一.结构以及配置：
1.nginx基本结构信息
Nginx+keepalived（两台）
192.168.1.77 nginx lb0
192.168.1.60 nginx lb1
192.168.1.80 vip0 lb0
192.168.1.81 vip1 lb1
域名：
192.168.1.80 www.crh.com
192.168.1.81 www.crh.com
2.软件安装和配置
#tar zxf keepalived-1.2.7
#cd keepalived-1.2.7
#./configure
#make && make install
安装完成之后，拷贝启动文件和配置文件等。
#cp /usr/local/etc/rc.d/init.d/keepalived /etc/init.d/keepalived
#cp /usr/local/etc/sysconfig/keepalived /etc/sysconfig/keepalived
#cp /usr/local/etc/keepalived/keepalived.conf /etc/keepalived/keepalived.conf
配置文件：
/home/ytb/server/conf/nginx_main.conf （nginx主配置文件）
/etc/keepalived/keepalived.conf (keepalived主配置文件）
3.启动脚本：
/etc/init.d/nginx start
/etc/inti.d/keepalived start
二.Keepalived配置
1.Nginx双主负载lb0服务器的keepalived配置文件的配置
nginx lb0 （192.168.1.77）
#vi /etc/keepalived/keepalived.conf
! Configuration File for keepalived
global_defs {
router_id Nginx_Id_1
}
vrrp_script chk_nginx {
script "/root/bin/check_http.sh" #定义检测脚本
interval 2 #检测间隔
weight 2
}
vrrp_instance Nginx1 {
state MASTER
interface eth1
virtual_router_id 51
priority 150
authentication {
auth_type PASS
auth_pass w3e4r5t5
}
track_script {
chk_nginx
}
virtual_ipaddress {
192.168.1.80
}
}
vrrp_instance Nginx2 {
state BACKUP
interface eth1
virtual_router_id 52
priority 110
authentication {
auth_type PASS
auth_pass w3e4r5t5
}
virtual_ipaddress {
192.168.1.81
}
}
检测脚本：
url="http://www.kankanapp.com/test.html"
status=$(/usr/bin/curl -s --head "$url" | awk '/HTTP/ {print $2}')
if [ "$status" != "200" ];then
/etc/init.d/keepalived stop
fi
注: 以上脚本是判断如果主服务器上的nginx 不正常,则马上停止主服务器的keepalived 工作, IP 会自动切换到从服务器上，keepalive 只检测两个keepalived 是否正常工作来决定切换vip，不使用此脚本如果nginx 挂了就不会自动切换了。
2.Nginx双主负载lb0服务器的keepalived配置文件的配置.
Nginx 负载均衡备份服务器(同时此服务器也担当其中的一台源服务器)
192.168.1.60 nginx lb1
#vi /etc/keepalived/keepalived.conf
! Configuration File for keepalived
global_defs {
router_id Nginx_Id_2
}
vrrp_script chk_nginx {
script "/root/bin/check_http.sh"
interval 2
weight 2
}
vrrp_instance Nginx1 {
state BACKUP
interface eth0
virtual_router_id 51
priority 110
authentication {
auth_type PASS
auth_pass w3e4r5t5
}
virtual_ipaddress {
192.168.1.80
}
}
vrrp_instance Nginx2 {
state MASTER
interface eth0
virtual_router_id 52
priority 150
authentication {
auth_type PASS
auth_pass w3e4r5t5
}
track_script {
chk_nginx
}
virtual_ipaddress {
192.168.1.81
}
}
LB1上面也有检测脚本，跟上面LB0的检测脚本一样，把他们放到后台运行就可以达到检测的效果了
三.切换测试
此架构是为了充分利用服务器，并负载均衡，两个vip 使用DNS 轮询的方式将域名www.crh.com 绑定到这两个IP 上
1.将nginx lb0 上的nginx 进程杀掉或down 机,vip0 会自动切换至nginx lb1 上.
2.当nginx lb0 服务器恢复以后,vip0 又切回lb0 上,为了保持负载均衡.
同理在lb1 上测试结果一样.
3.测试配图如下：
当lb0执行 pkill -9 nginx后
在LB0 上执行/etc/init.d/nginx start 启动nginx，并执行service keepalived start 后：
四，补充说明
1.此文档单纯是为了测试Nginx+KeepAlived 双主，不涉及nginx 配置
2.两个VIP 在应用正常的情况下一定要分布在两台机器上以达到负载均衡的目的,但是有可能在服务器故障恢复的时候出现session 不能同步的问题，是瞬间现象，如果想要不抢占，请把MASTER设成BACKUP,并添加参数nopreempt
