1…脑裂模拟

1）开启防火墙（两台lb开启防火墙，都出现vip10.0.0.3）

[root@lb01 ~]# systemctl start firewalld.service 
[root@lb01 ~]# ip addr | grep 10.0.0.3
    inet 10.0.0.3/32 scope global eth0
    
[root@lb02 ~]# systemctl start firewalld
[root@lb02 ~]# ip addr | grep 10.0.0.3
    inet 10.0.0.3/32 scope global eth0

2）做免密交互

[root@lb02 ~]#  ssh-keygen
[root@lb02 ~]#  ssh-copy-id 172.16.1.4
[root@lb02 ~]# ssh 172.16.1.4 'ip addr |grep 10.0.0.3'
    inet 10.0.0.3/32 scope global eth0
[root@lb02 ~]# echo $?
0
[root@lb02 ~]# ssh 172.16.1.4 'ip addr |grep 10.0.0.3'
[root@lb02 ~]# echo $?
1

2.脑裂解决的办法

#如果发生闹裂，则随机kill掉一台即可
#在备上编写检测脚本, 测试如果能ping通主并且备节点还有VIP的话则认为产生了脑裂

[root@lb02 ~]# vim check_keepalive.sh 
#!/bin/sh
vip=10.0.0.3
lb01_ip=172.16.1.4
while true;do
    ssh $lb01_ip 'ip addr | grep 10.0.0.3' &>/dev/null
    if [ $? -eq 0 -a `ip add|grep "$vip"|wc -l` -eq 1 ];then
        echo "ha is split brain.warning."
    else
        echo "ha is ok"
    fi
sleep 3
done

如下

[root@lb01 ~]# systemctl stop keepalived.service 

重点：在备上执行脚本，两台lb显示脑列。如果关闭主的keepalived，则会显示OK。

[root@lb02 ~]# sh check_keepalivded.sh 
ha is split brain.warning.
ha is split brain.warning.
ha is split brain.warning.

ha is ok
ha is ok
ha is ok

如果关闭lb01的eth0网卡，ifdown eth0，vip也就起不来。因为VIP绑定在eth0网卡上。这样就不会出现脑列了。

二、高可用keepalived

Nginx默认监听在所有的IP地址上，VIP会飘到一台节点上，相当于那台nginx多了VIP这么一个网卡，所以可以访问到nginx所在机器

但是…如果nginx宕机，会导致用户请求失败，但是keepalived没有挂掉不会进行切换，所以需要编写一个脚本检测Nginx的存活状态，如果不存活则kill掉keepalived

1.nginx故障切换脚本

[root@lb01 ~]# vim check_web.sh
#!/bin/sh
nginxpid=$(ps -ef | grep [n]ginx | wc -l)

#1.判断Nginx是否存活,如果不存活则尝试启动Nginx
if [ $nginxpid -eq 0 ];then
    systemctl start nginx &>/dev/null
    sleep 3
    #2.等待3秒后再次获取一次Nginx状态
    nginxpid=$(ps -ef | grep [n]ginx | wc -l) 
    #3.再次进行判断, 如Nginx还不存活则停止Keepalived,让地址进行漂移,并退出脚本  
    if [ $nginxpid -eq 0 ];then
        systemctl stop keepalived
    fi
fi

#给脚本增加执行权限
[root@lb01 ~]# chmod +x /root/check_web.sh

2.使用keepalived配置文件调用nginx切换脚本
1）配置抢占式时

#只需要在备节点配置
[root@lb02 ~]# vim /etc/keepalived/keepalived.conf 
global_defs {
    router_id lb02
}

#每5秒执行一次脚本，脚本执行内容不能超过5秒，否则会中断再次重新执行脚本
vrrp_script check_web {
    script "/root/check_web.sh"
    interval 5
}

vrrp_instance VI_1 {
    state BACKUP
    nopreempt
    interface eth0
    virtual_router_id 50
    priority 90
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        10.0.0.3
    }
    #调用计划的脚本
    track_script {
        check_web
    }
}

补充
如果开启防火墙的情况下，怎么保证不脑列。

[root@lb01 ~]# systemctl start firewalld.service 

            
  [root@lb01 ~]# firewall-cmd --add-protocol=vrrp

2）配置非抢占式时

#配置非抢占式时，两边都要配置脚本
[root@lb02 ~]# scp check_web.sh 172.16.1.4:/root

#主节点也要配置
[root@lb01 ~]# cat /etc/keepalived/keepalived.conf 
global_defs {
    router_id lb01
}

vrrp_script check_web {
    script "/root/check_web.sh"
    interval 5
}

vrrp_instance VI_1 {
    state BACKUP
    nopreempt
    interface eth0
    virtual_router_id 50
    priority 100
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        10.0.0.3
    }
    track_script {
        check_web
    }
}

.测试
将VIP所在机器nginx的配置文件修改错误
停止nginx
查看VIP是否切换