Keepalived+Nginx双主切换实验

一、实验架构图
在这里插入图片描述

二、实验环境说明

角色说明	ip地址
调度器1	192.168.126.141
调度器2	192.168.126.139
Web server1	192.168.126.128
Web server2	192.168.126.138
虚拟ip地址1	192.168.126.140
虚拟ip地址2	192.168.126.150

三、实验配置
注意事项：首先我们需要在两个调度器上可以互相解析主机名（需要与本地hosts文件相同），而且需要关闭iptables防火墙与selinux；而后要在调度器两端分别安装好keepalived和nginx程序包。
1）第一步，配置后端web服务器站点资源文件
web server1端：

[root@web1 ~]#vim /var/www/html/index.html
<h1>Page from web1</h1>
[root@master ~]# systemctl restart  httpd
[root@web1 ~]# ss -tunlp | grep 80   //web服务已经启动
tcp    LISTEN     0      128                   :::80                   :::*      users:(("httpd",6753,4),("httpd",6752,4),("httpd",6751,4),("httpd",6750,4),("httpd",6749,4),("httpd",6740,4))
[root@web1 ~]#

web server2端：

[root@db ~]# vim /var/www/html/index.html 
<h1>Page from web2</h1>
[root@db ~]# service  httpd restart 
停止 httpd：                                               [确定]
正在启动 httpd：httpd: apr_sockaddr_info_get() failed for db
httpd: Could not reliably determine the server's fully qualified domain name, using 127.0.0.1 for ServerName
                                                           [确定]
[root@db ~]# ss -tunlp | grep 80   //web服务已经正常启动
tcp    LISTEN     0      128                   :::80                   :::*      users:(("httpd",3213,4))
[root@db ~]#

2）配置nginx成为负载均衡调度器
我们需要在调度器端各自安装nginx程序包，先使nginx可以正常进行的web服务资源调度。
编译安装nginx程序请看：https://blog.csdn.net/Micky_Yang/article/details/88647953

[root@director1 ~]#vim /usr/local/nginx/conf/nginx.conf
http {
    include       mime.types;
    default_type  application/octet-stream;

    upstream webserv{   //在http配置段配置upstream服务器群，名称为webserv；
        server 192.168.126.128:80 weight=1;   //调度的后端服务器地址，权重为1；
        server 192.168.126.138:80 weight=1;   //调度的后端服务器地址，权重也为1，轮询方式；

    }



 location / {
            proxy_pass  http://webserv;     //在server端配置一个location匹配段，把所有请求本地的资源
            								调度到名称为webserv的负载均衡集群中；
        }


//把此配置文件拷贝到director2端一份，并在两端启动nginx
[root@director1 ~]# scp /usr/local/nginx/conf/nginx.conf  director2:/usr/local/nginx/conf/nginx.conf
nginx.conf                                                                         100% 2829     1.2MB/s   00:00    
[root@director1 ~]# 
[root@director1 ~]# /usr/local/nginx/sbin/nginx  
[root@director1 ~]# ss -tunlp | grep 80 
tcp    LISTEN     0      128       *:80                    *:*                   users:(("nginx",pid=79475,fd=6),("nginx",pid=79474,fd=6))
[root@director1 ~]# 
[root@director1 ~]# curl 192.168.126.141
<h1>Page from web1</h1>
[root@director1 ~]# curl 192.168.126.141
<h1>Page from web2</h1>
[root@director1 ~]# curl 192.168.126.141
<h1>Page from web1</h1>
[root@director1 ~]# curl 192.168.126.141  
<h1>Page from web2</h1>    //调度器1端已经成功实现负载均衡效果
[root@director1 ~]#

调度器2配置：

[root@director2 ~]# /usr/local/nginx/sbin/nginx 
[root@director2 ~]# ss -tunlp | grep 80 
tcp    LISTEN     0      128       *:80                    *:*                   users:(("nginx",pid=77953,fd=6),("nginx",pid=77952,fd=6))
[root@director2 ~]# 
[root@director2 ~]# curl 192.168.126.139
<h1>Page from web1</h1>
[root@director2 ~]# curl 192.168.126.139
<h1>Page from web2</h1>
[root@director2 ~]# curl 192.168.126.139
<h1>Page from web1</h1>
[root@director2 ~]# curl 192.168.126.139
<h1>Page from web2</h1>   //调度器2端也已经成功实现负载均衡效果
[root@director2 ~]#

3）配置keepalived，添加虚拟ip实现互为主备

调度器1配置：

! Configuration File for keepalived

global_defs {     //global全局配置段
   notification_email {
        root@localhost     //设置邮件的接受人地址
}
   notification_email_from  [email protected]   //发送邮件的人，此处可以自行定义；
   smtp_server 127.0.0.1    //配置邮件服务器地址，默认为本地即可；
   smtp_connect_timeout 30   //连接邮件服务器的超时时间；
   router_id LVS_DEVEL   
}

vrrp_script chk_mt {    //虚拟ip地址切换脚本
                script "[[ -f /etc/keepalived/down ]] && exit 1 || exit 0"
                interval 1
                weight -2
}

vrrp_script chk_nginx {    //检测nginx是否在线脚本；如果nginx服务不在线，则当前节点权重减10；
        script "killall -0 nginx &> /dev/null"
        interval 1
        weight -10

}

vrrp_instance VI_1 {   //第一个虚拟路由器实例1配置
    state MASTER    //初始状态为主角色；
    interface ens33   //虚拟地址所在的网卡接口名称；
    virtual_router_id 51  //虚拟路由器实例的id号，同一个虚拟路由器实例两端id号必须相同；
    priority 100   //优先级，数字越大，优先级越高；
    advert_int 1  //健康状态监测时长间隔；
    authentication {   //认证相关配置
        auth_type PASS  //字符认证
        auth_pass 1111  //认证密码
    }
    virtual_ipaddress {  
        192.168.126.140  //配置虚拟ip地址；   
    }
    track_script {
        chk_nginx    //此处调用上面的nginx状态检查脚本
    }
}
vrrp_instance VI_2 {   //虚拟路由器实例二的配置段
    state BACKUP    //因为是互相为主备，所以第二个虚拟路由器实例的初始状态为备节点；
    interface ens33   //虚拟地址所在的网卡接口名称；
    virtual_router_id 61 //虚拟路由器实例的id号，同一个虚拟路由器实例两端id号必须相同；
    priority 99    //优先级设定；
    advert_int 1  //状态检查时长间隔；
    authentication {   //认证类型
        auth_type PASS
        auth_pass 2222
    }
    virtual_ipaddress {    
        192.168.126.150  //虚拟路由器实例二的ip地址；
    }
    track_script {
        chk_nginx  //同样调用nginx健康状态检查脚本；
    }
}

调度器2配置：

! Configuration File for keepalived

global_defs {
   notification_email {
        root@localhost
}
   notification_email_from  [email protected]
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id LVS_DEVEL
}

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

vrrp_script chk_nginx {
        script "killall -0 nginx &> /dev/null"
        interval 1
        weight -10

}



vrrp_instance VI_1 {
    state BACKUP     //初始状态为备虚拟节点；
    interface eno16777736
    virtual_router_id 51   
    priority 99    //优先级要小于调度器1的虚拟路由器1的优先级；
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        192.168.126.140   //虚拟路由器1所关联的虚拟ip地址；
    }
    track_script {
        chk_nginx
    }
}

vrrp_instance VI_2 {
    state MASTER    //既然是双主模型，此时虚拟路由器2的配置应该为主，同样提供虚拟ip地址实现负载均衡服务
    interface eno16777736  //虚拟地址的网卡接口地址；
    virtual_router_id 61
    priority 100    //优先级要大于调度器1的虚拟路由器2的优先级；
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 2222
    }
    virtual_ipaddress {
        192.168.126.150  //配置虚拟路由器实例2的ip地址
    }
    track_script {
        chk_nginx   //同样都使用nginx健康状态检查脚本；
    }
}

4）启动服务，使用虚拟IP地址访问服务

调度器1：

[root@director1 ~]# systemctl restart  keepalived 
[root@director1 ~]# ip  addr list 
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    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 group default qlen 1000
    link/ether 00:0c:29:8f:12:c4 brd ff:ff:ff:ff:ff:ff
    inet 192.168.126.141/24 brd 192.168.126.255 scope global noprefixroute dynamic ens33
       valid_lft 1014sec preferred_lft 1014sec
    inet 192.168.126.140/32 scope global ens33    //第一个虚拟ip地址已经在调度器1
       valid_lft forever preferred_lft forever
    inet6 fe80::1729:4a5d:513a:e120/64 scope link noprefixroute 
       valid_lft forever preferred_lft forever
[root@director1 ~]# curl 192.168.126.141
<h1>Page from web1</h1>
[root@director1 ~]# curl 192.168.126.141
<h1>Page from web2</h1>
[root@director1 ~]# curl 192.168.126.141
<h1>Page from web1</h1>
[root@director1 ~]# curl 192.168.126.141
<h1>Page from web2</h1>   //实现调度服务已然没有任何问题
[root@director1 ~]#

调度器2：

[root@director2 ~]# systemctl restart keepalived 
[root@director2 ~]# ip addr list 
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    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: eno16777736: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 00:0c:29:60:18:fd brd ff:ff:ff:ff:ff:ff
    inet 192.168.126.139/24 brd 192.168.126.255 scope global noprefixroute dynamic eno16777736
       valid_lft 1490sec preferred_lft 1490sec
    inet 192.168.126.150/32 scope global eno16777736    //第二个虚拟路由器ip地址已在调度器2上
       valid_lft forever preferred_lft forever
    inet6 fe80::20c:29ff:fe60:18fd/64 scope link noprefixroute 
       valid_lft forever preferred_lft forever
[root@director2 ~]# curl 192.168.126.150
<h1>Page from web1</h1>
[root@director2 ~]# curl 192.168.126.150
<h1>Page from web2</h1>
[root@director2 ~]# curl 192.168.126.150
<h1>Page from web1</h1>
[root@director2 ~]# curl 192.168.126.150
<h1>Page from web2</h1>    //此时，调度器2也已然可以利用虚拟路由ip地址提供负载均衡服务
[root@director2 ~]#

5）模拟故障，查看是否可以实现虚拟ip地址切换
我们先停掉调度器1端的nginx负载均衡服务，查看虚拟ip地址1是否可以切换到调度器2上

[root@director1 ~]# /usr/local/nginx/sbin/nginx -s stop 
[root@director1 ~]# 
[root@director1 ~]# ip addr list 
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    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 group default qlen 1000
    link/ether 00:0c:29:8f:12:c4 brd ff:ff:ff:ff:ff:ff
    inet 192.168.126.141/24 brd 192.168.126.255 scope global noprefixroute dynamic ens33
       valid_lft 1396sec preferred_lft 1396sec  //此时已经没有虚拟ip地址1；
    inet6 fe80::1729:4a5d:513a:e120/64 scope link noprefixroute 
       valid_lft forever preferred_lft forever
3: ens36: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 00:0c:29:8f:12:ce brd ff:ff:ff:ff:ff:ff
[root@director1 ~]#

在调度器2端查看是否有从调度器1端切换过来的虚拟ip地址1

[root@director2 ~]# ip addr list 
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    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: eno16777736: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 00:0c:29:60:18:fd brd ff:ff:ff:ff:ff:ff
    inet 192.168.126.139/24 brd 192.168.126.255 scope global noprefixroute dynamic eno16777736
       valid_lft 1122sec preferred_lft 1122sec
    inet 192.168.126.150/32 scope global eno16777736
       valid_lft forever preferred_lft forever
    inet 192.168.126.140/32 scope global eno16777736   //此时已然有第一个虚拟ip地址
       valid_lft forever preferred_lft forever
    inet6 fe80::20c:29ff:fe60:18fd/64 scope link noprefixroute 
       valid_lft forever preferred_lft forever

我们再查看使用两个虚拟ip地址请求web资源是否正常

[root@director2 ~]# curl 192.168.126.140
<h1>Page from web1</h1>
[root@director2 ~]# curl 192.168.126.140
<h1>Page from web2</h1>
[root@director2 ~]# 
[root@director2 ~]# 
[root@director2 ~]# 
[root@director2 ~]# curl 192.168.126.150
<h1>Page from web1</h1>
[root@director2 ~]# curl 192.168.126.150
<h1>Page from web2</h1>
[root@director2 ~]# curl 192.168.126.150
<h1>Page from web1</h1>
[root@director2 ~]# curl 192.168.126.150
<h1>Page from web2</h1>    //已然没有任何问题，调度器2已接手两个虚拟ip地址，提供负载均衡服务器
[root@director2 ~]#

实验结果：通过以上实验，我们已经可以到达双主故障自动切换，提供负载均衡服务的高可用性，无论哪一端的调度器出现问题，提供虚拟服务器的ip地址都会切换到另一个调度器上，达到服务的高可用性。

Keepalived+Nginx双主切换实验

猜你喜欢