LVS + keepalived的实现

! Configuration File for keepalived  

global_defs {
   notification_email {
         linuxedu@foxmail.com
         mageedu@.com
   }
   notification_email_from kanotify@magedu.com
   smtp_connect_timeout
   smtp_server 127.0.0.1
   router_id LVS_DEVEL
}  

vrrp_script chk_schedown {
   script "[[ -f /etc/keepalived/down ]] && exit 1 || exit 0"
   interval
   weight -
}

vrrp_instance VI_1 {
    interface eth0
    state MASTER
    priority
    virtual_router_id
    garp_master_delay  

    authentication {
        auth_type PASS
        auth_pass password
    }  

    track_interface {
       eth0
    }  

    virtual_ipaddress {
        172.16.100.1/ dev eth0 label eth0:
    }  

    track_script {
        chk_schedown
    }
} 

virtual_server 172.16.100.1  {
    delay_loop
    lb_algo rr
    lb_kind DR
    persistence_timeout
    protocol TCP

#    sorry_server 192.168.200.200 

    real_server 172.16.100.11  {
        weight
        HTTP_GET {
            url {
              path /
              status_code
            }
            connect_timeout
            nb_get_retry
            delay_before_retry
        }
    }

    real_server 172.16.100.12  {
        weight
        HTTP_GET {
            url {
              path /
              status_code
            }
            connect_timeout
            nb_get_retry
            delay_before_retry
        }
    }
}

如果要使用TCP_CHECK检测各realserver的健康状态，那么，上面关于realserver部分的定义也可以替换为如下内容：
virtual_server 172.16.100.1  {
    delay_loop
    lb_algo rr
    lb_kind DR
    persistence_timeout
    protocol TCP

    sorry_server 127.0.0.1 

    real_server 172.16.100.11  {
        weight
        TCP_CHECK {
        tcp_port
            connect_timeout
        }
    }

    real_server 172.16.100.12  {
        weight
        TCP_CHECK {
        connect_port
            connect_timeout
          }
    }
}

说明：其中的sorry_server是用于定义所有realserver均出现故障时所用的服务器。