CentOS7使用keepalive实现nginx的高可用

其他 2020-01-19 13:32:47 阅读次数: 0

目录

1、环境说明... 1

2、获取nginx安装yum源... 1

3、安装nginx. 1

3、keepalive配置文件... 2

4.1、master节点192.168.10.178. 2

4.2、backup节点192.168.10.177. 4

5、nginx检查脚本... 5

6、测试高可用... 6

6.1、修改检查脚本... 6

6.2、确认当前nginx状态... 7

6.3、修改两节点的nginx欢迎页面... 8

6.4、未切换之前浏览器访问nginx. 9

6.5、测试切换nginx的vip. 9

6.6、master节点的故障恢复... 11

6.7、再次前台页面登陆nginx. 12

 

1、环境说明

master ：192.168.10.178

backup ：192.168.10.177

vip ：192.168.10.180

OS：CentOS Linux release 7.4.1708 (Core)

2、获取nginx安装yum源

rpm -ivh http://nginx.org/packages/centos/7/noarch/RPMS/nginx-release-centos-7-0.el7.ngx.noarch.rpm

wget -O /etc/yum.repos.d/CentOS-Base.repo http://mirrors.aliyun.com/repo/Centos-7.repo

 

3、安装nginx

yum -y install  nginx

这边有下载好的，可以直接rpm包安装

链接：https://pan.baidu.com/s/1eJPuzXps-pCJ3gMAr79tkA

提取码：n7ic

[root@mysql5 soft]# rpm -ivh nginx-1.16.1-1.el7.ngx.x86_64.rpm

warning: nginx-1.16.1-1.el7.ngx.x86_64.rpm: Header V4 RSA/SHA1 Signature, key ID 7bd9bf62: NOKEY

Preparing...                          ################################# [100%]

Updating / installing...

   1:nginx-1:1.16.1-1.el7.ngx         ################################# [100%]

----------------------------------------------------------------------

 

Thanks for using nginx!

 

Please find the official documentation for nginx here:

* http://nginx.org/en/docs/

 

Please subscribe to nginx-announce mailing list to get

the most important news about nginx:

* http://nginx.org/en/support.html

 

Commercial subscriptions for nginx are available on:

* http://nginx.com/products/

 

----------------------------------------------------------------------

[root@mysql5 soft]#

 

3、keepalive配置文件

4.1、master节点192.168.10.178

! Configuration File for keepalived

      

           global_defs {

        notification_email {

                [email protected]

                [email protected]

        }

              

        notification_email_from [email protected]

        smtp_server 127.0.0.1

        smtp_connect_timeout 30

        router_id MASTER-HA.177

           }

   

    #for check mysql scripts 

    vrrp_script chk_mysql_port {

        script "/opt/chk_mysql.sh"  

        interval 2                  

        weight -5                   

        fall 2                   

        rise 1                   

           }

              

    vrrp_instance VI_1 {

        state MASTER   

        interface ens33    

        mcast_src_ip 192.168.10.177

        virtual_router_id 51   

        priority 101           

        advert_int 1        

        authentication {  

             auth_type PASS

             auth_pass 1111    

         }

         virtual_ipaddress {   

             192.168.10.179

         }

             

        track_script {              

           chk_mysql_port            

        }

           }

          

           #for check nginx scripts 

    vrrp_script chk_nginx_port {

        script "/opt/chk_nginx.sh"

        interval 2

        weight -5

        fall 2

        rise 1

           }

          

           vrrp_instance VI_2 {

        state BACKUP   

        interface ens33    

        mcast_src_ip 192.168.10.177

        virtual_router_id 52

        priority 99

        advert_int 1

        authentication {

             auth_type PASS

             auth_pass 1111

         }

         virtual_ipaddress {   

             192.168.10.180

         }

 

        track_script {

           chk_nginx_port

        }

           }

主要看飘红的，上面的VI_1是之前做mysql的双主高可用的配置，可以不用看。

4.2、backup节点192.168.10.177

! Configuration File for keepalived

 

           global_defs {

        notification_email {

                [email protected]

                [email protected]

        }

 

        notification_email_from [email protected]

        smtp_server 127.0.0.1

        smtp_connect_timeout 30

        router_id MASTER-HA.178

        }

 

        vrrp_script chk_mysql_port {

            script "/opt/chk_mysql.sh"

            interval 2

            weight -5

            fall 2

            rise 1

        }

 

           vrrp_instance VI_1 {

                      state BACKUP

                      interface ens33

                      mcast_src_ip 192.168.10.178

                      virtual_router_id 51

                      priority 99

                      advert_int 1

                      authentication {

                                  auth_type PASS

                                  auth_pass 1111

                      }

                      virtual_ipaddress {

                                  192.168.10.179

                      }

 

        track_script {

           chk_mysql_port

        }

    }

   

           #for check nginx scripts 

    vrrp_script chk_nginx_port {

        script "/opt/chk_nginx.sh"

        interval 2

        weight -5

        fall 2

        rise 1

           }

          

           vrrp_instance VI_2 {

        state MASTER   

        interface ens33    

        mcast_src_ip 192.168.10.178

        virtual_router_id 52

        priority 101

        advert_int 1

        authentication {

             auth_type PASS

             auth_pass 1111

         }

         virtual_ipaddress {   

             192.168.10.180

         }

 

        track_script {

           chk_nginx_port

        }

           }

主要看飘红的，上面的VI_1是之前做mysql的双主高可用的配置，可以不用看。

 

5、nginx检查脚本

vi /opt/chk_nginx.sh

 

#!/bin/bash

 

#检测nginx是否启动了

 

A=`ps -C nginx --no-header |wc -l`

 

#如果nginx没有启动就启动nginx

   

if [ $A -eq 0 ];then

 

      #重启nginx

 

      systemctl start nginx

 

      #nginx重启失败，则停掉keepalived服务，进行VIP转移

 

      if [ `ps -C nginx --no-header |wc -l` -eq 0 ];then

           killall keepalived

      fi

fi

赋权：

chmod 775 /opt/chk_nginx.sh

 

6、测试高可用

6.1、修改检查脚本

让测试看到效果，需要把chk_nginx.sh检查脚本改一下，注释掉自启动nginx的命令systemctl start nginx：

[root@mysql5 yum.repos.d]# cat /opt/chk_nginx.sh

#!/bin/bash

 

#检测nginx是否启动了

 

A=`ps -C nginx --no-header |wc -l`

 

#如果nginx没有启动就启动nginx

   

if [ $A -eq 0 ];then

 

        #重启nginx

 

#       systemctl start nginx

 

        #nginx重启失败，则停掉keepalived服务，进行VIP转移

 

        if [ `ps -C nginx --no-header |wc -l` -eq 0 ];then

                killall keepalived

        fi

fi

6.2、确认当前nginx状态

现在nginx的vip 192.168.10.180在keepalive的master上192.168.10.178上

[root@mysql5 yum.repos.d]# ip a

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:29:27:e2:10 brd ff:ff:ff:ff:ff:ff

    inet 192.168.10.178/24 brd 192.168.10.255 scope global ens33

       valid_lft forever preferred_lft forever

    inet 192.168.10.180/32 scope global ens33

       valid_lft forever preferred_lft forever

    inet6 fda7:4ac0:63::736/128 scope global

       valid_lft forever preferred_lft forever

    inet6 fda7:4ac0:63:0:367b:f0fa:c290:8367/64 scope global noprefixroute

       valid_lft forever preferred_lft forever

    inet6 fe80::8d0:5ce9:59e6:cbb8/64 scope link

       valid_lft forever preferred_lft forever

[root@mysql5 yum.repos.d]#

 

我们确定一下keepalive的backup节点192.168.10.177上没有，防止防火墙等原因导致的集群脑裂：

[root@mysql5 html]# ip a

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:29:8b:41:06 brd ff:ff:ff:ff:ff:ff

    inet 192.168.10.177/24 brd 192.168.10.255 scope global ens33

       valid_lft forever preferred_lft forever

    inet 192.168.10.179/32 scope global ens33

       valid_lft forever preferred_lft forever

    inet6 fda7:4ac0:63::8ec/128 scope global

       valid_lft forever preferred_lft forever

    inet6 fda7:4ac0:63:0:9f26:83d:6510:a6dd/64 scope global noprefixroute

       valid_lft forever preferred_lft forever

    inet6 fe80::58:dc06:f8c4:84b3/64 scope link

       valid_lft forever preferred_lft forever

[root@mysql5 html]#

确实没有vip为192.168.10.180的ip，上面的192.168.10.179是mysql双主集群高可用的vip。

6.3、修改两节点的nginx欢迎页面

然后我们为了看到更好区分的效果，将两台主机的nginx欢迎页面改成带各自不同表示的样子

cat /etc/nginx/nginx.conf

找到：

include /etc/nginx/conf.d/*.conf;

然后找到：

[root@mysql5 html]# cd /etc/nginx/conf.d/

[root@mysql5 conf.d]# ls

default.conf

[root@mysql5 conf.d]# cat default.conf

找到：

location / {

        root   /usr/share/nginx/html;

        index  index.html index.htm;

    }

然后修改对应的index.html，这个就是nginx默认的页面欢迎文件：

[root@mysql5 conf.d]# cd /usr/share/nginx/html

[root@mysql5 html]# ls

50x.html  index.html

[root@mysql5 html]#

我这里在<h1>Welcome to nginx!10.177</h1>这一行后面加个本机的ip标示，可以随意，能区分就行：

[root@mysql5 html]# cat index.html

<!DOCTYPE html>

<html>

<head>

<title>Welcome to nginx!10.177</title>

<style>

    body {

        width: 35em;

        margin: 0 auto;

        font-family: Tahoma, Verdana, Arial, sans-serif;

    }

</style>

</head>

<body>

<h1>Welcome to nginx!10.177</h1>

<p>If you see this page, the nginx web server is successfully installed and

working. Further configuration is required.</p>

 

<p>For online documentation and support please refer to

<a href="http://nginx.org/">nginx.org</a>.<br/>

Commercial support is available at

<a href="http://nginx.com/">nginx.com</a>.</p>

 

<p><em>Thank you for using nginx.</em></p>

</body>

</html>

另外一台nginx同理一样修改。

6.4、未切换之前浏览器访问nginx

此时我们通过浏览器访问nginx的vip192.168.10.180，出来的页面会显示的是10.178的地址：

http://192.168.10.180/

 

6.5、测试切换nginx的vip

在master192.168.10.178上停nginx

[root@mysql5 html]# systemctl stop nginx

停完根据我们的配置文件，最多等1秒，就会执行检查脚本，检查到nginx停掉后，就立马发生vip切换

[root@mysql5 html]# ip a

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:29:27:e2:10 brd ff:ff:ff:ff:ff:ff

    inet 192.168.10.178/24 brd 192.168.10.255 scope global ens33

       valid_lft forever preferred_lft forever

    inet6 fda7:4ac0:63::736/128 scope global

       valid_lft forever preferred_lft forever

    inet6 fda7:4ac0:63:0:367b:f0fa:c290:8367/64 scope global noprefixroute

       valid_lft forever preferred_lft forever

    inet6 fe80::8d0:5ce9:59e6:cbb8/64 scope link

       valid_lft forever preferred_lft forever

[root@mysql5 html]#

此时发现vip192.168.10.180已经不在master192.168.10.178上了。

我们去backup192.168.10.177上看一下，已经飘过来了

[root@mysql5 html]# ip a

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:29:8b:41:06 brd ff:ff:ff:ff:ff:ff

    inet 192.168.10.177/24 brd 192.168.10.255 scope global ens33

       valid_lft forever preferred_lft forever

    inet 192.168.10.179/32 scope global ens33

       valid_lft forever preferred_lft forever

    inet 192.168.10.180/32 scope global ens33

       valid_lft forever preferred_lft forever

    inet6 fda7:4ac0:63::8ec/128 scope global

       valid_lft forever preferred_lft forever

    inet6 fda7:4ac0:63:0:9f26:83d:6510:a6dd/64 scope global noprefixroute

       valid_lft forever preferred_lft forever

    inet6 fe80::58:dc06:f8c4:84b3/64 scope link

       valid_lft forever preferred_lft forever

[root@mysql5 html]#

5.6、再次浏览器访问nginx

此时我们在用浏览器登陆nginx的vip192.168.10.180地址，发现页面已经变成backup的地址了。

 

6.6、master节点的故障恢复

此时我们在master192.168.10.178上将nginx在起起来：

systemctl start nginx

发现ip又会被master节点192.168.10.177抢占回来，因为我们配置了主备节点的优先级，主节点优先级要高，所以当主节点故障恢复后，将会重新抢占vip。

[root@mysql5 html]# ip a

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:29:27:e2:10 brd ff:ff:ff:ff:ff:ff

    inet 192.168.10.178/24 brd 192.168.10.255 scope global ens33

       valid_lft forever preferred_lft forever

    inet 192.168.10.180/32 scope global ens33

       valid_lft forever preferred_lft forever

    inet6 fda7:4ac0:63::736/128 scope global

       valid_lft forever preferred_lft forever

    inet6 fda7:4ac0:63:0:367b:f0fa:c290:8367/64 scope global noprefixroute

       valid_lft forever preferred_lft forever

    inet6 fe80::8d0:5ce9:59e6:cbb8/64 scope link

       valid_lft forever preferred_lft forever

[root@mysql5 html]#

保险起见，我们还是检查一下backup节点192.168.10.177：

[root@mysql5 html]# ip a

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:29:8b:41:06 brd ff:ff:ff:ff:ff:ff

    inet 192.168.10.177/24 brd 192.168.10.255 scope global ens33

       valid_lft forever preferred_lft forever

    inet 192.168.10.179/32 scope global ens33

       valid_lft forever preferred_lft forever

    inet6 fda7:4ac0:63::8ec/128 scope global

       valid_lft forever preferred_lft forever

    inet6 fda7:4ac0:63:0:9f26:83d:6510:a6dd/64 scope global noprefixroute

       valid_lft forever preferred_lft forever

    inet6 fe80::58:dc06:f8c4:84b3/64 scope link

       valid_lft forever preferred_lft forever

发现没有nginx的vip192.168.10.180了。

6.7、再次前台页面登陆nginx

前台登陆nginx的页面，发现又回到了192.168.10.178的master节点上了。

 

kadwf123
发布了177 篇原创文章 · 获赞 43 · 访问量 46万+
他的留言板 关注

猜你喜欢

转载自blog.csdn.net/kadwf123/article/details/100538234
今日推荐
周排行
成为C++高手之宏与枚举
在CAD二次开发中使用进度条
Js插件ECharts，HighCharts学习网址整理
Celery提交任务出错(on windows.)
cephfs内核客户端性能追踪
thinkphp中PHPExcel用法
EntityFramework动态组合多排序字段
汇编语言（八）实验9 根据材料编程
安装ubuntu后必须做的事情（对我而言）
JS函数式编程
每日归档
更多
2024-10-22(0)
2024-10-21(0)
2024-10-20(0)
2024-10-19(0)
2024-10-18(0)
2024-10-17(0)
2024-10-16(0)
2024-10-15(0)
2024-10-14(0)
2024-10-13(0)

代码天地 © 2018 codetd.com

境外服务器   最新电视剧2022