keepalived+haproxy实现高可用负载均衡集群

2019独角兽企业重金招聘Python工程师标准>>>

keepalived+haproxy 实现高可用、负载均衡服务器集群

实验环境：iptables和selinux关闭

前端高可用负载均衡调度主机：vm7 172.25.28.7   vm8 172.25.28.8

后台应用服务器：                   vm1 172.25.28.1   vm2 172.25.28.4

配置如下：#######################################################################

分别在两台调度主机上面修改/etc/haproxy/haproxy.cf

global

log         127.0.0.1 local2

chroot      /var/lib/haproxy

pidfile     /var/run/haproxy.pid

maxconn     4000

user        haproxy

group       haproxy

daemon

stats socket /var/lib/haproxy/stats

defaults

mode                    http

log                     global

option                  httplog

option                  dontlognull

option http-server-close

option forwardfor       except 127.0.0.0/8

option                  redispatch

retries                 3

timeout http-request    10s

timeout queue           1m

timeout connect         10s

timeout client          1m

timeout server          1m

timeout http-keep-alive 10s

timeout check           10s

maxconn                 3000

listen stats_auth *:8080

stats enable

stats uri /status

stats auth admin:westos

stats refresh 5s

listen www.westos.org *:80

balance roundrobin

server web1 172.25.28.1:80 check weight 3

server web2 172.25.28.4:80 check weight 1

server backup 127.0.0.1:8081 backup

两台调度主机：分别修改/etc/keepalived/keepalived.conf

MASTER节点:

! Configuration File for keepalived

vrrp_script check_haproxy {      #使用haproxy_check 脚本对后台服务器做一个健康检查

script "/opt/haproxy_check"    #此处目录为脚本的绝对路径

interval 5

weight 2

}

global_defs {

notification_email {

root@localhost

}

notification_email_from keepalived@vm7.example.com

smtp_server 127.0.0.1

smtp_connect_timeout 30

router_id LVS_DEVEL

}

vrrp_instance VI_1 {

state MASTER     #备机的为BACKUP

interface eth0

virtual_router_id 51

priority 100                      #两台主备调度器的值设置必须不一样

advert_int 1

authentication {

auth_type PASS

auth_pass 1111

}

virtual_ipaddress {

172.25.28.100   #对外虚拟ip

}

track_script {

check_haproxy

}

}

BACKUP:

! Configuration File for keepalived

vrrp_script check_haproxy {

script "/opt/haproxy_check"

interval 5

weight 2

}

global_defs {

notification_email {

root@localhost

}

notification_email_from keepalived@vm7.example.com

smtp_server 127.0.0.1

smtp_connect_timeout 30

router_id LVS_DEVEL

}

vrrp_instance VI_1 {

state BACKUP     #注意修改此处

interface eth0

virtual_router_id 51

priority 50     注意修改此处

advert_int 1

authentication {

auth_type PASS

auth_pass 1111

}

virtual_ipaddress {

172.25.28.100     #对外虚拟ip

}

track_script {

check_haproxy

}

}

分别在两台调度主机上面操作：

添加检测后台服务器健康状况的脚本：在/etc/keepalived/keepalived.conf 中的对应目录下添加文件/opt/haproxy_check

#!/bin/bash

/etc/init.d/haproxy status &> /dev/null || /etc/init.d/haproxy restart &> /dev/null

if [ $? -ne 0 ];then

/etc/init.d/keepalived stop &> /dev/null

fi

测试：分别在两台后台应用服务器上面开启httpd服务，并写上首页文件。 并且要在安装浏览器的主机的/etc/hosts下面写解析。

1.高可用测试：停止master上的keepalived服务，查看backup是否会接管并且根据脚本会不会自动启动haproxy服务

2.负载均衡测试：访问www.westos.org，看到页面在两个后台服务器上切换表示成功！

3.故障切换测试：任意关闭后台服务器上面的httpd服务，查看是否haproxy可以及时发现故障主机，并且将服务转移到正常节点上来执行。

转载于:https://my.oschina.net/u/2368584/blog/479954