nginx+keepalived实现负载均衡和高可用

环境准备

	IP	VIP	环境
客户端	192.168.134.174
Master	192.168.134.170	192.168.134.100	需要配置nginx负载均衡
Backup	192.168.134.172	192.168.134.100	需要配置nginx负载均衡
web1服务器	192.168.134.171
web2服务器	192.168.134.173

1、首先安装nginx服务器（此处采用yum安装）

wget -O /etc/yum.repos.d/epel.repo https://mirrors.aliyun.com/repo/epel-7.repo
yum install nginx -y

2、修改nginx的配置文件（配置负载均衡）

cd /etc/nginx/conf.d/
vim web.confupstream webPools {server 192.168.134.171;server 192.168.134.173;
}
server {location / {proxy_pass http://webPools;}
}

3、启动nginx

systemctl start nginx

4、修改keepalived配置文件

主节点：
! Configuration File for keepalivedglobal_defs {notification_email {[e-mail]@[e-mail]}notification_email_from [e-mail]_server 192.168.200.1smtp_connect_timeout 30router_id LVS_DEVEL1
}vrrp_instance VI_1 {state MASTER# nopreemptinterface ens33virtual_router_id 51priority 100advert_int 1authentication {auth_type PASSauth_pass 1111}virtual_ipaddress {192.168.134.100}
}
备节点
修改：state MASTERpriority 80

此时可以实现keepalived的故障切换和nginx负载均衡，但是如果nginx的主服务器出现故障，那么此时无法实现客户端的正常访问，即需要新的配置来实现高可用，因此，利用·vrrp-script（利用VIP漂移实现服务的可用）去监控集群资源。

5、重新修改keepalived配置文件，添加vrrp-script，在实例中还要调用。

主备节点做一样的修改
! Configuration File for keepalivedglobal_defs {notification_email {[e-mail]@[e-mail]}notification_email_from [e-mail]_server 192.168.200.1smtp_connect_timeout 30router_id LVS_DEVEL1
}
vrrp_script check_nginx {  # 自定义资源监控脚本script "killall -0 nginx"interval 2
}
vrrp_instance VI_1 {state MASTER# nopreemptinterface ens33virtual_router_id 51priority 100advert_int 1authentication {auth_type PASSauth_pass 1111}track_script {check__nginx}virtual_ipaddress {192.168.134.100}
}

6、还可以用脚本检测服务的方法

cat check_#!/bin/bashtotal=$(ps -C nginx --no-header | wc -l)if [ $total -eq 0 ]
thennginx_status=1
elsenginx_status=0
fiexit $nginx_status在keepalived文件中也进行修改vrrp_script check_nginx {#script "killall -0 nginx"script "/etc/keepalived/check_nginx.sh"interval 2
}

7、还有一种使用notify的·方法

写notify.sh脚本#!/bin/bashcase "$1" inmaster)nmap localhost -p 80 | grep "80/tcp open"if [ $? -ne 0 ];thensystemctl start nginxfi;;
backup)nginx_psr=`ps -C nginx --no-header | wc -l`if [ $nginx_psr -ne 0 ];thensystemctl stop nginxfi;;
*)echo "Usage:$0 master|backup";;
esacchmod +x notify.sh  # 给脚本增加执行权限然后在keepalived文件中修改就行，修改下面几个地方
vrrp_script check_nginx {#script "killall -0 nginx"script "/etc/keepalived/check_nginx.sh"interval 2weight -30
}track_script {check_nginx}virtual_ipaddress {192.168.134.100}notify_master "/etc/keepalived/notify.sh master"notify_backup "/etc/keepalived/notify.sh backup"
}

8、测试如下：

在停掉nginx后，仍旧可以正常访问systemctl stop nginx
结果如下：
web test page ,ip is 192.168.134.173 192.168.122.1 
web test page ,ip is 192.168.134.171 192.168.122.1 
web test page ,ip is 192.168.134.173 192.168.122.1 
web test page ,ip is 192.168.134.173 192.168.122.1 
web test page ,ip is 192.168.134.171 192.168.122.1 
web test page ,ip is 192.168.134.171 192.168.122.1