HAProxy+KeepAlived实现web服务高可用、动静分离等
时间：2014-05-03 14:37:46
收藏：0
阅读：610
大致规划：
| 主机 | IP | 描述 |
| VIP | 192.168.0.222 | 对外提供高可用IP |
| haproxy+keepalived (node1) | 192.168.0.111 | haproxy为后端两台WEB服务的做动静分离；keepalived为haproxy做高可用。 |
| haproxy+keepalived (node2) | 192.168.0.112 | |
| WEB (node3) | 192.168.0.113 | 提供静态请求响应 |
| Apache+PHP+MySQL (node4) | 192.168.0.114 | 提供动态请求响应 |
一、KeepAlived的简介以原理
Keepalived的作用是检测web服务器的状态，如果有一台web服务器死机，或工作出现故障，Keepalived将检测到，并将有故障的web服务器从系统中剔除，当web服务器工作正常后Keepalived自动将web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web服务器。
Layer3,4&7工作在IP/TCP协议栈的IP层，TCP层，及应用层,原理分别如下：
Layer3：Keepalived使用Layer3的方式工作式时，Keepalived会定期向服务器群中的服务器发送一个ICMP的数据包（既我们平时用的Ping程序）,如果发现某台服务的IP地址没有激活，Keepalived便报告这台服务器失效，并将它从服务器群中剔除，这种情况的典型例子是某台服务器被非法关机。Layer3的方式是以服务器的IP地址是否有效作为服务器工作正常与否的标准。在本文中将采用这种方式。
Layer4:如果您理解了Layer3的方式，Layer4就容易了。Layer4主要以TCP端口的状态来决定服务器工作正常与否。如web server的服务端口一般是80，如果Keepalived检测到80端口没有启动，则Keepalived将把这台服务器从服务器群中剔除。
Layer7：Layer7就是工作在具体的应用层了，比Layer3,Layer4要复杂一点，在网络上占用的带宽也要大一些。Keepalived将根据用户的设定检查服务器程序的运行是否正常，如果与用户的设定不相符，则Keepalived将把服务器从服务器群中剔除。
--引用百度百科
二、安装相关软件
WEB和LAMP搭建这里不做赘述；安装完成后可以测试是否正常访问。
测试两台机器正常访问。
为node1和node2各自安装keepalived和haproxy；为了简便；yum安装即可
[root@node1 ~]# rpm -q keepalived haproxy
keepalived-1.2.7-3.el6.x86_64
haproxy-1.4.24-2.el6.x86_64
-----------------------------------------------
[root@node2 ~]# rpm -q keepalived haproxy
keepalived-1.2.7-3.el6.x86_64
haproxy-1.4.24-2.el6.x86_64
[root@node2 ~]#三、配置haproxy实现动静分离
这里在node1上演示；node2同样的配置：
[root@node1 ~]# vim /etc/haproxy/haproxy.cfg
#这里对应global段和default段不做任何修改；但是日志需要在global段开启；
#其余的全部注释；任何重新添加以下内容；
#具体含义上一篇以做解释；这里就不做详细介绍
frontend web
bind *:80
acl url_static path_beg -i /static /images /javascript /stylesheets
acl url_static path_end -i .jpg .gif .png .css .js .html .htm
acl url_dynamic path_end -i .php
use_backend static if url_static
use_backend dynamic if url_dynamic
default_backend static
backend static
balance roundrobin
server node3 192.168.0.113:80 check
backend dynamic
balance roundrobin
server node4 192.168.0.114:80 check
listen stats
mode http
bind *:1234
stats enable
stats refresh 3s
stats hide-version
stats uri /admin?stats
stats realm HAProxy\ Statistics
stats auth admin:haproxy
stats admin if TRUE
-------------------------------------------------------------------
#复制一份到node2即可；各自启动haproxy
[root@node1 ~]# scp /etc/haproxy/haproxy.cfg node2:/etc/haproxy/
haproxy.cfg 100% 3896 3.8KB/s 00:00
[root@node1 ~]#查看stats信息：
如后端有多台机器；则可以实现负载均衡；这里未做实例。
四、配置keepalived实现haproxy的高可用
同样在node1上配置；但是node2需要更改两个地方：
[root@node1 ~]# vim /etc/keepalived/keepalived.conf
! Configuration File for keepalived
global_defs { #全局配置
notification_email {
root@localhost #接受邮件方
}
notification_email_from soul@localhost #发件人
smtp_server 127.0.0.1 #邮件服务器
smtp_connect_timeout 30 #超时时长
router_id LVS_DEVEL #ID；随意即可
}
vrrp_script chk_down { #添加监控规则
script "[[ -f /root/down ]] && exit 1 || exit 0"
#上面说明如果在/roo/下有down文件；就失败；转移到从上；否则正常
interval 1 #多久监控一次
weight 2 #权重
}
vrrp_script chk_haproxy { #监控haproxy服务的
script "pidof haproxy &> /dev/null && exit 0 || exit 1"
interval 1
weight 2
}
vrrp_instance VI_1 { #添加一个实例
state MASTER #定义主从 #注意node2从的上为BACKUP
interface eth0 #网络接口
virtual_router_id 222 #虚拟路由ID；根据该ID生成虚拟MAC；保证其唯一性
priority 100 #优先级别；BACKUP要低于MASTER
advert_int 1 #心跳广播间隔
authentication { #认证
auth_type PASS #明文认证
auth_pass 1111 #password
}
virtual_ipaddress { #虚拟IP
192.168.0.222
}
track_script { #脚本追踪；生效上述定义脚本
chk_down
chk_haproxy
}
notify_master "/etc/keepalived/notify.sh master" #邮件通知
notify_backup "/etc/keepalived/notify.sh backup"
notify_fault "/etc/keepalived/notify.sh fault"
}
#更改以上信息；其余配置文件中的信息全部注释即可
---------------------------------------------------------
#复制一份到node2；根据上述描述进行更改提供haproxy脚本：
[root@node1 ~]# vim /etc/keepalived/notify.sh
#!/bin/bash
# Author: MageEdu <linuxedu@foxmail.com>
# description: An example of notify script
#
vip=192.168.0.222
contact=‘root@localhost‘
notify() { #定义发送邮件格式等信息
mailsubject="`hostname` to be $1: $vip floating"
mailbody="`date ‘+%F %H:%M:%S‘`: vrrp transition, `hostname` changed to be $1"
echo $mailbody | mail -s "$mailsubject" $contact
}
case "$1" in
master) #根据参数进行执行对应命令
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
#完成后复制一份到node2即可
#启动keepalived即可五、测试haproxy的高可用
[root@node1 ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UNKNOWN qlen 1000
link/ether 00:0c:29:5e:1e:4f brd ff:ff:ff:ff:ff:ff
inet 192.168.0.111/16 brd 192.168.255.255 scope global eth0
inet 192.168.0.222/32 scope global eth0
inet6 fe80::20c:29ff:fe5e:1e4f/64 scope link
valid_lft forever preferred_lft forever
#查看虚拟IP目前在node1上；
测试访问正常；下面down掉node1看下；
#由于配置文件定义了一个down的脚本；直接建立文件即可
[root@node1 ~]# touch down
[root@node1 ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UNKNOWN qlen 1000
link/ether 00:0c:29:5e:1e:4f brd ff:ff:ff:ff:ff:ff
inet 192.168.0.111/16 brd 192.168.255.255 scope global eth0
inet6 fe80::20c:29ff:fe5e:1e4f/64 scope link
valid_lft forever preferred_lft forever
You have new mail in /var/spool/mail/root #这里也提示有邮件信息
------------------------------------------------------------------
[root@node2 ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:df:70:b6 brd ff:ff:ff:ff:ff:ff
inet 192.168.0.112/16 brd 192.168.255.255 scope global eth0
inet 192.168.0.222/32 scope global eth0
#测试已转移到node2上同时测试页面也是正常的。
测试haproxy脚本是否能够执行
[root@node1 ~]# service haproxy stop
Stopping haproxy: [ OK ]
[root@node1 ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UNKNOWN qlen 1000
link/ether 00:0c:29:5e:1e:4f brd ff:ff:ff:ff:ff:ff
inet 192.168.0.111/16 brd 192.168.255.255 scope global eth0
inet6 fe80::20c:29ff:fe5e:1e4f/64 scope link
----------------------------------------------------------------------
[root@node2 ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:df:70:b6 brd ff:ff:ff:ff:ff:ff
inet 192.168.0.112/16 brd 192.168.255.255 scope global eth0
inet 192.168.0.222/32 scope global eth0
---------------------------------------------------------------------
[root@node1 ~]# tail -5 /var/log/messages
May 2 18:25:10 node1 Keepalived_vrrp[5331]: VRRP_Script(chk_haproxy) failed
May 2 18:25:10 node1 Keepalived_vrrp[5331]: VRRP_Instance(VI_1) Received higher prio advert
May 2 18:25:10 node1 Keepalived_vrrp[5331]: VRRP_Instance(VI_1) Entering BACKUP STATE
May 2 18:25:10 node1 Keepalived_vrrp[5331]: VRRP_Instance(VI_1) removing protocol VIPs.
May 2 18:25:10 node1 Keepalived_healthcheckers[5330]: Netlink reflector reports IP 192.168.0.222 removed
#可以查看系统日志也记录详细信息
#测试启动haproxy后；虚拟IP会自动转回。到此；HAProxy+KeepAlived配置以全部完成；相关功能也都已测试成功。
如有错误；恳请纠正。
本文出自 “Soul” 博客，请务必保留此出处http://chenpipi.blog.51cto.com/8563610/1405554
评论（0）