网站前端服务器高可用方案

本文档编写的目的是为了实现前端nginx代理的高可用，去除前端的单点故障，保证在前端网络出现问题、单台服务器挂掉、nginx崩溃等问题的时候能自动切换到另外一台服务器，保障服务的正常运行。

 

本方案采用nginx+keepalived来保证前端服务器的高可用。在前端网络层面、系统层面、nginx进程层面由keepalived来保证高可用，在代理后方的网络层面、系统层面、nginx可用性等由nginx自身的故障检查机制来保证高可用。

需要提供和一个网通IP和一个电信IP（和原来的IP在一个子网），用来做虚拟IP。

的安装（两台服务器都需要安装）

wget

ln -s /usr/src/kernels/2.6.18-164.el5-i686/ /usr/src/linux

tar zxvf ipvsadm-1.24.tar.gz

cd ipvsadm-1.24

make && make install

 

wget
tar zxvf keepalived-1.1.19.tar.gz
cd keepalived-1.1.19
./configure --prefix=/usr/local/keepalived
make
make install
cp /usr/local/keepalived/sbin/keepalived /usr/sbin/
cp /usr/local/keepalived/etc/sysconfig/keepalived /etc/sysconfig/
cp /usr/local/keepalived/etc/rc.d/init.d/keepalived /etc/init.d/
mkdir /etc/keepalived

的配置

的配置

vim /etc/keepalived/keepalived.conf

#Configuration File for keepalived

 

global_defs {

notification_email {

               [e-mail]

        }

        notification_email_from [e-mail]

        smtp_server mail.test.com

        smtp_connect_timeout 30

        router_id LVS2

}

 

 

vrrp_script check_nginx {

          #检测nginx是否挂掉的脚本

     script "/etc/keepalived/check_nginx.sh"

     interval 2

     weight 2

     }

 

 

vrrp_instance loadbalance {

        state MASTER

        interface eth1

        virtual_router_id 52

        priority 180

        #mtp_alert

        advert_int 1

                  #发生切换以后报警的脚本

        notify /etc/keepalived/alert.sh

 

authentication {

        auth_type PASS

        auth_pass ufsoft

}

 

track_script {

     check_nginx

     }

 

virtual_ipaddress {

                   #电信和网通的虚拟IP

         143.20.33.28

142.30.31.124

}

}

的配置

vim /etc/keepalived/keepalived.conf

#Configuration File for keepalived

 

global_defs {

notification_email {

               [e-mail]

        }

        notification_email_from [e-mail]

        smtp_server mail.test.com

        smtp_connect_timeout 30

        router_id LVS2

}

 

 

vrrp_script check_nginx {

     script "/etc/keepalived/check_nginx.sh"

     interval 2

     weight 2

     }

 

 

vrrp_instance loadbalance {

        state BACKUP

        interface eth1

        virtual_router_id 52

        priority 150

        #mtp_alert

        advert_int 1

        notify /etc/keepalived/alert.sh

 

authentication {

        auth_type PASS

        auth_pass ufsoft

}

 

track_script {

     check_nginx

     }

 

virtual_ipaddress {

         143.20.33.28

142.30.31.124

}

}

 

nginx脚本配置（28、29都需要配置）

vim /etc/keepalived/check_nginx.sh

#!/bin/bash

if [ `ps aux | grep "nginx: master process" | grep -v "grep"|wc -l` -eq 0 ];then

        /usr/local/nginx/sbin/nginx

        sleep 5

        if [ `ps -aux | grep "nginx: master process" | grep -v "grep"|wc -l` -eq 0 ];then

                /etc/rc.d/init.d/keepalived stop

        fi

fi

chmod +x /etc/keepalived/check_nginx.sh

（两台服务器都需要配置）

vim /etc/keepalived/alert.sh

#!/bin/bash

group_instance=$1

group_name=$2

event=$3

#用来设置邮件报警

/etc/keepalived/sendmail.pl "keepalived problem, $group_instance $group_name $event please check it out!!!" "keepalived"

#用来设置短信报警

/etc/keepalived/sendsms.pl 手机号码 $group_instance $group_name $event 'keepalived problem,please check it out!!!'

chmod +x /etc/keepalived/alert.sh

的管理和维护

的启动

/etc/rc.d/init.d/keepalived start

测试没有问题以后将此启动语句加入到/etc/rc.local自动启动文件中！

的停止

/etc/rc.d/init.d/keepalived stop

keepalived的日志

tail -f  /var/log/messages

通过上面的命令可以查看到日志中keeplived发生切换的过程及发生切换的时间等信息，对于错误的排查很关键。

要测试各个情况下keepalived的切换情况，停掉nginx、停掉keepalived、网络中断等。另外还需要在测试环境测试高负载情况下各种情况的切换速度等。对各个影响做到心中有数！

DNS

更改智能DNS的记录，使域名对应的记录指向两个虚拟IP。

 Keepalived上线后需要添加对应的keepalived的监控，以备出现问题后第一时间响应，监控的内容包括keepalived进程的监控、虚拟IP连通性监控。

   为方便大家学习，制作了pdf文档，下载请点击： 网站前端服务器高可用方案.pdf