使用nginx+keepalived做双机热备+负载均衡

一、keepalived作用说明
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。

二、keepalived的安装
我们采用源代码的方式来安装keepalived.

cd /usr/local/src
wget

tar -xvf keepalived-1.2.20.tar.gz
cd keepalived-1.2.20
./configure --prefix=/usr/local/keepalived
make
make install

1.建立服务启动脚本，以便使用service命令控制；
(1)复制文件
cp /usr/local/keepalived/etc/rc.d/init.d/keepalived  /etc/init.d/keepalived
(2)添加执行权限
chmod +x /etc/init.d/keepalived

2.修改/etc/init.d/keepalived文件
将/etc/sysconfig/keepalived修改为.  /usr/local/keepalived/etc/sysconfig/keepalived

3.在上述行下添加内容(将keepalived主程序所在路径导入到环境变量PATH中)
PATH="$PATH:/usr/local/keepalived/sbin"
export PATH

4.修改/usr/local/keepalived/etc/sysconfig/keepalived文件，设置正确的服务启动参数
KEEPALIVED_OPTIONS="-D -f /usr/local/keepalived/etc/keepalived/keepalived.conf"

5.至此，keepalived配置完毕，启动
service keepalived restart

6.设置服务为开机启动
chkconfig keepalived on

三、keepalived的配置

1.主机说明
本次共涉及到4台主机。
主机1(安装nginx+keepalived)为主负载均衡机器，IP：192.168.109.210
主机2(安装nginx+keepalived)为备负载均衡机器，IP：192.168.109.207
主机3(安装apache)为Web服务器，IP：192.168.109.208
主机4(安装apache)为Web服务器，IP：192.168.109.209

2.搭建说明

(1)首先使用主机1为负载均衡机器(停掉主机2上的nginx),并启动主机3、4上的apache。
(2)通过配置主机1上nginx，使其实现负载均衡，让所有请求均匀分配到主机3、4上来处理请求；
(3)在实现负载均衡后，为了实现服务持续可用，需对负载均衡器配置主从切换。当主机1出现故障时，通过keepalived迅速切换到主机2上。
(4)主机3和主机4上分别在/var/www/html目录下放置有页面index.html,页面内容为

web1.test.com

web2.test.com

3.搭建步骤

(1)配置nginx，主机1和主机2的配置文件相同
配置文件名：/usr/local/nginx/conf/nginx.conf

#配置进程数，一般与CPU核心数相同或者核心的倍数
worker_processes  1;
#error_log  logs/error.log;
#error_log  logs/error.log  notice;
#error_log  logs/error.log  info;
#access_log  logs/access.log  main;
events {
       #设置单个进程的最大连接数
       worker_connections  1024;
       }
http {
        include       mime.types;
        default_type  application/octet-stream;
        sendfile        on;
        keepalive_timeout  65;
        #    access_log  logs/access.log  main;
        server {
            listen       80;
            server_name  localhost;
            location / {
                root   html;
                #自定义错误页面
                index  index.html index.htm;
            }
            error_page   500 502 503 504  /50x.html;
            location = /50x.html {
                root   html;
            }
}
    #设置负载均衡服务器组
    upstream myserv {
    server 192.168.109.208:80  weight=1;
    server 192.168.109.209:80  weight=1;
    }
    server {
        #设置监听VIP（虚拟IP）,，该IP在主机3、4的配置文件中有配置
        listen 192.168.109.215:80;
        server_name myserv;
        #        access_log  /var/logs/access.log  main;
        location / {
                index  index.html index.htm;
                #设置反向代理
                proxy_pass    http://myserv;
                proxy_set_header   X-Real-IP        $remote_addr;
        }
    }
}

(2)配置keepalived

说明：分别在主机1和2上安装keepalived. 其中1为主，2为备。配置文件名：/etc/keepalived/keepalived.conf
         
(1)主机1上的keepalived.conf

! Configuration File for keepalived

global_defs {
    notification_email {
        #设置邮件收件人
        [e-mail]
        [e-mail]
    }
    #设置邮件发件人
    notification_email_from [e-mail]
    smtp_connect_timeout 3
    smtp_server 127.0.0.1
    router_id LVS_DEVEL
}

cript chk_schedown {
    script "[[ -f /etc/keepalived/down ]] && exit 1 || exit 0"
    interval 2
    weight -2
}
#第一种主备切换方案[当监测到主的nginx为0时，启动它；如果启动不了就杀掉主的keepalived]

   vrrp_script chk_http_port {
#    script "/usr/local/nginx/nginx_check.sh"
#    interval 2
#    weight -2
#}
   #第二种主备切换方案[当监测到主的nginx进程是否存活，若没有则降低权重]
   vrrp_script chk_http_port {
   script "killall -0 nginx"
   interval 2
   weight -2
}

rrp_script chk_schedown {

vrrp_instance VI_1 {
    interface eth0
    #配置主从，MASTER为主，BACKUP为备
    state MASTER
    #设置优先级，数值越大，优先级越高，都是主比备大
    priority 101
    virtual_router_id 51
    garp_master_delay 1

    authentication {
    auth_type PASS
    auth_pass password
    }

    track_interface {
    eth0
    }
    #设置虚拟IP地址，主备虚拟IP地址相同
    virtual_ipaddress {
    192.168.109.215
    }

    track_script {
    chk_schedown
    }
}

(2)备机2上的keepalived.conf

! Configuration File for keepalived

global_defs {
    notification_email {
        #设置邮件收件人
        [e-mail]
        [e-mail]
    }
#设置邮件发件人
notification_email_from [e-mail]
smtp_connect_timeout 3
smtp_server 127.0.0.1
router_id LVS_DEVEL
}

cript chk_schedown {
    script "[[ -f /etc/keepalived/down ]] && exit 1 || exit 0"
    interval 2
    weight -2
}
rrp_script chk_schedown {

vrrp_instance VI_1 {
    interface eth0
    #配置主从，MASTER为主，BACKUP为备
    state BACKUP
    #设置优先级，数值越大，优先级越高，都是主比备大
    priority 100
    virtual_router_id 51
    garp_master_delay 1

    authentication {
    auth_type PASS
    auth_pass password
    }

    track_interface {
    eth0
    }
    #设置虚拟IP地址，主备虚拟IP地址相同
    virtual_ipaddress {
    192.168.109.215
    }

    track_script {
    chk_schedown
    }
}

至此，keepalived配置完毕！

4.测试验证

(1)验证负载均衡
在浏览器地址栏输入虚拟IP：192.168.109.215，看是否会看到主机3和主机4对应的页面内容:


并且这2个页面是交替出现。此时即到达了负载均衡。

(2)验证keepalived主备切换

停止主机1上的nginx和keepalived，然后重新刷新页面，看是否还能打开，若能打开，说明keepalived切换成功。
注意：同一台机器上的nginx和keepalived必须同时停止才可以！！不能只停nginx！不然不会生效！



5.编写nginx监控脚本

vim /usr/local/nginx/nginx_check.sh
#!/bin/bash
# varsion 0.0.2
# 根据一网友说这样做不科学，如果nginx服务起来了，但是我把keepalived 杀掉了，我的理由是，如果nginx死掉了，我觉得就很难在起来，再有就是nagios当然要给你报警了啊。不过这位同学说的有道理，所以就稍加改了一下脚本
A=`ps -C nginx --no-header |wc -l`                ## 查看是否有 nginx进程 把值赋给变量A
if [ $A -eq 0 ];then                                         ## 如果没有进程值得为零
        /usr/local/nginx/sbin/nginx
        sleep 3
        if [ `ps -C nginx --no-header |wc -l` -eq 0 ];then
            killall keepalived                        ## 则结束 keepalived 进程
        fi
fi

参考地址：http://deidara.blog.51cto.com/400447/302402/