Nginx-keepalived+Nginx实现高可用集群

Keepalived+Nginx 高可用集群（主从模式）

集群架构图：

说明：Keepalived机器同样是nginx负载均衡器。

1）实验环境准备（此处都是使用的centos7系统）

# cat /etc/redhat-release
CentOS Linux release 7.4. (Core)

在所有节点上面进行配置

# systemctl stop firewalld         //关闭防火墙
# sed -i 's/^SELINUX=.*/SELINUX=disabled/' /etc/sysconfig/selinux        //关闭selinux，重启生效
# setenforce         　　　　　　　　//关闭selinux，临时生效
# ntpdate .centos.pool.ntp.org    //时间同步
# yum install nginx -y        　　　//安装nginx

2）配置后端web服务器（两台一样）

# echo "`hostname` `ifconfig ens33 |sed -n 's#.*inet \(.*\)netmask.*#\1#p'`" > /usr/share/nginx/html/index.html        //准备测试文件，此处是将主机名和ip写到index.html页面中

# vim /etc/nginx/nginx.conf        //编辑配置文件
user nginx;
worker_processes auto;
error_log /var/log/nginx/error.log;
pid /run/nginx.pid;
include /usr/share/nginx/modules/*.conf;
events {
    worker_connections 1024;
}
http {
    log_format  main  '$remote_addr - $remote_user [$time_local] "$request" '
                      '$status $body_bytes_sent "$http_referer" '
                      '"$http_user_agent" "$http_x_forwarded_for"';
    access_log  /var/log/nginx/access.log  main;
    sendfile            on;
    tcp_nopush          on;
    tcp_nodelay         on;
    keepalive_timeout   65;
    types_hash_max_size 2048;
    include             /etc/nginx/mime.types;
    default_type        application/octet-stream;
    include /etc/nginx/conf.d/*.conf;
    server {
        listen       80;
        server_name  www.mtian.org;
        location / {
            root         /usr/share/nginx/html;
        }
    access_log    /var/log/nginx/access.log main;
    }
}

# systemctl start nginx    //启动nginx
# systemctl enable nginx    //加入开机启动

3）配置LB服务器（两台都一样）

# vim /etc/nginx/nginx.conf
user nginx;
worker_processes auto;
error_log /var/log/nginx/error.log;
pid /run/nginx.pid;
include /usr/share/nginx/modules/*.conf;
events {
    worker_connections 1024;
}
http {
    log_format  main  '$remote_addr - $remote_user [$time_local] "$request" '
                      '$status $body_bytes_sent "$http_referer" '
                      '"$http_user_agent" "$http_x_forwarded_for"';
    access_log  /var/log/nginx/access.log  main;
    sendfile            on;
    tcp_nopush          on;
    tcp_nodelay         on;
    keepalive_timeout   65;
    types_hash_max_size 2048;
    include             /etc/nginx/mime.types;
    default_type        application/octet-stream;
    include /etc/nginx/conf.d/*.conf;
    upstream backend {
    server 192.168.1.33:80 weight=1 max_fails=3 fail_timeout=20s;
    server 192.168.1.34:80 weight=1 max_fails=3 fail_timeout=20s;
    }
    server {
        listen       80;
        server_name  www.mtian.org;
        location / {
        proxy_pass http://backend;
        proxy_set_header Host $host:$proxy_port;
        proxy_set_header X-Forwarded-For $remote_addr;
        }
    }
}

# systemctl start nginx     //启动nginx
# systemctl enable nginx    //加入开机自启动

4）在测试机（192.168.1.35）上面添加host解析，并测试lb集群是否正常。（测试机任意都可以，只要能访问lb节点。）

[root@node01 ~]# cat /etc/hosts
127.0.0.1   localhost localhost.localdomain localhost4 localhost4.localdomain4
::         localhost localhost.localdomain localhost6 localhost6.localdomain6
192.168.1.32    www.mtian.org
192.168.1.31    www.mtian.org

// 测试时候轮流关闭lb1 和 lb2 节点，关闭后还是能够访问并看到轮循效果即表示 nginx lb集群搭建成功。
[root@node01 ~]# curl www.mtian.org
web01 192.168.1.33
[root@node01 ~]# curl www.mtian.org
web02 192.168.1.34
[root@node01 ~]# curl www.mtian.org
web01 192.168.1.33
[root@node01 ~]# curl www.mtian.org
web02 192.168.1.34
[root@node01 ~]# curl www.mtian.org
web01 192.168.1.33
[root@node01 ~]# curl www.mtian.org
web02 192.168.1.34

5）上面步骤成功后，开始搭建keepalived，在两台 lb节点上面安装keepalived（也可以源码编译安装、此处直接使用yum安装）

# yum install keepalived -y

6）配置 LB-01节点

[root@LB- ~]# vim /etc/keepalived/keepalived.conf
! Configuration File for keepalived

global_defs {
   notification_email {
    @qq.com
   }
   smtp_server 192.168.200.1
   smtp_connect_timeout
   router_id LVS_DEVEL
}

vrrp_instance VI_1 {
    state MASTER
    interface ens33
    virtual_router_id
    priority
    advert_int
    authentication {
        auth_type PASS
        auth_pass
    }
    virtual_ipaddress {
    192.168.1.110/24 dev ens33 label ens33:1
    }
}

[root@LB- ~]# systemctl start keepalived     //启动keepalived
[root@LB- ~]# systemctl enable keepalived    //加入开机自启动

[root@LB- ~]# ip a    //查看IP，会发现多出了VIP 192.168.1.110
......
: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu  qdisc pfifo_fast state UP qlen
    link/ether :0c:::: brd ff:ff:ff:ff:ff:ff
    inet 192.168.1.31/ brd 192.168.1.255 scope global ens33
       valid_lft forever preferred_lft forever
    inet 192.168.1.110/24 scope global secondary ens33:1
       valid_lft forever preferred_lft forever
    inet6 fe80::20c:29ff:fe94:/ scope link
       valid_lft forever preferred_lft forever
......

7）配置 LB-02节点

[root@LB- ~]# vim /etc/keepalived/keepalived.conf
! Configuration File for keepalived

global_defs {
   notification_email {
    @qq.com
   }
   smtp_server 192.168.200.1
   smtp_connect_timeout
   router_id LVS_DEVEL
}

vrrp_instance VI_1 {
    state BACKUP
    interface ens33
    virtual_router_id
    priority
    advert_int
    authentication {
        auth_type PASS
        auth_pass
    }
    virtual_ipaddress {
    192.168.1.110/24 dev ens33 label ens33:1
    }
}

[root@LB- ~]# systemctl start keepalived        //启动keepalived
[root@LB- ~]# systemctl enable keepalived    //加入开机自启动

[root@LB- ~]# ifconfig   //查看IP，此时备节点不会有VIP（只有当主挂了的时候，VIP才会飘到备节点）
ens33: flags=<UP,BROADCAST,RUNNING,MULTICAST>  mtu
        inet 192.168.1.32  netmask 255.255.255.0  broadcast 192.168.1.255
        inet6 fe80::20c:29ff:feab:  prefixlen   scopeid 0x20<link>
        ether :0c::ab::  txqueuelen   (Ethernet)
        RX packets   bytes  (16.9 MiB)
        RX errors   dropped   overruns   frame
        TX packets   bytes  (406.0 KiB)
        TX errors   dropped  overruns   carrier   collisions
......

8）在测试机器上面访问 Keepalived上面配置的VIP 192.168.1.110

[root@node01 ~]# curl 192.168.1.110
web01 192.168.1.33
[root@node01 ~]# curl 192.168.1.110
web02 192.168.1.34
[root@node01 ~]# curl 192.168.1.110
web01 192.168.1.33
[root@node01 ~]# curl 192.168.1.110
web02 192.168.1.34 

//关闭LB-01 节点上面keepalived主节点。再次访问
[root@LB- ~]# systemctl stop keepalived
[root@node01 ~]#
[root@node01 ~]# curl 192.168.1.110
web01 192.168.1.33
[root@node01 ~]# curl 192.168.1.110
web02 192.168.1.34
[root@node01 ~]# curl 192.168.1.110
web01 192.168.1.33
[root@node01 ~]# curl 192.168.1.110
web02 192.168.1.34 

//此时查看LB-01 主节点上面的IP ，发现已经没有了 VIP
[root@LB- ~]# ifconfig
ens33: flags=<UP,BROADCAST,RUNNING,MULTICAST>  mtu
        inet 192.168.1.31  netmask 255.255.255.0  broadcast 192.168.1.255
        inet6 fe80::20c:29ff:fe94:  prefixlen   scopeid 0x20<link>
        ether :0c::::  txqueuelen   (Ethernet)
        RX packets   bytes  (17.1 MiB)
        RX errors   dropped   overruns   frame
        TX packets   bytes  (1016.4 KiB)
        TX errors   dropped  overruns   carrier   collisions
...

//查看LB-02 备节点上面的IP，发现 VIP已经成功飘过来了
[root@LB- ~]# ifconfig
ens33: flags=<UP,BROADCAST,RUNNING,MULTICAST>  mtu
        inet 192.168.1.32  netmask 255.255.255.0  broadcast 192.168.1.255
        inet6 fe80::20c:29ff:feab:  prefixlen   scopeid 0x20<link>
        ether :0c::ab::  txqueuelen   (Ethernet)
        RX packets   bytes  (16.9 MiB)
        RX errors   dropped   overruns   frame
        TX packets   bytes  (419.9 KiB)
        TX errors   dropped  overruns   carrier   collisions 

ens33:: flags=<UP,BROADCAST,RUNNING,MULTICAST>  mtu
        inet 192.168.1.110  netmask 255.255.255.0  broadcast 0.0.0.0
        ether :0c::ab::  txqueuelen   (Ethernet)
...

到此，Keepalived+Nginx高可用集群（主从）就搭建完成了。

Keepalived+Nginx 高可用集群（双主模式）

将keepalived做成双主模式，其实很简单，就是再配置一段新的vrrp_instance（实例）规则，主上面加配置一个从的实例规则，从上面加配置一个主的实例规则。

集群架构图：

说明：还是按照上面的环境继续做实验，只是修改LB节点上面的keepalived服务的配置文件即可。此时LB-01节点即为Keepalived的主节点也为备节点，LB-02节点同样即为Keepalived的主节点也为备节点。LB-01节点默认的主节点VIP（192.168.1.110），LB-02节点默认的主节点VIP（192.168.1.210）

1）配置 LB-01 节点

[root@LB- ~]# vim /etc/keepalived/keepalived.conf   //编辑配置文件，增加一段新的vrrp_instance规则
! Configuration File for keepalived

global_defs {
   notification_email {
    @qq.com
   }
   smtp_server 192.168.200.1
   smtp_connect_timeout
   router_id LVS_DEVEL
}

vrrp_instance VI_1 {
    state MASTER
    interface ens33
    virtual_router_id
    priority
    advert_int
    authentication {
        auth_type PASS
        auth_pass
    }
    virtual_ipaddress {
    192.168.1.110/ dev ens33 label ens33:
    }
}

vrrp_instance VI_2 {
    state BACKUP
    interface ens33
    virtual_router_id 52
    priority 100
    advert_int 1
    authentication {
    auth_type PASS
    auth_pass 2222
    }
    virtual_ipaddress {
    192.168.1.210/24 dev ens33 label ens33:2
    }
}

[root@LB- ~]# systemctl restart keepalived    //重新启动keepalived

// 查看LB-01 节点的IP地址，发现VIP（192.168.1.110）同样还是默认在该节点
[root@LB- ~]# ip a
: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu  qdisc pfifo_fast state UP qlen
    link/ether :0c:::: brd ff:ff:ff:ff:ff:ff
    inet 192.168.1.31/ brd 192.168.1.255 scope global ens33
       valid_lft forever preferred_lft forever
    inet 192.168.1.110/ scope global secondary ens33:
       valid_lft forever preferred_lft forever
    inet6 fe80::20c:29ff:fe94:/ scope link
       valid_lft forever preferred_lft forever

2）配置 LB-02 节点

[root@LB- ~]# vim /etc/keepalived/keepalived.conf    //编辑配置文件，增加一段新的vrrp_instance规则
! Configuration File for keepalived

global_defs {
   notification_email {
    @qq.com
   }
   smtp_server 192.168.200.1
   smtp_connect_timeout
   router_id LVS_DEVEL
}

vrrp_instance VI_1 {
    state BACKUP
    interface ens33
    virtual_router_id
    priority
    advert_int
    authentication {
        auth_type PASS
        auth_pass
    }
    virtual_ipaddress {
    192.168.1.110/ dev ens33 label ens33:
    }
}

vrrp_instance VI_2 {
    state MASTER
    interface ens33
    virtual_router_id 52
    priority 150
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 2222
    }
    virtual_ipaddress {
        192.168.1.210/24 dev ens33 label ens33:2
    }
}

[root@LB- ~]# systemctl restart keepalived    //重新启动keepalived

// 查看LB-02节点IP，会发现也多了一个VIP（192.168.1.210），此时该节点也就是一个主了。
[root@LB- ~]# ip a
: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu  qdisc pfifo_fast state UP qlen
    link/ether :0c::ab:: brd ff:ff:ff:ff:ff:ff
    inet 192.168.1.32/ brd 192.168.1.255 scope global ens33
       valid_lft forever preferred_lft forever
    inet 192.168.1.210/ scope global secondary ens33:
       valid_lft forever preferred_lft forever
    inet6 fe80::20c:29ff:feab:/ scope link
       valid_lft forever preferred_lft forever

3）测试

[root@node01 ~]# curl 192.168.1.110
web01 192.168.1.33
[root@node01 ~]# curl 192.168.1.110
web02 192.168.1.34
[root@node01 ~]# curl 192.168.1.210
web01 192.168.1.33
[root@node01 ~]# curl 192.168.1.210
web02 192.168.1.34

// 停止LB-01节点的keepalived再次测试
[root@LB- ~]# systemctl stop keepalived
[root@node01 ~]# curl 192.168.1.110
web01 192.168.1.33
[root@node01 ~]# curl 192.168.1.110
web02 192.168.1.34
[root@node01 ~]# curl 192.168.1.210
web01 192.168.1.33
[root@node01 ~]# curl 192.168.1.210
web02 192.168.1.34

测试可以发现我们访问keepalived中配置的两个VIP都可以正常调度等，当我们停止任意一台keepalived节点，同样还是正常访问；到此，keepalived+nginx高可用集群（双主模式）就搭建完成了。