使用Keepalived+Nginx实现Nginx高可用的负载均衡

接上一文章<<CentOS 6.5高可用集群LVS+Keepalived>>

本文主要是配置Nginx、Keeplive，至于Nginx的配置就省略了

1、服务器规划

服务器IP 　　服务

192.168.80.77 　VIP

192.168.80.188 　Keepalived(Master)、Nginx(Backup)

192.168.80.189 　Keepalived(Slave)、Nginx(Backup)

2、目标

所有的请求都通过1.77虚拟服务转发给1.188服务器，而189作为188的备份，当188服务挂掉了，自动切换到189

3、Nginx配置

安装省略......，Nginx 安装目录：/usr/local/nginx

1） 188、189配置index.html来区分是哪台nginx工作：

在188上配置：

echo "192.168.80.188" > /usr/local/nginx/html/index.html

在189上配置

echo "192.168.80.189" > /usr/local/nginx/html/index.html

验证方法：

1）、首先用IP访问各自的nginx，看index.html页面内容是否为当前服务器的IP地址

2、配置keepalived

按照上面的安装方法，keepalived的配置文件在/etc/keepalived/keepalived.conf。主、从服务器的配置相关联但有所不同。如下：

Master:

! Configuration File for keepalived

global_defs {
   notification_email {
     acassen@firewall.loc
   }
   notification_email_from Alexandre.Cassen@firewall.loc
   smtp_server 192.168.200.1
   smtp_connect_timeout 30
   router_id LVS_MASTER
}

vrrp_instance VI_1 {
    state MASTER
    interface eth0
    virtual_router_id 51
    priority 100
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        192.168.80.77
    }
}

virtual_server 192.168.80.77 80 {
    delay_loop 6
    lb_algo rr
    lb_kind DR
    nat_mask 255.255.255.0
    persistence_timeout 50
    protocol TCP

    real_server 192.168.80.188 80 {
        weight 1
		TCP_CHECK {
			connect_timeout 3
			nb_get_retry 3
			delay_before_retry 3
		}
	}

	real_server 192.168.80.189 80 {
        weight 1
        TCP_CHECK {
            connect_timeout 3
            nb_get_retry 3
            delay_before_retry 3
        }
    }
}

Backup:

! Configuration File for keepalived

global_defs {
   notification_email {
     acassen@firewall.loc
   }
   notification_email_from Alexandre.Cassen@firewall.loc
   smtp_server 192.168.200.1
   smtp_connect_timeout 30
   router_id LVS_BACKUP
}

vrrp_instance VI_1 {
    state BACKUP
    interface eth0
    virtual_router_id 51
    priority 90
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        192.168.80.77
    }
}

virtual_server 192.168.80.77 80 {
    delay_loop 6
    lb_algo rr
    lb_kind DR
    nat_mask 255.255.255.0
    persistence_timeout 50
    protocol TCP

    real_server 192.168.80.188 80 {
        weight 1
		TCP_CHECK {
			connect_timeout 3
			nb_get_retry 3
			delay_before_retry 3
		}
	}

	real_server 192.168.80.189 80 {
        weight 1
        TCP_CHECK {
            connect_timeout 3
            nb_get_retry 3
            delay_before_retry 3
        }
    }
}

验证：

先后在主、从服务器上启动keepalived: /etc/init.d/keepalived start
在主服务器上查看是否已经绑定了虚拟IP： ip addr
停止主服务器上的keepalived: /etc/init.d/keepalived stop 然后在从服务器上查看是否已经绑定了虚拟IP：
启动主服务器上的keepalived，看看主服务器能否重新接管虚拟IP

让keepalived监控NginX的状态

经过前面的配置，如果主服务器的keepalived停止服务，从服务器会自动接管VIP对外服务；一旦主服务器的keepalived恢复，会重新接管VIP。但这并不是我们需要的，我们需要的是当NginX停止服务的时候能够自动切换。

keepalived支持配置监控脚本，我们可以通过脚本监控NginX的状态，如果状态不正常则进行一系列的操作，最终仍不能恢复NginX则杀掉keepalived，使得从服务器能够接管服务。

如何监控NginX的状态

最简单的做法是监控NginX进程，更靠谱的做法是检查NginX端口，最靠谱的做法是检查多个url能否获取到页面。

如何尝试恢复服务

如果发现NginX不正常，重启之。等待3秒再次校验，仍然失败则不再尝试。

根据上述策略很容易写出监控脚本。这里使用nmap检查nginx端口来判断nginx的状态，记得要首先安装nmap。监控脚本如下:

#!/bin/sh
# check nginx server status
NGINX=/usr/local/nginx/sbin/nginx
PORT=80

nmap 127.0.0.1 -p $PORT | grep "$PORT/tcp open"
#echo $?
if [ $? -ne 0 ];then
    #$NGINX -s stop
    $NGINX
    sleep 3
    nmap 127.0.0.1 -p $PORT | grep "$PORT/tcp open"
    #[ $? -ne 0 ] && /etc/init.d/keepalived stop
    [ $? -ne 0 ] && killall keepalived
fi

不要忘了设置脚本的执行权限，否则不起作用。

假设上述脚本放在/usr/local/nginx/chk_nginx.sh，则keepalived.conf中增加如下配置：

vrrp_script chk_http_port {
    script "/usr/local/nginx/chk_nginx.sh"
    interval 2
    weight 2
}

track_script {
    chk_http_port
}

增加完成后的配置如下：

Master:

! Configuration File for keepalived

global_defs {
   notification_email {
     acassen@firewall.loc
   }
   notification_email_from Alexandre.Cassen@firewall.loc
   smtp_server 192.168.200.1
   smtp_connect_timeout 30
   router_id LVS_MASTER
}

vrrp_instance VI_1 {
    state MASTER
    interface eth0
    virtual_router_id 51
    priority 100
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        192.168.80.77
    }
    track_script {
        chk_http_port
    }
}

vrrp_script chk_http_port {
    script "/usr/local/nginx/chk_nginx.sh"
    interval 2
    weight 2
}

virtual_server 192.168.80.77 80 {
    delay_loop 6
    lb_algo rr
    lb_kind DR
    nat_mask 255.255.255.0
    persistence_timeout 50
    protocol TCP

    real_server 192.168.80.188 80 {
        weight 1
        TCP_CHECK {
            connect_timeout 3
            nb_get_retry 3
            delay_before_retry 3
        }
    }

    real_server 192.168.80.189 80 {
        weight 1
        TCP_CHECK {
            connect_timeout 3
            nb_get_retry 3
            delay_before_retry 3
        }
    }
}

Backup:

! Configuration File for keepalived

global_defs {
   notification_email {
     acassen@firewall.loc
   }
   notification_email_from Alexandre.Cassen@firewall.loc
   smtp_server 192.168.200.1
   smtp_connect_timeout 30
   router_id LVS_BACKUP
}

vrrp_instance VI_1 {
    state BACKUP
    interface eth0
    virtual_router_id 51
    priority 90
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        192.168.80.77
    }
    track_script {
        chk_http_port
    }
}

vrrp_script chk_http_port {
    script "/usr/local/nginx/chk_nginx.sh"
    interval 2
    weight 2
}

virtual_server 192.168.80.77 80 {
    delay_loop 6
    lb_algo rr
    lb_kind DR
    nat_mask 255.255.255.0
    persistence_timeout 50
    protocol TCP

    real_server 192.168.80.188 80 {
        weight 1
        TCP_CHECK {
            connect_timeout 3
            nb_get_retry 3
            delay_before_retry 3
        }
    }

    real_server 192.168.80.189 80 {
        weight 1
        TCP_CHECK {
            connect_timeout 3
            nb_get_retry 3
            delay_before_retry 3
        }
    }
}

更进一步，为了避免启动keepalived之前没有启动nginx , 可以在/etc/init.d/keepalived的start中首先启动nginx:

start() {
    /usr/local/nginx/sbin/nginx
    sleep 3
    echo -n $"Starting $prog: "
    daemon keepalived ${KEEPALIVED_OPTIONS}
    RETVAL=$?
    echo
    [ $RETVAL -eq 0 ] && touch /var/lock/subsys/$prog
}

参考：http://www.cnblogs.com/holbrook/archive/2012/10/25/2738475.html#sec-5

来源：oschina

链接：https://my.oschina.net/u/92865/blog/716866

标签

nginx

nginx集群

keepalived