CentOS 6.3下部署LVS(NAT)+keepalived实现高性能高可用负载均衡

一、简介

VS/NAT原理图：

二、系统环境

实验拓扑：

系统平台：CentOS 6.3

Kernel：2.6.32-279.el6.i686

LVS版本：ipvsadm-1.26

keepalived版本：keepalived-1.2.4

三、安装

0、安装LVS前系统需要安装popt-static，kernel-devel，make，gcc，openssl-devel，lftp，libnl*，popt*

1、在两台Director Server上分别配置LVS+Keepalived

LVS install -------------

[root@CentOS-LVS_MASTER ~]# wget

[root@CentOS-LVS_MASTER ~]# ln -s /usr/src/kernels/2.6.32-279.el6.i686//usr/src/linux/

[root@CentOS-LVS_MASTER ~]# tar zxvf ipvsadm-1.26.tar.gz

[root@CentOS-LVS_MASTER ~]# cd ipvsadm-1.26

[root@CentOS-LVS_MASTER ipvsadm-1.26]# make && make install

Keepalived install -------------

[root@CentOS-LVS_MASTER ~]# wget

[root@CentOS-LVS_MASTER ~]# tar zxvf keepalived-1.2.4.tar.gz

[root@CentOS-LVS_MASTER ~]# cd keepalived-1.2.4

[root@CentOS-LVS_MASTER keepalived-1.2.4]# ./configure && make && make install

######### 将keepalived做成启动服务，方便管理##########

[root@CentOS-LVS_MASTER ~]# cp /usr/local/etc/rc.d/init.d/keepalived /etc/init.d/

[root@CentOS-LVS_MASTER ~]# cp /usr/local/etc/sysconfig/keepalived /etc/sysconfig/

[root@CentOS-LVS_MASTER ~]# mkdir /etc/keepalived/

[root@CentOS-LVS_MASTER ~]# cp /usr/local/etc/keepalived/keepalived.conf /etc/keepalived/

[root@CentOS-LVS_MASTER ~]# cp /usr/local/sbin/keepalived /usr/sbin/

[root@CentOS-LVS_MASTER ~]# service keepalived start | stop

2、开启路由转发

[root@CentOS-LVS_MASTER ~]# vim /etc/sysctl.conf 
net.ipv4.ip_forward = 1
[root@CentOS-LVS_MASTER ~]# sysctl -p

3、配置Keepalived

[root@CentOS-LVS_MASTER ~]# less /etc/keepalived/keepalived.conf

! Configuration File for keepalived

global_defs {
   router_id LVS_MASTER #BACKUP上修改为LVS_BACKUP }

vrrp_instance VI_1 {
    state MASTER #BACKUP上修改为BACKUP interface eth0
    virtual_router_id 51
    priority 100 #BACKUP上修改为80 advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        10.0.0.227
    }
}

vrrp_instance LAN_GATEWAY {
    state MASTER #BACKUP上修改为LVS_BACKUP interface eth1
    virtual_router_id 52
    priority 100 #BACKUP上修改为80 advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        192.168.10.10
    }
}

virtual_server 10.0.0.227 80 {
    delay_loop 6
    lb_algo rr
    lb_kind NAT
#   persistence_timeout 5
    protocol TCP

    real_server 192.168.10.4 80 {
        weight 3
        TCP_CHECK {
            connect_timeout 3
            nb_get_retry 3
            delay_before_retry 3
            connect_port 80
        }
    }
    
    real_server 192.168.10.5 80 {
        weight 3
        TCP_CHECK {
            connect_timeout 3
            nb_get_retry 3
            delay_before_retry 3
            connect_port 80
        }
    }
}

BACKUP服务器同上配置，先安装lvs再安装keepalived，然后配置/etc/keepalived/keepalived.conf，只需将批注部分改一下即可。

4、分别在2台Real Server上面设置网关

把网关都设置成：192.168.10.10

5、在2台RealServer中分别配置HTTP

[root@WEB1 ~]# yum -y install httpd
[root@WEB1 ~]# cd /var/www/html/
[root@WEB1 html]# cat index.html 

WEB1/192.168.10.4

[root@WEB1 html]# /etc/init.d/httpd start
另一台机器配置一样，过程略。

6、分别在CentOS-LVS_MASTER、CentOS-LVS_BACKUP上执行service keepalived start启动keepalived就可实现负载均衡及高可用集群了；

[root@CentOS-LVS_MASTER keepalived]# service keepalived start

四、测试

####高可用性测试####

模拟故障，将CentOS-LVS_MASTER上的keepalived服务停掉，然后观察CentOS-LVS_BACKUP上的日志，信息如下

从日志中可知，主机出现故障后，备机立刻检测到，此时备机变为MASTER角色，并且接管了主机的虚拟IP资源，最后将虚拟IP绑定在etho设备上。

将CentOS-LVS_MASTER 上的keepalived服务开启后，CentOS-LVS_BACKUP的日志状态。

从日志可知，备机在检测到主机重新恢复正常后，释放了虚拟IP资源重新成为BACKUP角色

####故障切换测试####

故障切换是测试当某个节点出现故障后，Keepalived监制模块是否能及时发现然后屏蔽故障节点，同时将服务器转移到正常节点来执行。

将WEB2节点服务停掉，假设这个节点出现故障，然后主、备机日志信息如下

从以上可以看出，Keepalived监控模块检测到192.168.10.5这台主机出现故障后，将WEB2从集群系统中剔除掉了。 此时访问只能看到WEB1了)

重新启动WEB2节点的服务，日志信息如下：

Keepalived监控模块检测到192.168.10.5这台主机恢复正常后，又将此节点加入集群系统中，再次访问就可以访问到WEB2页面了）