CDH 7.1.4 大数据平台搭建

CDH 7.1.4 大数据平台搭建

CDH 7.1.4 大数据平台搭建-新手指南

• 环境介绍
• • 笔记本硬件
  • 软件及版本
• 环境准备
• • 虚拟系统安装
  • 挂载离线安装ISO文件
  • 安装httpd搭建http服务器
  • 下载CM和CDH安装包
  • 安装createrepo软件
  • 打包CM文件
  • 配置hostname和hosts
  • • hostname
    • hosts
  • 建立互信
  • 重新制作ISO的repo，方便其他机器访问通过http
  • 升级软件和系统内核
  • 禁用SELinux
  • 关闭防火墙
  • 集群时钟同步
  • 设置swap
  • 设置透明大页面
  • 关闭iptables
  • 设置limits
  • 安装Mysql
  • 安装jdbc驱动
  • 安装PostgreSQL(未安装)
  • 安装JDK
• CM安装
• • 初始化数据库
  • 启动Cloudera Manager Server
  • 登录CM
  • 界面
• CDH集群安装

环境介绍

笔记本硬件

硬盘 1T SSD
内存 48G
处理器 I5 8代

软件及版本

系统 Win10 家庭版
虚拟机 VMware 15.5.1
Linux rhel-server-7.9-x86_64-dvd.iso Redhat Linux 7.9 下载地址
CM Cloudera Manager CM下载地址
CDH Cloudera Runtime集群 CDH下载地址

环境准备

虚拟系统安装

准备安装5个节点
192.168.101.141 C1(主节点) 10G
192.168.101.142 C2 8G
192.168.101.143 C3 8G
192.168.101.144 C4 8G
192.168.101.145 C5 8G
每个节点1个CPU
每个节点100G硬盘
电脑之前试过2G内存 20G 内存和空间都不够
改为4G内存 60G硬盘 内存不够，空间也快满了
现改为8G内存 100G硬盘空间

1.安装虚拟机配置

2.克隆虚拟机*4

3.开启5台虚拟机

挂载离线安装ISO文件

为了使用yum安装软件，方便安装软件，系统安装后未配置的情况如下

yum repolist

1.创建软件挂载路径

mkdir -p /soft

2.将系统ISO镜像文件复制到该路径下（使用Xftp 7）

3.修改yum源

cd /pos.d/

po

添加如下内容，先为了安装httpd，安装之后再改为http方式进行repo分发。

[iso]
name=iso
baseurl=file:///mnt/iso/
gpgcheck=0
enabled=1

创建文件夹挂载磁盘镜像

mkdir -p /mnt/iso
mount /soft/rhel-server-7.9-x86_64-dvd.iso /mnt/iso

成功挂载

执行yum repolist看是否成功

yum repolist

安装httpd搭建http服务器

yum -y install httpd

安装成功
启动http服务

systemctl enable httpd
systemctl start httpd

关闭防火墙后访问

systemctl disable firewalld.service 禁止开机启动
systemctl stop firewalld.service 关闭防火墙

用浏览器访问

下载CM和CDH安装包

CM Cloudera Manager CM下载地址
CDH Cloudera Runtime集群 CDH下载地址

下载后将其放置在http服务器路径下，用于其他机器访问安装软件
/var/www/html/cm
/var/www/html/cdh

截图

安装createrepo软件

yum -y install createrepo

打包CM文件

cd /var/www/html/cm

createrepo .

配置hostname和hosts

hostname

/etc/hostname文件

cat /etc/hostname

修改为C1

hostnamectl set-hostname CDH1

并将其余几台电脑改名

hostnamectl set-hostname CDH2
hostnamectl set-hostname CDH3
hostnamectl set-hostname CDH4
hostnamectl set-hostname CDH5

hosts

/etc/hosts

cat /etc/hosts
vi /etc/hosts

追加如下，里面主机的名称重新修改为CDH1-5了，每个节点都需要修改

192.168.101.141 CDH1
192.168.101.142 CDH2
192.168.101.143 CDH3
192.168.101.144 CDH4
192.168.101.145 CDH5

保存退出

建立互信

在管理节点生成秘钥，并放到其他节点上

ssh-keygen

一直按回车

将秘钥分发给其他节点

for i in {2..5}; do ssh-copy-id -i ~/.ssh/id_rsa.pub CDH$i ; done

需要输入yes和密码

检查互信

for i in {2..5}; do ssh CDH$i " hostname -A" ; done

互信配置完成

重新制作ISO的repo，方便其他机器访问通过http

创建对应文件夹，将ISO对应内容全部拷贝过去

mkdir -p /var/www/html/iso/
cp -r /mnt/iso/* /var/www/html/iso/

修改yum的repo文件为http派发

vi /pos.po

192.168.101.141/iso/

[iso]
name=iso
#baseurl=file:///mnt/iso/
baseurl=192.168.101.141/iso/
gpgcheck=0
enabled=1

192.168.101.141 主机master ip

repo分发到其他电脑

for i in {2..5}; do scp /pos.po CDH$i:/pos.po; done

升级软件和系统内核

for i in {1..5}; do ssh CDH$i " yum -y update " ; done

禁用SELinux

for i in {1..5}; do ssh CDH$i 'echo "SELINUX=disabled" > /etc/selinux/config '  ;done

关闭防火墙

for i in {2..5}; do ssh CDH$i 'systemctl disable firewalld.service'  ;done
for i in {2..5}; do ssh CDH$i 'systemctl stop firewalld.service '  ;done

集群时钟同步

卸载chrony

for i in {1..5}; do ssh CDH$i 'yum -y remove chrony'  ;done

安装ntp和ntpdate

for i in {1..5}; do ssh CDH$i 'yum install -y ntp ntpdate '  ;done
for i in {1..5}; do ssh CDH$i ' timedatectl set-timezone Asia/Shanghai'  ;done
for i in {1..5}; do ssh CDH$i 'chkconfig --level 345 ntpd on '  ;done

主节点时钟修改为

vi /f

#server 0.p iburst
#server 1.p iburst
#server 2.p iburst
#server 3.p iburst
server  127.127.1.0     # local clock
fudge   127.127.1.0 stratum 10

根据实际情况编辑时钟同步
其余节点按照主节点同步
注释四行，追加

server 192.168.101.141

停止服务，同步NTP

for i in {2..5}; do ssh CDH$i ' systemctl stop ntpd'  ;done
for i in {2..5}; do ssh CDH$i ' ntpdate CDH1'  ;done

报这个错误说明，ntpd没有停止掉

正常同步

验证时钟同步

for i in {1..5}; do ssh CDH$i 'systemctl enable ntpd '  ;done
for i in {1..5}; do ssh CDH$i 'systemctl start ntpd '  ;done
for i in {1..5}; do ssh CDH$i ' ntpq -p'  ;done

设置swap

for i in {1..5}; do ssh CDH$i ' echo "vm.swappiness = 1" >> /f'  ;done
for i in {1..5}; do ssh CDH$i ' sysctl vm.swappiness=1'  ;done

设置透明大页面

如下脚本追加到vi /etc/rc.d/rc.local

if test -f /sys/kernel/mm/transparent_hugepage/enabled; then
echo never > /sys/kernel/mm/transparent_hugepage/enabled
fi
if test -f /sys/kernel/mm/transparent_hugepage/defrag; then
echo never > /sys/kernel/mm/transparent_hugepage/defrag
fi

for i in {2..5}; do scp /etc/rc.d/rc.local CDH$i:/etc/rc.d/rc.local; done

每个节点执行

echo never > /sys/kernel/mm/transparent_hugepage/defrag
echo never > /sys/kernel/mm/transparent_hugepage/enabled

for i in {1..5}; do ssh CDH$i ' echo never > /sys/kernel/mm/transparent_hugepage/defrag'  ;done
for i in {1..5}; do ssh CDH$i ' echo never > /sys/kernel/mm/transparent_hugepage/enabled'  ;done

同步到其他节点

关闭iptables

执行后，发现没有该服务

for i in {1..5}; do ssh CDH$i 'systemctl stop iptables'  ;done
for i in {1..5}; do ssh CDH$i 'chkconfig iptables off'  ;done 

设置limits

for i in {1..5}; do ssh CDH$i 'echo "hdfs - nofile 32768" >> /etc/f';done
for i in {1..5}; do ssh CDH$i 'echo "mapred - nofile 32768" >> /etc/f';done
for i in {1..5}; do ssh CDH$i 'echo "hbase - nofile 32768" >> /etc/f';done
for i in {1..5}; do ssh CDH$i 'echo "hdfs - noproc 32768" >> /etc/f';done
for i in {1..5}; do ssh CDH$i 'echo "mapred - noproc 32768" >> /etc/f';done
for i in {1..5}; do ssh CDH$i 'echo "hbase - noproc 32768" >> /etc/f';done

安装Mysql

wget .noarch.rpm

rpm -arch.rpm

检查yum

yum repolist enabled | grep mysql

按照mysql

yum install -y mysql-community-server

启动MySQL

systemctl start  mysqld.service

查看状态

systemctl status mysqld.service

查找初始密码

grep "password" /var/log/mysqld.log

kjNAA_K11;_#

进入数据库

mysql -uroot -p

修改密码

set password=password("Cdh123456-");

停止服务

systemctl stop mysqld

修改配置文件
最好先做备份

vi /etc/myf

追加

character_set_server=utf8
init_connect='SET NAMES utf8'
transaction-isolation = READ-COMMITTED
key_buffer_size = 32M
max_allowed_packet = 32M
thread_stack = 256K
thread_cache_size = 64
query_cache_limit = 8M
query_cache_size = 64M
query_cache_type = 1
max_connections = 550
binlog_format = mixed
read_buffer_size = 2M
read_rnd_buffer_size = 16M
sort_buffer_size = 8M
join_buffer_size = 8M
innodb_file_per_table = 1
innodb_flush_log_at_trx_commit = 2
innodb_log_buffer_size = 64M
innodb_buffer_pool_size = 4G
innodb_thread_concurrency = 8
innodb_flush_method = O_DIRECT
innodb_log_file_size = 512M
sql_mode = STRICT_ALL_TABLES

如下

设置mysql自启动

sudo systemctl enable mysqld
sudo systemctl start mysqld

配置数据库

sudo /usr/bin/mysql_secure_installation

n y n y y

链接数据库

mysql -uroot -p

创建数据库

CREATE DATABASE scm DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON scm.* TO 'scm'@'%' IDENTIFIED BY 'Cdh123456-';
CREATE DATABASE amon DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON amon.* TO 'amon'@'%' IDENTIFIED BY 'Cdh123456-';
CREATE DATABASE rman DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON rman.* TO 'rman'@'%' IDENTIFIED BY 'Cdh123456-';
CREATE DATABASE hue DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON hue.* TO 'hue'@'%' IDENTIFIED BY 'Cdh123456-';
CREATE DATABASE hive DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON hive.* TO 'hive'@'%' IDENTIFIED BY 'Cdh123456-';
CREATE DATABASE ranger DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON ranger.* TO 'rangeradmin'@'%' IDENTIFIED BY 'Cdh123456-';
CREATE DATABASE nav DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON nav.* TO 'nav'@'%' IDENTIFIED BY 'Cdh123456-';
CREATE DATABASE navms DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON navms.* TO 'navms'@'%' IDENTIFIED BY 'Cdh123456-';
CREATE DATABASE oozie DEFAULT CHARACTER SET utf8 DEFAULT COLLATE utf8_general_ci;
GRANT ALL ON oozie.* TO 'oozie'@'%' IDENTIFIED BY 'Cdh123456-';

FLUSH PRIVILEGES; # ...

show databases;

安装jdbc驱动

wget .1.

tar zxvf mysql-connector-java-5.1.

for i in {1..5}; do ssh CDH$i " mkdir -p /usr/share/java/" ; done

for i in {1..5}; do scp /soft/mysql-connector-java-5.1.46/mysql-connector-java-5.1.46-bin.jar CDH$i:/usr/share/java/mysql-connector-java.jar; done

安装PostgreSQL(未安装)

wget .noarch.rpm

rpm -arch.rpm

ls -lrt /pos.d/

yum list postgresql*

我们这里安装PostgreSQL10的数据库

sudo yum -y install postgresql10-server

安装JDK

for i in {1..5}; do ssh CDH$i 'yum -y install java-1.8.0-openjdk.x86_64 java-1.8.0-openjdk-headless.x86_64 java-1.8.0-openjdk-devel.x86_64'  ;done

CM安装

vi /pos.d/po

新增po

分发

for i in {2..5}; do scp /pos.d/po CDH$i:/pos.d/po; done

安装

yum repolist
yum -y install cloudera-manager-daemons cloudera-manager-agent cloudera-manager-server

初始化数据库

语法

sudo /opt/cloudera/cm/schema/scm_prepare_database.sh [options] <databaseType> <databaseName> <databaseUser> <password>

sudo /opt/cloudera/cm/schema/scm_prepare_database.sh mysql scm scm Cdh123456-

启动Cloudera Manager Server

systemctl start cloudera-scm-server
systemctl status cloudera-scm-server

查看启动日志

sudo tail -f /var/log/cloudera-scm-server/cloudera-scm-server.log

当看到如下，准备就绪

INFO WebServerImpl:com.f.WebServerImpl: Started Jetty server

检查端口是否监听

netstat -lnpt | grep 7180

登录CM

192.168.101.141:7180

admin/admin

界面

CDH集群安装

删除多余远程URL

测试链接