一、keepalived高可用介绍

解决单点故障，提升服务高可用

基于VRRP协议设计

原理：
将多个物理设备放入到一个VRRP组中，通过VRRP组形成浮动IP，通过优先级进行主备选举，优先级高为主，浮动IP会配置到主设备上，同时主设备会发送心跳以通知自己的状态，备设备连续一段时间接收不到心跳，则认为主宕机，会自动接替主的工作

心跳：
默认以组播的方式发送心跳，地址224.0.0.18

影响心跳的因素：
1、防火墙
2、时间不同步
3、网络堵塞，多块网卡分担流量

二、keepalived高可用设计

1、两台haproxy负载均衡器配置一致

客户端通过DNS解析两台haproxy都可正常访问 业务

2、在haproxy上分别安装keepalived

[root@master_haproxy ~]# yum install -y keepalived
[root@slave_haproxy ~]# yum install -y keepalived

3、配置keepalived实现高可用

规划浮动IP： 192.168.140.100

主：

[root@master_haproxy ~]# cat /etc/keepalived/keepalived.conf
! Configuration File for keepalived

global_defs {
   router_id master_haproxy
}

vrrp_instance haproxy {
    state MASTER
    interface ens33
    virtual_router_id 88
    priority 100
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass redhat
    }
    virtual_ipaddress {
        192.168.140.100
    }
}

备：

[root@slave_haproxy ~]# cat /etc/keepalived/keepalived.conf
! Configuration File for keepalived

global_defs {
   router_id slave_haproxy
}

vrrp_instance haproxy {
    state BACKUP
    interface ens33
    virtual_router_id 88
    priority 50
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass redhat
    }
    virtual_ipaddress {
        192.168.140.100
    }
}

4、分别在两个调度器上查看浮动IP

主：

[root@master_haproxy ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 00:0c:29:cc:6b:2f brd ff:ff:ff:ff:ff:ff
    inet 192.168.140.10/24 brd 192.168.140.255 scope global noprefixroute ens33
       valid_lft forever preferred_lft forever
    inet 192.168.140.100/32 scope global ens33
       valid_lft forever preferred_lft forever
    inet6 fe80::20c:29ff:fecc:6b2f/64 scope link 
       valid_lft forever preferred_lft forever

正常情况下，浮动只会出现的优先级高的设备上

备：

[root@slave_haproxy ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 00:0c:29:29:e5:43 brd ff:ff:ff:ff:ff:ff
    inet 192.168.140.11/24 brd 192.168.140.255 scope global noprefixroute ens33
       valid_lft forever preferred_lft forever
    inet6 fe80::20c:29ff:fe29:e543/64 scope link 
       valid_lft forever preferred_lft forever

5、测试客户端通过浮动IP可正常访问业务

C:\Users\admin>ping blog.linux.com

正在 Ping blog.linux.com [192.168.140.100] 具有 32 字节的数据:
来自 192.168.140.100 的回复: 字节=32 时间<1ms TTL=64
来自 192.168.140.100 的回复: 字节=32 时间<1ms TTL=64
来自 192.168.140.100 的回复: 字节=32 时间=1ms TTL=64

6、模拟故障，将优先级高的设备停机，测试客户端还可正常访问

7、配置脚本，检测虚拟服务故障，实现浮动IP转移

[root@master_haproxy ~]# cat /etc/keepalived/check_haproxy.sh 
#!/bin/bash
#

netstat -tunlp | grep haproxy &> /dev/null

if [ $? -ne 0 ]; then
   killall  keepalived
fi

[root@master_haproxy ~]# ls -l /etc/keepalived/
total 8
-rwxr-xr-x 1 root root 107 Jun 12 11:04 check_haproxy.sh
-rw-r--r-- 1 root root 469 Jun 12 10:59 keepalived.conf

[root@master_haproxy ~]# cat /etc/keepalived/keepalived.conf 

..........
vrrp_script check_haproxy {      // 定义外部脚本 
   script "/etc/keepalived/check_haproxy.sh"
   interval 1
}

vrrp_instance haproxy {
   ..............
    track_script {      // 调用外部脚本 
       check_haproxy
    }
}

测试：关掉虚拟服务，测试浮动IP的转移