Cnetos7.5离线安装CDH5.15.1

1.环境准备

1.1 系统环境

1.1.1 基础配置信息

基础配置信息	配置
PC机1	Intel I7-8700 + 16G内存
PC机2	Intel xenon E5-1603 + 8G内存
虚拟机VMware Workstation Pro	版本15.0

1.1.2 虚拟化服务器配置

服务器名称	IP地址	配置	角色
master	192.168.3.181	操作系统:centos7.5,CPU*4,内存:2G
slave183	192.168.3.183	操作系统:centos7.5,CPU*4,内存:2G
slave184	192.168.3.184	操作系统:centos7.5,CPU*4,内存:2G
slave185	192.168.3.185	操作系统:centos7.5,CPU*4,内存:2G
slave186	192.168.3.186	操作系统:centos7.5,CPU*6,内存:4G
slave187	192.168.3.187	操作系统:centos7.5,CPU*6,内存:4G
slave188	192.168.3.188	操作系统:centos7.5,CPU*6,内存:4G
slave189	192.168.3.189	操作系统:centos7.5,CPU*6,内存:4G

1.2 安装JDK

JDK版本:

[root@localhost /]# java -version
java version "1.8.0_161"
Java(TM) SE Runtime Environment (build 1.8.0_161-b12)
Java HotSpot(TM) 64-Bit Server VM (build 25.161-b12, mixed mode)

1.3 修改静态IP,主机名,配置HOSTS

以改主机名为master,IP为192.168.3.181为例

1.3.1 修改静态IP:

使用ifconfig命令查看本机网卡信息.
如果提示ifconfig没有安装,则使用 yum install -y net-tools安装即可.
若此时虚拟机无法联网,也可使用ip addr命令查看网卡信息

[root@localhost /]# ifconfig
ens33: flags=4163  mtu 1500
        inet 192.168.3.181  netmask 255.255.255.0  broadcast 192.168.3.255
        inet6 fe80::9cfa:d47a:8c84:cdb6  prefixlen 64  scopeid 0x20
        inet6 fe80::26b1:3a23:6c81:de93  prefixlen 64  scopeid 0x20
        inet6 fe80::4756:77ae:310c:febe  prefixlen 64  scopeid 0x20
        ether 00:0c:29:86:b9:5d  txqueuelen 1000  (Ethernet)
        RX packets 18639  bytes 1521733 (1.4 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 880  bytes 71977 (70.2 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

lo: flags=73  mtu 65536
        inet 127.0.0.1  netmask 255.0.0.0
        inet6 ::1  prefixlen 128  scopeid 0x10
        loop  txqueuelen 1000  (Local Loopback)
        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 0

从打印的网卡信息中看到我们的网卡名为:ens33,现在修改网络配置.

使用 vi /etc/sysconfig/network-scripts/ifcfg-ens33打开网卡配置文件

改为对应地址然后保存.

使用service network restart或者systemctl restart network重启网络服务,然后再使用ifconfig查看本机网络配置是否正确.

对slave系列服务器做对应修改,使得IP改为192.168.3.183-192.168.3.189

1.3.2 修改主机名:

使用:vi /etc/hostname修改主机名称,这里我们将主机名改为master.
改完配置文件之后需要重启生效,这里我们再执行:hostname master,这样可以达到不启动也修改了hostname,重启后更改也生效的效果.
执行: exit,退出重新登陆,就会发现[root@localhost ~]#变成了[root@master ~]#

对slave系列服务器做对应修改,使得hostName改为slave183-slave189

1.3.3 配置HOSTS:

对每台服务其配置HOSTS:

192.168.3.181 master
192.168.3.183 slave183
192.168.3.184 slave184
192.168.3.185 slave185
192.168.3.186 slave186
192.168.3.187 slave187
192.168.3.188 slave188
192.168.3.189 slave189

1.4 关闭防火墙

systemctl status firewalld 查看防火墙状态
systemctl stop firewalld 关闭防火墙
systemctl disable firewalld 禁止防火墙开机启动
systemctl status firewalld 启动防火墙

使用 systemctl stop firewalld并使用 systemctl disable firewalld禁止防火墙.

1.5 SSH配置免密登陆

这一步master和slave节点操作略有区别.

• 首先操作master:

1. ssh-keygen -t rsa #一路回车到完成.会在~/.ssh下面生成2个文件 id_rsa，id_rsa.pub

[root@master ~]# ssh-keygen -t rsa
Generating public/private rsa key pair.
Enter file in which to save the key (/root/.ssh/id_rsa): 
Created directory '/root/.ssh'.
Enter passphrase (empty for no passphrase): 
Enter same passphrase again: 
Your identification has been saved in /root/.ssh/id_rsa.
Your public key has been saved in /root/.ssh/id_rsa.pub.
The key fingerprint is:
SHA256:ET9Wd//uqIcVwnLzLRbTBdIGkO2h5TQuh6P3eItOeZI root@master
The key's randomart image is:
+---[RSA 2048]----+
|        . .++oo..|
|         o..*oo.o|
|        . +X + .o|
|         o*.X + o|
|        S. * + =.|
|        . .o  =..|
|         .Eo.+ ..|
|         ..++ .o |
|         .o.o+. .|
+----[SHA256]-----+
[root@master ~]# 

2. ssh-copy-id -i ~/.ssh/id_rsa.pub root@master #将公钥拷贝到本机的authorized_keys上

• 操作slave节点(在其它节点分别执行 ):

1. ssh-keygen -t rsa #一路回车到完成.会在~/.ssh下面生成2个文件 id_rsa，id_rsa.pub
2. ssh-copy-id -i ~/.ssh/id_rsa.pub root@master #注意此处不变，将公钥拷贝到master的authorized_keys上.

操作完成后,各slave节点即可免密码登陆master.登陆测试:

ssh master

[root@slave183 ~]# ssh master
Last login: Thu Oct 11 09:47:35 2018 from slave183
[root@master ~]# 

成功免密登陆master,在所有slave节点都进行以上测试,校验免密登陆是否成功.

• 继续操作master,在master节点上,将authorized_keys分发到其他节点服务器：

scp ~/.ssh/authorized_keys root@slave183:~/.ssh/
…
scp ~/.ssh/authorized_keys root@slave189:~/.ssh/

[root@master ~]# scp ~/.ssh/authorized_keys root@slave183:~/.ssh/
root@slave183's password: 
authorized_keys                               100% 3158     1.9MB/s   00:00    

操作完成后,master即可免密码登陆各slave节点,slave节点亦可免密登陆其它slave节点.
master登陆节点测试:

ssh slave183

[root@master ~]# ssh slave183
Last login: Thu Oct 11 09:48:09 2018 from master
[root@slave183 ~]# exit
登出
Connection to slave183 closed.

slave节点互相登陆测试:

ssh slave184

[root@slave183 ~]# ssh slave184
Last login: Thu Oct 11 10:19:55 2018 from slave183
[root@slave184 ~]# exit
登出
Connection to slave184 closed.

1.6 配置时间同步服务NTP

集群中所有主机必须保持时间同步，如果时间相差较大会引起各种问题。 具体思路如下：

1. master节点作为ntp服务器与外界对时中心同步时间，随后对所有slave节点提供时间同步服务。
2. 所有slave节点以master节点为基础同步时间。

• 安装NTP服务,所有节点执行：

yum install -y ntp
yum install -y ntpdate #ntpdate会作为依赖服务在ntp服务安装的时候被安装.因此安装完ntp之后这里不执行也是可以的.

• 在启动ntp服务之前先进行一次时间同步:

ntpdate -u s2c.time.edu.cn

[root@master ~]# ntpdate -u s2c.time.edu.cn
14 Oct 17:36:21 ntpdate[12912]: adjust time server 202.112.10.36 offset -0.000184 sec

• 设置ntp服务自启:

禁用chronyd,centos7.5默认使用chronyd作为时间同步服务,不禁用的话ntp服务无法自启.
systemctl disable chronyd #禁用chronyd
设置ntp服务自启:
systemctl enable ntpd #ntp自启

1.6.1 修改master节点ntp配置

使用s2c.time.edu.cn作为NTP服务器,加入到配置文件中

• 打开配置文件: vi /etc/ntp.conf,将server s2c.time.edu.cn prefer加入到配置文件中,并且注释原来的NTP服务器:

# Use public servers from the pool.ntp.org project.
# Please consider joining the pool (http://www.pool.ntp.org/join.html).
server  s2c.time.edu.cn prefer

#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

• 配置完成后重启下master,重启完成查看ntp服务是否自启,并查看ntp同步状态:

使用systemctl status ntpd 查看ntp是否启动.

[root@master ~]# systemctl status ntpd
● ntpd.service - Network Time Service
   Loaded: loaded (/usr/lib/systemd/system/ntpd.service; enabled; vendor preset: disabled)
   Active: active (running) since 四 2018-10-11 11:28:16 CST; 54s ago
  Process: 688 ExecStart=/usr/sbin/ntpd -u ntp:ntp $OPTIONS (code=exited, status=0/SUCCESS)
 Main PID: 696 (ntpd)
   CGroup: /system.slice/ntpd.service
           └─696 /usr/sbin/ntpd -u ntp:ntp -g

10月 11 11:28:16 master ntpd[696]: Listening on routing socket on fd #20 f...es
10月 11 11:28:16 master systemd[1]: Started Network Time Service.
10月 11 11:28:16 master ntpd[696]: Deferring DNS for s2c.time.edu.cn 1
10月 11 11:28:16 master ntpd[696]: 0.0.0.0 c016 06 restart
10月 11 11:28:16 master ntpd[701]: signal_no_reset: signal 17 had flags 4000000
10月 11 11:28:16 master ntpd[696]: 0.0.0.0 c012 02 freq_set kernel 0.000 PPM
10月 11 11:28:16 master ntpd[696]: 0.0.0.0 c011 01 freq_not_set
10月 11 11:28:18 master ntpd_intres[701]: DNS s2c.time.edu.cn -> 202.112.10.36
10月 11 11:28:19 master ntpd[696]: Listen normally on 4 ens33 192.168.3.18...23
10月 11 11:28:19 master ntpd[696]: new interface(s) found: waking up resolver
Hint: Some lines were ellipsized, use -l to show in full.

使用ntpstat 查看ntp是否启动(启动后需要等候几分钟才会正常同步).

1.6.2修改所有slave节点配置

• 修改所有slave节点的配置文件
  打开配置文件: vi /etc/ntp.conf,将server master加入到配置文件中,并且注释原来的NTP服务器:

# the administrative functions.
restrict 127.0.0.1
restrict ::1

# Hosts on local network are less restricted.
#restrict 192.168.1.0 mask 255.255.255.0 nomodify notrap

# Use public servers from the pool.ntp.org project.
# Please consider joining the pool (http://www.pool.ntp.org/join.html).

server master
#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

• 配置完成后重启各节点,重启完成查看ntp服务是否自启,并查看ntp同步状态.

[root@slave184 ~]# ntpstat 
synchronised to NTP server (192.168.3.181) at stratum 6 
   time correct to within 1201 ms
   polling server every 64 s

至此,NTP服务配置完成.

1.7 安装Mysql

只需要在master下安装mysql,其他节点不需要安装.安装mysql过程在此省略,可参考: Centos7.5安装mysql5.7.23-二进制包方式安装-完美安装,一步到位.

1.7.1 创建数据库

Service	Database	User
hbase	hbase	hbase
Activity Monitor	amon	amon
Reports Manager	rman	rman
Hue	hue	hue
Hive Metastore Server	metastore	hive
Sentry Server	sentry	sentry
Cloudera Navigator Audit Server	nav	nav
Cloudera Navigator Metadata Server	navms	navms
Oozie	oozie	oozie
hive	hive	hive

#使用下面局域创建表并创建用户
CREATE DATABASE <database> DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON <database>.* TO ''@'%' IDENTIFIED BY '<password>';


#以下是具体创建语句
CREATE DATABASE hbase DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON hbase.* TO 'hbase'@'%' IDENTIFIED BY 'root';
CREATE DATABASE amon DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON amon.* TO 'amon'@'%' IDENTIFIED BY 'root';
CREATE DATABASE rman DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON rman.* TO 'rman'@'%' IDENTIFIED BY 'root';
CREATE DATABASE hue DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON hue.* TO 'hue'@'%' IDENTIFIED BY 'root';
CREATE DATABASE sentry DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON sentry.* TO 'sentry'@'%' IDENTIFIED BY 'root';
CREATE DATABASE metastore DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON metastore.* TO 'hive'@'%' IDENTIFIED BY 'root';
CREATE DATABASE nav DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON nav.* TO 'nav'@'%' IDENTIFIED BY 'root';
CREATE DATABASE navms DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON navms.* TO 'navms'@'%' IDENTIFIED BY 'root';
CREATE DATABASE oozie DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON oozie.* TO 'oozie'@'%' IDENTIFIED BY 'root';
CREATE DATABASE hive DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON hive.* TO 'hive'@'%' IDENTIFIED BY 'root';

2.安装Cloudera Manager

所需文件	解压放置位置
cloudera-manager-centos7-cm5.15.1_x86_64.tar.gz	/opt/

2.1下载所需文件

下载地址: CM5.15.1下载

版本选择: 我们使用centos7.5,因此选择centos7版本下载.

2.2 上传,解压文件

使用文件上传工具将下载的cloudera-manager-centos7-cm5.15.1_x86_64.tar.gz上传至master节点,并解压至/opt(这里我将文件上传至/usr/local/share/目录下面).

[root@master ~]# tar zxvf /usr/local/share/cloudera-manager-centos7-cm5.15.1_x86_64.tar.gz -C /opt/
[root@master opt]# ls
cloudera  cm-5.15.1

2.3 修改agent配置文件

修改 /opt/cm-5.0.0/etc/cloudera-scm-agent/config.ini 下面的 serer_host:

[root@master opt]# vi /opt/cm-5.15.1/etc/cloudera-scm-agent/config.ini 

#修改内容:
server_host=master

2.4创建scm数据库

• 将mysql驱动复制到/opt/cm-5.15.1/share/cmf/lib/

[root@master opt]# cp /usr/local/share/mysql-connector-java-5.1.47-bin.jar  /opt/cm-5.15.1/share/cmf/lib/mysql-connector-java.jar

• 执行建表脚本

[root@master opt]# /opt/cm-5.15.1/share/cmf/schema/scm_prepare_database.sh mysql cm -hmaster -uroot -proot --scm-host master  scm scm scm

格式是:scm_prepare_database.sh 数据库类型 数据库 服务器 用户名 密码
–scm-host Cloudera_Manager_Server所在的机器，后面那三个不知道代表什么，直接照抄官网的了。

2.5 复制Agents端到各个slave节点

[root@master ~]# scp -r /opt/cm-5.15.1 slave183:/opt/
......
[root@master ~]# scp -r /opt/cm-5.15.1 slave189:/opt/

2.6 创建cloudera-scm用户（所有节点）

useradd --system --home=/opt/cm-5.15.1/run/cloudera-scm-server --shell=/bin/false --comment "Cloudera SCM User" cloudera-scm

[root@master opt]# useradd --system --home=/opt/cm-5.15.1/run/cloudera-scm-server --shell=/bin/false --comment "Cloudera SCM User" cloudera-scm

2.7 更改文件夹所属用户

• master节点:

[root@master opt]# chown -R cloudera-scm:cloudera-scm /opt/cloudera
[root@master opt]# chown -R cloudera-scm:cloudera-scm /opt/cm-5.15.1

• slave节点:

[root@master opt]# chown -R cloudera-scm:cloudera-scm /opt/cm-5.15.1

2.8 根据Cloudera Manager的建议修改/proc/sys/vm/swappiness

echo 10 > /proc/sys/vm/swappiness

2.9 关闭SELinux

vi /etc/selinux/config

SELINUX=disabled

2.10 启动Cloudera Manager的Server和Agent

#在master执行
/opt/cm-5.15.1/etc/init.d/cloudera-scm-server start
#在各个slave节点执行
/opt/cm-5.15.1/etc/init.d/cloudera-scm-agent start

# 查看运行状态
/opt/cm-5.15.1/etc/init.d/cloudera-scm-agent status
/opt/cm-5.15.1/etc/init.d/cloudera-scm-server status

# 查看server日志
tail -f /opt/cm-5.15.1/log/cloudera-scm-server/cloudera-scm-server.log 
# 查看agent日志
tail -f /opt/cm-5.15.1/log/cloudera-scm-agent/cloudera-scm-agent.log

注意server首次启动不要立即关闭或重启，因为首次启动会自动创建相关表以及数据，如果因为特殊原因中途退出，请先删除所有表以及数据之后再次启动，否则将会出现启动不成功的情况。

启动master中server服务时报这个错误:
/opt/cm-5.15.1/etc/init.d/cloudera-scm-server:行109: pstree: 未找到命令
原因：因为系统是最小化安装，默认没有安装
解决：yum install -y psmisc # 安装psmisc

访问http://192.168.3.181:7180:

至此Cloudera Manager安装成功.

3.安装CDH5

所需文件	解压放置位置
CDH-5.15.1-1.cdh5.15.1.p0.4-el7.parcel.sha1	/opt/cloudera/parcel-repo
CDH-5.15.1-1.cdh5.15.1.p0.4-el7.parcel	/opt/cloudera/parcel-repo
manifest.json	/opt/cloudera/parcel-repo

3.1 下载所需文件

下载地址:CDH5.15.1下载.

版本选择: el5-el7对应centos Red Hat 5.X - centos Red Hat 7.X,jessie对应debian发行版本等等…我们用的是centos7.5,选择el7版本,下载.

3.2 上传,移动文件

将下载好的:

CDH-5.15.1-1.cdh5.15.1.p0.4-el7.parcel,
CDH-5.15.1-1.cdh5.15.1.p0.4-el7.parcel.sha1,
manifest.json

三个文件上传到master的/opt/cloudera/parcel-repo/目录下,并将CDH-5.15.1-1.cdh5.15.1.p0.4-el7.parcel.sha1重命名为CDH-5.15.1-1.cdh5.15.1.p0.4-el7.parcel.sha,否则会报找不到hash文件.

3.3 图形化安装

访问http://192.168.3.181:7180,账号密码都是admin.