Haproxy+keepalived实现负载均衡
Haproxy简介： Haproxy是一个开源的高性能的反向代理或者说是负载均衡服务软件之一，它支持双机热备、虚拟主机、基于TCP和HTTP应用代理等功能。其配置简单，而且拥有很好的对服务器节点的健康检查功能（相当于keepalived健康检查）。Haproxy实现了一种事件驱动, 单一进程模型，此模型支持非常大的并发连接数，HAProxy特别适用于那些负载特大的web站点，这些站点通常又需要会话保持或七层处理。HAProxy运行在当前的硬件上，完全可以支持数以万计的并发连接。并且它的运行模式使得它可以很简单安全的整合进您当前的架构中， 同时可以保护你的web服务器不被暴露到网络上。
Keepalived简介：
Keepalived的作用是检测服务器的状态，如果有一台web服务器宕机，或工作出现故障，Keepalived将检测到，并将有故障的服务器从系统中剔除，同时使用其他服务器代替该服务器的工作，当服务器工作正常后Keepalived自动将服务器加入到服务器群中，这些工作全部自动完成，不需要人工干涉，需要人工做的只是修复故障的服务器。
一、实验环境
(Centos7)
Haporxy1 + keepalived : 10.0.0.128
Haproxy2 + keepalived : 10.0.0.129
VIP1 : 10.0.0.250
Web1 : 10.0.0.130:80
Web2 : 10.0.0.131:80
拓扑图：
二、方案实施
安装及配置web服务(Web1、Web2).
1.安装httpd服务
# yum install -y httpd
# yum install -y httpd
2.修改配置文件中的ServerName,Web1为img.zxb.com,Web2为test.zxb.com
Web1:ServerName img.zxb.com:80
Web2:ServerName test.zxb.com:80
3.编辑网页内容
# echo "web1:10.0.0.130" >/var/www/html/index.html
# echo "web2:10.0.0.131" >/var/www/html/index.html
4.启动web服务
# systemctl start httpd
# systemctl start httpd
5.编辑自身hosts文件
# cat /etc/hosts
10.0.0.130 img.zxb.com
# cat /etc/hosts
10.0.0.131 test.zxb.com
配置Haproxy及keepalived相关配置
1.安装Haproxy及keepalived
# yum install -y keepalived haproxy
# yum install -y keepalived haproxy
2.配置haproxy(两台配置一样)
#vim /etc/haproxy/haproxy.cfg
默认
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
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
添加
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
frontendmain *:80
option forwardforexcept 127.0.0.0/8
rspadd X-via:\ haproxy/www.zxb.com
rspidel Server.*
stats enable #在frontend添加监控页面
stats uri /zxb?stats #自定义监控页面的url
stats realm Stats\ Page\ Area #去掉空格转译
stats auth zxb:zxb #配置帐号密码
stats refresh 5s #每5s刷新一次监控页面
stats hide-version #隐藏版本信息
stats admin if TRUE
maxconn 10000 #定义最大并发为10000
acl url_img hdr(host) -i img.zxb.com
acl url_test hdr(host) -i test.zxb.com
use_backend img_web if url_img
use_backend test_web if url_test
backend img_web
balance roundrobin
server web1 10.0.0.130:80 check
backend test_web
balance roundrobin
server web2 10.0.0.131:80 check
4.启动haproxy
# systemctl start haproxy
# systemctl start haproxy
3.配置keepalived
Master:
#
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
! Configuration File for keepalived
vrrp_script check_haproxy {
script "/etc/keepalived/check_haproxy.sh"
#检测脚本，该路径脚本存放的位置
interval 2
#每隔2秒钟检测一次
# weight 2 #优先级减2
}
global_defs {
#全局配置
notification_email {
#发生事件切换的时候，发送的邮箱
acassen@firewall.loc
}
router_id LVS_DEVEL
}
vrrp_instance VI_2 {
state MASTER
#两种角色,MASTER or BACKUP
interface ens33
#实例绑定的网卡
virtual_router_id 50 #VRID 标记
priority 100 #优先级
advert_int 1
#MASTER 与BACKUP 之间同步检查的时间间隔，单位为秒
authentication {
auth_type PASS
#PASS#认证的方式，支持 PASS
auth_pass 1111
#认证的密码
}
virtual_ipaddress {
10.0.0.250
#指定漂移地址（VIP)
}
track_script {
check_haproxy
}
}
Backup:
#
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
! Configuration File for keepalived
vrrp_script check_haproxy {
script "
/etc/keepalived/check_haproxy.sh
"
interval 2
weight 2
}
global_defs {
notification_email {
acassen@firewall.loc
}
router_id LVS_DEVEL
}
vrrp_instance VI_2 {
state
BACKUP
interface ens33
virtual_router_id 50
priority 80
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
10.0.0.250
}
track_script {
check_haproxy
}
}
4.编辑haproxy检测脚本（两台配置一样）
#cat /etc/keepalived/check_haproxy.sh
#!/bin/bash
haproxy=`ps -C haproxy --no-heading| wc -l`
if [ $haproxy -eq 0 ];then
systemctl stop keepalived
fi
三.测试
1.访问 haproxy 自带的监控页面，url为haproxy服务器的IP地址
url用户名密码为http://10.0.0.128/zxb?stats
https://s5.51cto.com/oss/201710/23/43256cd81eda936dbd6187de04ba6df6.png-wh_500x0-wm_3-wmp_4-s_1025407540.png
2.编辑客户端10.0.0.150:hosts文件
# cat /etc/hosts
10.0.0.100 img.zxb.com test.zxb.com
3.启动Haproxy1和Haproxy2的keepalived,由于Haproxy1的priority较高，所以VIP首先在Haproxy1
# systemctl start keepalived
# systemctl start keepalived
# systemctl status keepalived
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
● keepalived.service - LVS and VRRP High Availability Monitor
Loaded: loaded (/usr/lib/systemd/system/keepalived.service; disabled; vendor preset: disabled)
Active: active (running) since 日 2017-10-22 23:34:43 CST; 30min ago
Process: 17871 ExecStart=/usr/sbin/keepalived $KEEPALIVED_OPTIONS (code=exited, status=0/SUCCESS)
Main PID: 17872 (keepalived)
CGroup: /system.slice/keepalived.service
├─17872 /usr/sbin/keepalived -D
├─17873 /usr/sbin/keepalived -D
└─17874 /usr/sbin/keepalived -D
10月 22 23:58:10 zxb Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
10月 22 23:58:10 zxb Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
10月 22 23:58:10 zxb Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
10月 22 23:58:10 zxb Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
10月 22 23:58:16 zxb Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
10月 22 23:58:16 zxb Keepalived_vrrp: VRRP_Instance(VI_2)
Sending/queueing gratuitous ARPs on ens33 for 10.0.0.250
10月 22 23:58:16 zxb Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
10月 22 23:58:16 zxb Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
10月 22 23:58:16 zxb Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
10月 22 23:58:16 zxb Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
4.访问
# curl img.zxb.com
web1:10.0.0.130
# curl test.zxb.com
web2:10.0.0.131
5.当停止master上的haproxy服务时，keepalived配置文件检测到haproxy脚本后，停止keepalived服务，VIP从Master飘到Backup
# ip addr
1
2
3
4
5
6
7
8
9
10
11
12
13
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1
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 qlen 1000
link/ether 00:0c:29:64:e3:62 brd ff:ff:ff:ff:ff:ff
inet 10.0.0.128/24 brd 10.0.0.255 scope global ens33
valid_lft forever preferred_lft forever
inet6 fe80::8a1e:39ba:9beb:4aa9/64 scope link
valid_lft forever preferred_lft forever
# systemctl status keepalived
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
● keepalived.service - LVS and VRRP High Availability Monitor
Loaded: loaded (/usr/lib/systemd/system/keepalived.service; disabled; vendor preset: disabled)
Active: active (running) since 一 2017-10-23 17:25:23 CST; 16min ago
Process: 3915 ExecStart=/usr/sbin/keepalived $KEEPALIVED_OPTIONS (code=exited, status=0/SUCCESS)
Main PID: 3916 (keepalived)
CGroup: /system.slice/keepalived.service
├─3916 /usr/sbin/keepalived -D
├─3917 /usr/sbin/keepalived -D
└─3920 /usr/sbin/keepalived -D
10月 23 17:41:33 zxb2 Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
10月 23 17:41:33 zxb2 Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
10月 23 17:41:33 zxb2 Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
10月 23 17:41:33 zxb2 Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
10月 23 17:41:38 zxb2 Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
10月 23 17:41:38 zxb2 Keepalived_vrrp: VRRP_Instance(VI_2)
Sending/queueing gratuitous ARPs on ens33 for 10.0.0.250
10月 23 17:41:38 zxb2 Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
10月 23 17:41:38 zxb2 Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
10月 23 17:41:38 zxb2 Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
10月 23 17:41:38 zxb2 Keepalived_vrrp: Sending gratuitous ARP on ens33 for 10.0.0.250
# ip addr
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1
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 qlen 1000
link/ether 00:0c:29:23:e3:43 brd ff:ff:ff:ff:ff:ff
inet 10.0.0.129/24 brd 10.0.0.255 scope global ens33
valid_lft forever preferred_lft forever
inet 10.0.0.250/32 scope global ens33
valid_lft forever preferred_lft forever
inet6 fe80::bcdf:e60f:7858:6933/64 scope link
valid_lft forever preferred_lft forever
6.再访问
# curl img.zxb.com
web1:10.0.0.130
# curl test.zxb.com
web2:10.0.0.131
end
！！！！
页:
[1]