www.hnbian.cn/posts/f2bc4737.html

7、系统环境变量：

# 所有节点都需要修改
[hadoop@hadoop1 bin]$ vim ~/.bashrc

export HBASE_HOME=/home/hadoop/apps/hbase-1.2.6
export PATH=$PATH:$HBASE_HOME/bin
    
source ~/.bashrc

8、启动 hbase 集群：

[hadoop@hadoop1 bin]$ pwd
/home/hadoop/apps/hbase-1.2.6/bin
[hadoop@hadoop1 bin]$ start-hbase.sh
starting master, logging to /home/hadoop/apps/hbase-1.2.6/logs/hbase-hadoop-master-hadoop1.out
Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0
Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0
hadoop3: starting regionserver, logging to /home/hadoop/apps/hbase-1.2.6/logs/hbase-hadoop-regionserver-hadoop3.out
hadoop2: starting regionserver, logging to /home/hadoop/apps/hbase-1.2.6/logs/hbase-hadoop-regionserver-hadoop2.out
hadoop1: starting regionserver, logging to /home/hadoop/apps/hbase-1.2.6/logs/hbase-hadoop-regionserver-hadoop1.out
hadoop3: starting master, logging to /home/hadoop/apps/hbase-1.2.6/logs/hbase-hadoop-master-hadoop3.out

注意：在哪个集群启动，哪个就是主节点 Master，启动 hbase 之前一定先启动 zookeeper、hadoop！！！

9、检查：

[hadoop@hadoop1 bin]$ jps
1668 NameNode
2164 NodeManager
4054 Jps
3590 HMaster
1495 QuorumPeerMain
3719 HRegionServer
1768 DataNode
2061 ResourceManager

[hadoop@hadoop2 hbase-1.2.6]$ jps
1440 QuorumPeerMain
2305 Jps
2149 HRegionServer
1622 NodeManager
1511 DataNode

[hadoop@hadoop3 bin]$ jps
2722 Jps
1682 NodeManager
1436 QuorumPeerMain
1501 DataNode
2654 HRegionServer

若有节点 HRegionServer 进程没起来就手动启动

# 启动 master 进程 
hbase-daemon.sh start master

# 启动HRegionServer进程
hbase-daemon.sh start regionserver

10、web-ui：

http://192.168.131.137:16010/

11、测试：

[hadoop@hadoop1 bin]$ hbase shell
SLF4J: Class path contains multiple SLF4J bindings.
SLF4J: Found binding in [jar:file:/home/hadoop/apps/hbase-1.2.6/lib/slf4j-log4j12-1.7.5.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: Found binding in [jar:file:/home/hadoop/apps/hadoop-2.7.5/share/hadoop/common/lib/slf4j-log4j12-1.7.10.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
SLF4J: Actual binding is of type [org.slf4j.impl.Log4jLoggerFactory]
HBase Shell; enter 'help<RETURN>' for list of supported commands.
Type "exit<RETURN>" to leave the HBase Shell
Version 1.2.6, rUnknown, Mon May 29 02:25:32 CDT 2017

hbase(main):001:0> list
TABLE                                                                                                               
0 row(s) in 0.6420 seconds

=> []
hbase(main):002:0> helo 'list'
NoMethodError: undefined method `helo' for #<Object:0x3ab35b9c>


hbase(main):004:0> create 't1', {NAME => 'f1'}, {NAME => 'f2'}, {NAME => 'f3'}
0 row(s) in 1.6840 seconds

=> Hbase::Table - t1
hbase(main):005:0> list
TABLE                                                                                                               
t1                                                                                                                  
1 row(s) in 0.0290 seconds

=> ["t1"]
hbase(main):006:0> desc 't1'
Table t1 is ENABLED                                                                                                 
t1                                                                                                                  
COLUMN FAMILIES DESCRIPTION                                                                                         
{NAME => 'f1', BLOOMFILTER => 'ROW', VERSIONS => '1', IN_MEMORY => 'false', KEEP_DELETED_CELLS => 'FALSE', DATA_BLOC
K_ENCODING => 'NONE', TTL => 'FOREVER', COMPRESSION => 'NONE', MIN_VERSIONS => '0', BLOCKCACHE => 'true', BLOCKSIZE 
=> '65536', REPLICATION_SCOPE => '0'}                                                                               
{NAME => 'f2', BLOOMFILTER => 'ROW', VERSIONS => '1', IN_MEMORY => 'false', KEEP_DELETED_CELLS => 'FALSE', DATA_BLOC
K_ENCODING => 'NONE', TTL => 'FOREVER', COMPRESSION => 'NONE', MIN_VERSIONS => '0', BLOCKCACHE => 'true', BLOCKSIZE 
=> '65536', REPLICATION_SCOPE => '0'}                                                                               
{NAME => 'f3', BLOOMFILTER => 'ROW', VERSIONS => '1', IN_MEMORY => 'false', KEEP_DELETED_CELLS => 'FALSE', DATA_BLOC
K_ENCODING => 'NONE', TTL => 'FOREVER', COMPRESSION => 'NONE', MIN_VERSIONS => '0', BLOCKCACHE => 'true', BLOCKSIZE 
=> '65536', REPLICATION_SCOPE => '0'}                                                                               
3 row(s) in 0.3510 seconds

参考文章：

https://www.cnblogs.com/clsn/p/10300487.html#63-%E5%90%AF%E5%8A%A8hbase%E9%9B%86%E7%BE%A4
https://www.cnblogs.com/qingyunzong/p/8668880.html#_label2_0
https://blog.csdn.net/u011066470/article/details/106305238
https://www.cnblogs.com/ityouknow/p/7343996.html