haproxy的高可用

一、简介

软件负载均衡一般通过两种方式来实现：基于操作系统的软负载实现和基于第三方应用的软负载实现。

LVS 就是基于 Linux 操作系统实现的一种软负载，HAProxy就是开源的并且基于第三应用实现的软负载。HAProxy 相比 LVS 的使用要简单很多，功能方面也很丰富。当前，HAProxy 支持两种主要的代理模式:"tcp"即 4 层（大多用于邮件服务器、内部协议通信服务器等）和 7 层（HTTP）在 4 层模式下， HAproxy仅在客户端和服务器之间转发双向流量。 7 层模式下， HAProxy 会分析协议，并且能通过允许、拒绝、交换、增加、修改或者删除请求(request)或者回应(response)里指定内容来控制协议，这种操作要基于特定规则。

详情可以HAProxy 官方网站(http://haproxy.1wt.eu)可以下载配置说明文档(configuration.txt)和架构文件(architecture.txt)作为参考。

二、拓扑图

三、配置过程

注：

OS：Centos 6.5x86_64

己经安装的包组 :

1

#yum groupinstall -y "Development tools" "Server Platform Development"

前提：

HAproxy A与B要做到

主机名解析

时间同步

无**登录

1、HAproxy A配置

安装keepalived、haproxy

1

#yum install -y keepalived haproxy

配置keepalived

编辑/etc/keepalived/keepalived.conf

!  Configuration File for keepalived

global_defs {

   notification_email {

         [email protected]

         [email protected]

   }

   notification_email_from [email protected]

   smtp_connect_timeout 3

   smtp_server 127.0.0.1

   router_id LVS_DEVEL
}
vrrp_script chk_haproxy {

    script "killall -0 haproxy"

    interval 1

    weight 2
}
vrrp_instance VI_1 {

    interface eth0

    state MASTER

    priority 201

    virtual_router_id 109

    garp_master_delay 1

    authentication {

        auth_type PASS

        auth_pass password

    }

    track_interface {

       eth0

    }

    virtual_ipaddress {

        172.16.1.103/16 dev eth0 label eth0:0

    }

    track_script {

        chk_haproxy

    }

    notify_master "/etc/keepalived/notify.sh master"  

    notify_backup "/etc/keepalived/notify.sh backup"

    notify_fault "/etc/keepalived/notify.sh fault"
}
vrrp_instance VI_2 {

    interface eth0

    state BACKUP

    priority 99

    virtual_router_id 52

    garp_master_delay 1

    authentication {

        auth_type PASS

        auth_pass password

    }

    track_interface {

       eth0

    }

    virtual_ipaddress {

        172.16.1.109/16 dev eth0 label eth0:1

    }

    track_script {

        chk_haproxy

    }
}

通知脚本

#!/bin/bash
# description: An example of notify script
#
vip=172.16.1.103

contact='[email protected]'
notify() {

    mailsubject="`hostname` to be $1: $vip floating"

    mailbody="`date '+%F\ %T'`: vrrp transition, `hostname` changed to be $1"

    echo $mailbody | mail -s "$mailsubject" $contact
}

case "$1" in

    master)

        notify master

        /etc/rc.d/init.d/haproxy start

        exit 0

    ;;

    backup)

        notify backup

        /etc/rc.d/init.d/haproxy stop

        exit 0

    ;;

    fault)

        notify fault

        /etc/rc.d/init.d/haproxy stop

        exit 0

    ;;

    *)

        echo 'Usage: `basename $0` {master|backup|fault}'

        exit 1

    ;;
esac
#chmod +x /etc/keepalived/notify.sh

配置haproxy

编辑配置文件 /etc/haproxy/haproxy.cfg

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
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78

#---------------------------------------------------------------------
# Example configuration for a possible web application.  See the
# full configuration options online.
#
#   http://haproxy.1wt.eu/download/1.4/doc/configuration.txt
#
#---------------------------------------------------------------------
#---------------------------------------------------------------------
# Global settings
#---------------------------------------------------------------------

global #全局配置区域

    log         127.0.0.1 local2  #日志将通过rsyslog进行归档记录

    chroot      /var/lib/haproxy #运行的安装路径

    pidfile     /var/run/haproxy.pid #pid文件存放的位置

    maxconn     4000  #最大连接

    user        haproxy  #运行haproxy的用户

    group       haproxy  #运行haprixy的组

    daemon     #以后台模式运行haproxy

    # turn on stats unix socket

    stats socket /var/lib/haproxy/stats
#---------------------------------------------------------------------
# common defaults that all the 'listen' and 'backend' sections will
# use if not designated in their block
#---------------------------------------------------------------------
defaults

    mode                    http  #工作模式

    log                     global #记录日志

    option                  httplog

    option                  dontlognull #不记录健康检查的日志信息

    option http-server-close  #启用服务器端主动关闭

    option forwardfor       except 127.0.0.0/8 #传递客户端IP

    option                  redispatch #当后端服务器组中的某一台主机故障后，能够自动将请求重定向到组内的其它主机

    retries                 3 #请求重试次数

    timeout http-request    10s #http请求超时时间

    timeout queue           1m #一个请求在队列里的超时时间

    timeout connect         10s #连接服务器超时时间

    timeout client          1m #客户端超时时间

    timeout server          1m #客户端超时时间

    timeout http-keep-alive 10s

    timeout check           10s #心跳检测超时时间

    maxconn                 3000 #最大连接数
#---------------------------------------------------------------------
# main frontend which proxys to the backends
#---------------------------------------------------------------------
frontend  proxy *:80

    acl url_static       path_beg       -i /static /images /javascript /stylesheets

    acl url_static       path_end       -i .jpg .gif .png .css .js

    use_backend static          if url_static

    default_backend             dynamic
#---------------------------------------------------------------------
# static backend for serving up images, stylesheets and such
#---------------------------------------------------------------------

backend static  #后端调度

    balance     roundrobin #调度算法

    server      web2 192.168.1.108:80 inter 1500 rise 2 fall 3  check maxconn 5000
#----------------------------------------
listen statistics

    mode http  # http 7 层模式

    bind *:8080 #监听地址

    stats enable #启用状态监控

    stats auth admin:essun #验证的用户与密码

    stats uri /admin?status #访问路径

    stats admin if TRUE #如果验证通过了就允许登录

    stats refresh 6s #每6秒刷新一次

    acl allow src 172.16.1.0/24 #允许的IP地址

    tcp-request content accept if allow #如果允许的地址段就允许访问

    tcp-request content reject #拒绝非法连接
#---------------------------------------------------------------------
# round robin balancing between the various backends
#---------------------------------------------------------------------
backend dynamic

    balance     roundrobin

    server      web2 192.168.1.40:80 check inter 1500 rise 2 fall 3   maxconn 5000
#服务器定义，serverid为web2，check inter 1500是检测心跳频率
#rise 2是2次正确认为服务器可用
#fall 3是3次失败认为服务器不可用
#最大连接数据为5000
server  web3 192.168.1.104:80 check inter 1500 rise 2 fall 3   maxconn 5000

将此文件同时也复制到HAproxy B上一份

2、HAproxy B的配置

安装keepalived、haproxy

1

#yum install -y keepalived haproxy

修改keepalived在配置文件（/etc/keepalived/keepalived.conf）

! Configuration File for keepalived

global_defs {

   notification_email {

         [email protected]

         [email protected]

   }

   notification_email_from [email protected]

   smtp_connect_timeout 3

   smtp_server 127.0.0.1

   router_id LVS_DEVEL
}
vrrp_script chk_haproxy {

    script "killall -0 haproxy"

    interval 1

    weight 2
}
vrrp_instance VI_1 {

    interface eth0

    state BACKUP

    priority 200

    virtual_router_id 109

    garp_master_delay 1

    authentication {

        auth_type PASS

        auth_pass password

    }

    track_interface {

       eth0

    }

    virtual_ipaddress {

        172.16.1.103/16 dev eth0 label eth0:0

    }

    track_script {

        chk_haproxy

    }

}
vrrp_instance VI_2 {

    interface eth0

    state MASTER

    priority 100

    virtual_router_id 52

    garp_master_delay 1

    authentication {

        auth_type PASS

        auth_pass password

    }

    track_interface {

       eth0

    }

    virtual_ipaddress {

        172.16.1.109/16 dev eth0 label eth0:1

    }

    track_script {

        chk_haproxy

    }

    notify_master "/etc/keepalived/notify.sh master"  

    notify_backup "/etc/keepalived/notify.sh backup"

    notify_fault "/etc/keepalived/notify.sh fault"
}

修改通知脚本

#!/bin/bash
# description: An example of notify script
#

vip=172.16.1.109

contact='[email protected]'
notify() {

    mailsubject="`hostname` to be $1: $vip floating"

    mailbody="`date '+%F\ %T'`: vrrp transition, `hostname` changed to be $1"

    echo $mailbody | mail -s "$mailsubject" $contact
}

case "$1" in

    master)

        notify master

        /etc/rc.d/init.d/haproxy start

        exit 0

    ;;

    backup)

        notify backup

        /etc/rc.d/init.d/haproxy stop

        exit 0

    ;;

    fault)

        notify fault

        /etc/rc.d/init.d/haproxy stop

        exit 0

    ;;

    *)

        echo 'Usage: `basename $0` {master|backup|fault}'

        exit 1

    ;;
esac
#chmod +x /etc/keepalived/notify.sh

由于HAproxy A中的haporxy配置与HAporxy B 的配置文件相同从HAproxy A中发过来一份放在同一目录下即可

1

#scp -p /etc/haproxy/haproxy.cnf 192.168.1.109:/etc/haproxy/

3、测试一下keepalived功能

HAproxy B 上面的ip地址

将ha2上的keepalived停止后，ip地址己经转移到了ha1上了

当ha2启动后，172.16.1.109还是会回到ha2上面。

4、安装后端的web服务

web1 静态页面（192.168.1.108）

#yum install -y httpd

[[email protected] ~]# cd /var/www/html/

[[email protected] html]# echo "<h1>这是一个静态页面，地址为192.168.1.108</h1>" > index.html
[[email protected] html]# ll

total 4

-rw-r--r-- 1 root root 59 May  3 12:50 index.html
[[email protected] html]# cat index.html

<h1>这是一个静态页面，地址为192.168.1.108</h1>
[[email protected] html]# service httpd start

Starting httpd: httpd: apr_sockaddr_info_get() failed for essun.node3.com

httpd: Could not reliably determine the server's fully qualified domain name, using 127.0.0.1 for ServerName

                                                           [  OK  ]

[[email protected] html]# curl http://192.168.1.108

<h1>这是一个静态页面，地址为192.168.1.108</h1>

在/var/www/html中放一张图片，仅供测试

web2 动态页面 (192.168.1.40)

[[email protected] yum.repos.d]# yum install -y  httpd php php-mysql mysql-server mysql-devel

[[email protected] yum.repos.d]# cd /var/www/html/

[[email protected] html]# vim index.php

[[email protected] html]# service httpd restart
Stopping httpd:                                            [FAILED]

Starting httpd: httpd: apr_sockaddr_info_get() failed for essun.node4.com

httpd: Could not reliably determine the server's fully qualified domain name, using 127.0.0.1 for ServerName

                                                           [  OK  ]

[[email protected] html]# cat index.php

<h1>我是动态页面，地址是192.168.1.40</h1>
<?php

    phpinfo();
?>

[[email protected] html]# curl  -I http://192.168.1.40/index.php

HTTP/1.1 200 OK
Date: Sat, 03 May 2014 05:11:47 GMT

Server: Apache/2.2.15 (CentOS)

X-Powered-By: PHP/5.3.3
Connection: close

Content-Type: text/html; charset=UTF-8

web3 动态页面（192.168.1.104）

[[email protected] yum.repos.d]# yum install -y  httpd php php-mysql mysql-server mysql-devel

[[email protected] yum.repos.d]# cd /var/www/html/

[[email protected] html]# vim index.php

[[email protected] yum.repos.d]# service httpd restart
Stopping httpd:                                            [FAILED]
Starting httpd:                                            [  OK  ]

[[email protected] html]# cat index.php

<h1>我也是动态页面，地址是192.168.1.104</h1>
<?php

    phpinfo();
?>

[[email protected] yum.repos.d]# curl -I http://192.168.1.104

HTTP/1.1 200 OK
Date: Sat, 03 May 2014 05:14:22 GMT

Server: Apache/2.2.15 (CentOS)

X-Powered-By: PHP/5.3.3
Connection: close

Content-Type: text/html; charset=UTF-8

四、测试

1、利用两个vip任意一个测试一下

静态页面测试

动态页面测试web2 (192.168.1.40)

动态页面测试 web3 （192.168.1.104）

监控页面，验证用户身份

验证通过后

其中一个keepalived宕机后完不会影响到服务的正常的运行

本文转自 jinlinger 51CTO博客，原文链接：http://blog.51cto.com/essun/1405893，如需转载请自行联系原作者