image.png
上面的拓扑图，LB1和LB2是两个Nginx反向代理，后端RS1和RS2是两个Web服务器。
接下来我们通过keepalived实现高可用的nginx反向代理集群。
keepalived的实现本质是：
当LB1和LB2都正常运行时，把VIP 172.16.80.201分配给高优先级的LB1。用户访问时，就由LB1提供反向代理。
当LB1出现故障时，keepalived就会把VIP漂移到LB2，由LB2提供反向代理功能。从而保证了业务的正常运行。
RS1和RS2配置
1、安装Nginx
[root@RS1 ~]#yum install nginx -y
[root@RS2 ~]#yum install nginx -y
2、配置访问页面，并启动RS1和RS2的Nginx服务
[root@RS1 ~]#vim /usr/share/nginx/html/index.html
RS1 172.16.80.100
[root@RS2 ~]#vim /usr/share/nginx/html/index.html
RS2 172.16.80.102
客户端访问RS1和RS2效果如下：
web页面如下：
image.png
LB1和LB2的配置
1、LB1和LB2的Nginx负载均衡配置
[root@LB1 nginx]#vim /etc/nginx/nginx.conf
worker_processes 4;
events {
worker_connections 1024;
}
http {
include mime.types;
default_type application/octet-stream;
server_tokens off;
sendfile on;
keepalive_timeout 65;
upstream www.server.pools {
server 172.16.80.100:80 weight=1;
server 172.16.80.102:80 weight=1;
}
server {
listen 80;
server_name www.keepalived.com;
location / {
proxy_pass http://www.server.pools;
proxy_set_header Host $host;
proxy_set_header X-Forwardered-For $remote_addr;
}
}
}
LB2的Nginx配置与LB1一样。配置后重启nginx服务。客户端分别访问172.16.80.101和172.16.80.103，可以看到LB实现了负责均衡的反向代理：
image.png
2、LB1和LB2通过keepalived实现高可用
安装keepalived
[root@LB1 ~]#yum install keepalived ipvsadm -y
[root@LB2 ~]#yum install keepalived ipvsadm -y
配置VRRP实例，LB1为master，LB22为backup。
VS1配置：
[root@LB1 ~]#vim /etc/keepalived/keepalived.conf
! Configuration File for keepalived
global_defs {
notification_email {
root@localhost
}
notification_email_from keadmin@localhost
smtp_server 127.0.0.1
smtp_connect_timeout 30
router_id CentOS7A.luo.com
vrrp_mcast_group4 224.0.0.22
}
vrrp_instance VI_1 {
state MASTER
interface ens33
virtual_router_id 51
priority 150
advert_int 1
authentication {
auth_type PASS
auth_pass haha
}
virtual_ipaddress {
172.16.80.201/16
}
}
LB2配置
[root@LB2 ~]#vim /etc/keepalived/keepalived.conf
! Configuration File for keepalived
global_defs { <==全局配置段开始
notification_email {
root@localhost <==故障邮件的收件人
}
notification_email_from keadmin@localhost <==故障邮件的发件人
smtp_server 127.0.0.1 <==故障邮件的发件服务器
smtp_connect_timeout 30
router_id CentOS7B.luo.com <==路由器ID
vrrp_mcast_group4 224.0.0.22 <==组播地址
} <==全局配置段结束
vrrp_instance VI_1 { <==实例名字为VI_1，备节点的实例名字要和主节点的相同
state BACKUP <==状态为backup
interface ens33 <==通信的接口
virtual_router_id 51 <==实例ID
priority 100 <==优先级
advert_int 1 <==通信检查时间间隔
authentication {
auth_type PASS <==认证类型
auth_pass haha <==认证密码
}
virtual_ipaddress {
172.16.80.201/16 <==虚拟IP
}
}
配置完后，启动VS1和VS2的keepalived
image.png
image.png
上图可以看到，LB1成功拿到VIP 172.16.80.201。
验证
image.png
客户端访问http://172.16.80.201时，被轮询调度到RS1和RS2。
当我把LB1关机后，VIP就会被keepalived分配到LB2。这时客户端再次访问http://172.16.80.201，仍然被轮询调度到后端RS服务器，用户是无法感知到LB1的故障的。
