keepalived+httpd 做web服务的高可用

场景：

　　环境中有两台httpd服务器，一台做主，一台做备用；平时只用主向外提供http服务；当主宕机后，keepalived把vip绑定到备机上去，这样就由备机提供http服务了。

　　**keepalvied 这个软件还是更新的比较快的，主要问题再于它的文档并没能跟上来，不同版本之间的差别比较大，我这里用的是1.2.13版

【0】环境说明

主机ip            系统版本                keepavlied版本
192.168.80.131　　centos6.6　　　　　　　　keepalived-1.2.13
192.168.80.132　　centos6.6　　　　　　　　keepalived-1.2.13

【1】安装keepalived 所需要的依赖

yum -y install gcc ipvsadm openssl-devel popt-devel kernel-devel

**由于keepalived 通常来说是和lvs配合用的，虽然这里还用不着lvs，但是我还是把ipvsadm安装上了

【2】编译安装keepalived

tar -xzvf keepalived-1.2..tar.gz
cd keepalived-1.2.
./configure --prefix=/usr/local/keepalived --with-kernel-dir=/usr/src/kernels/2.6.-.el6.x86_64
make
make install

【3】修改192.168.80.131主机上的/usr/local/keepalived/etc/keepalived.conf

! Configuration File for keepalived

global_defs {
   router_id LVS_001B #router_id 这要个唯一
}

vrrp_instance VI_1 {
    state MASTER
    interface eth0
    virtual_router_id 51 #这个virtual_router_id 在两台机器上要相同
    priority
    advert_int
    authentication {
        auth_type PASS
        auth_pass
    }
    virtual_ipaddress {
        192.168.80.80/
    }
}

virtual_server 192.168.80.80  {
    delay_loop
    lb_algo rr
    lb_kind DR
    persistence_timeout
    protocol TCP
    real_server 192.168.80.131  {
        weight
        notify_down /tmp/stop_keepalive.sh #当检测到real_server 宕机的时候就会执行这个脚本
        TCP_CHECK {
            connect_port
            connect_timeout
            nb_get_retry
            delay_before_retry
        }
    }
}

【4】修改192.168.80.132主机上的/usr/local/keepalived/etc/keepalived.conf

! Configuration File for keepalived

global_defs {
   router_id LVS_001B
}

vrrp_instance VI_1 {
    state BACKUP
    interface eth0
    virtual_router_id
    priority
    advert_int
    authentication {
        auth_type PASS
        auth_pass
    }
    virtual_ipaddress {
        192.168.80.80/
    }
}

virtual_server 192.168.80.80  {
    delay_loop
    lb_algo rr
    lb_kind DR
    persistence_timeout
    protocol TCP
    real_server 192.168.80.132  {
        weight
        notify_down /tmp/stop_keepalive.sh
        TCP_CHECK {
            connect_port
            connect_timeout
            nb_get_retry
            delay_before_retry
        }
    }
}

【5】脚本/tmp/stop_keepalived.sh的内容如下：

#!/bin/bash
pkill keepalived

【6】启动keepalived程序

nohup /usr/local/keepalived/sbin/keepalived -D -f /usr/local/keepalived/etc/keepalived/keepalived.conf &