在之前的博文中，我搭建了一个1台HAproxy+2台Apache的WEB负载均衡测试环境，然而1台HAproxy终归存在单点故障的风险，如果能使用2台HAproxy做高可用，形成主备群集，对外提供一个VIP在两台HAproxy之间浮动，一旦主的故障，备机立马接管VIP，保持HAproxy仍旧可用，那就比较完美了，其实很简单，Keepalived就能实现这样的功能，且在RedHat 7的安装光盘中自带了keepalived的RPM包，通过yum安装非常方便。通过在两台HAproxy上分别安装Keepalived，形成一主一备，使用VRRP协议检测心跳信号，VIP可以在两台HAproxy之间浮动，实现HAproxy的高可用群集架构。架构示意图如下：
关于Apache和HAproxy的安装配置可以参考本人之前的博文：
http://jiangjianlong.blog.51cto.com/3735273/1887279
http://jiangjianlong.blog.51cto.com/3735273/1858650
一、测试环境
计算机名 |
操作系统 |
IP地址 |
VIP |
安装的软件 |
RH7-HAproxy01 |
Redhat 7.2 |
192.168.10.19 |
192.168.10.18 |
Haproxy1.5.14、keepalived 1.2.13 |
RH7-HAproxy02 |
Redhat 7.2 |
192.168.10.20 |
Haproxy1.5.14、keepalived 1.2.13 |
|
RH7-Apache01 |
Redhat 7.2 |
192.168.10.21 |
—— |
Apache httpd 2.4 |
RH7-Apache02 |
Redhat 7.2 |
192.168.10.22 |
—— |
Apache httpd 2.4 |
二、安装配置keepalived
1、在两台HAproxy上分别使用yum install keepalived命令安装keepalived
[root@RH7-HAproxy01 ~]# yum install keepalived
[root@RH7-HAproxy02 ~]# yum install keepalived
2、编辑RH7-HAproxy01的keepalived配置文件，将其配置成MASTER
[root@RH7-HAproxy01~]# cd /etc/keepalived/
[root@RH7-HAproxy01keepalived]# ls
keepalived.conf
[root@RH7-HAproxy01keepalived]# cp keepalived.conf keepalived.conf.BAK
[root@RH7-HAproxy01keepalived]# ls
keepalived.conf keepalived.conf.BAK
[root@RH7-HAproxy01keepalived]# vi keepalived.conf
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
|
! Configuration Filefor keepalived
global_defs {
notification_email {
acassen@firewall.loc
failover@firewall.loc
sysadmin@firewall.loc
}
notification_email_fromAlexandre.Cassen@firewall.loc
smtp_server 192.168.200.1
smtp_connect_timeout 30
router_id LVS_DEVEL
}
#监测haproxy进程状态，每2秒执行一次
vrrp_scriptchk_haproxy {
script
"/etc/keepalived/chk_haproxy.sh"
interval 2
weight 2
fall 3
}
vrrp_instance VI_1 {
state MASTER
#将HR7-HAproxy01标示为MASTER
interface eno50332208
#指定网络接口
virtual_router_id 51
priority 100
#MASTER权重要高于BACKUP
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
192.168.10.18
#VIP
}
track_script {
chk_haproxy
#监测haproxy进程状态
}
}
|
3、编辑RH7-HAproxy02的keepalived配置文件，将其配置成BACKUP
[root@RH7-HAproxy02 ~]# cd /etc/keepalived/
[root@RH7-HAproxy02 keepalived]# ls
keepalived.conf
[root@RH7-HAproxy02 keepalived]# cp keepalived.confkeepalived.conf.BAK
[root@RH7-HAproxy02 keepalived]# ls
keepalived.conf keepalived.conf.BAK
[root@RH7-HAproxy01keepalived]# vi keepalived.conf
[root@RH7-HAproxy02~]# vi /etc/keepalived/keepalived.conf
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
|
global_defs {
notification_email {
acassen@firewall.loc
failover@firewall.loc
sysadmin@firewall.loc
}
notification_email_fromAlexandre.Cassen@firewall.loc
smtp_server 192.168.200.1
smtp_connect_timeout 30
router_id LVS_DEVEL
}
#监测haproxy进程状态，每2秒执行一次
vrrp_scriptchk_haproxy {
script
"/etc/keepalived/chk_haproxy.sh"
interval 2
weight 2
}
vrrp_instance VI_1 {
state BACKUP
#将HR7-HAproxy02标识为BACKUP
interface eno50332208
#指定网络接口
virtual_router_id 51
priority 99
#MASTER权重要高于BACKUP
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
192.168.10.18
#VIP
}
track_script {
chk_haproxy
#监测haproxy进程状态
}
}
|
4、在2台HAproxy上创建chk_haproxy.sh脚本，内容如下：
vi /etc/keepalived/chk_haproxy.sh
|
1
2
3
4
5
6
7
8
9
10
11
12
13
|
#!/bin/bash
status=$(
ps
aux|grephaproxy |
grep
-
v
grep
|
grep
-
v
bash
|
wc
-l)
if
[
"${status}"
=
"0"
];
then
#判断haproxy是否已经启动
systemctl start haproxy
#如果没有启动，则启动haproxy程序
sleep
3
#睡眠3秒钟以等待haproxy完全启动
status2=$(psaux|
grep
haproxy |
grep
-
v
grep
|
grep
-
v
bash
|
wc
-l)
if
[
"${status2}"
=
"0"
];
then
systemctl stop keepalived
#如果haproxy还是没有启动，则将keepalived停掉，这样VIP会自动漂移到另外一台haproxy
fi
fi
|
5、对chk_haproxy.sh脚本赋予执行权限
6、重启RH7-HAproxy01的keepalived服务，并查看IP，确认VIP成功绑定
7、重启RH7-HAproxy02的keepalived服务，并查看IP，确认无VIP绑定
8、设置keepalived服务开机自动启动
9、使用浏览器访问VIP确认能正常访问到后端的WEB页面，负载均衡功能也正常
本文出自江健龙的技术博客http://jiangjianlong.blog.51cto.com/3735273/1905149
三、keepalived主备切换测试
1、停止haproxy服务测试chk_haproxy.sh脚本，从日志可以看出一停止haproxy服务马上就自动启动起来了，chk_haproxy.sh脚本成功生效
root@RH7-HAproxy01 ~]# systemctl stop haproxy
[root@RH7-HAproxy01 ~]# tail -f /var/log/messages
2、停止keepalived服务测试主备切换，通过日志可以看到RH7-HAproxy01的keepalived服务停止后释放了VIP，而RH7-HAproxy02马上进入MASTER状态并绑定了VIP，通过浏览器进行访问VIP发现仍然正常访问
[root@RH7-HAproxy01 ~]# tail -f /var/log/messages
[root@RH7-HAproxy02 ~]# tail -f /var/log/messages
3、将RH7-HAproxy01的keepalived服务启动，由于它原本是配置成MASTER，一旦服务起来后，马上接管VIP，我们观察日志也可以很直观地看出来
[root@RH7-HAproxy01 ~]# tail -f /var/log/messages
[root@RH7-HAproxy02 ~]# tail -f /var/log/messages
4、直接将RH7-HAproxy01关机，观察RH7-HAproxy02的日志也可以看出VIP的转移过程
[root@RH7-HAproxy01 ~]#poweroff
[root@RH7-HAproxy02 ~]# tail -f /var/log/messages
本文出自江健龙的技术博客http://jiangjianlong.blog.51cto.com/3735273/1905149
关于Apache和HAproxy的安装配置可以参考本人之前的博文：
http://jiangjianlong.blog.51cto.com/3735273/1887279
http://jiangjianlong.blog.51cto.com/3735273/1858650
本文转自Mr大表哥jianlong1990 博客，原文链接： http://blog.51cto.com/jiangjianlong/1905149 如需转载请自行联系原作者
