Keepalived 和Nginx搭建高可用集群

Keepalived 高可用集群

架构：
node1:192.168.205.10  (keepalived主负载服务器)
node2:192.168.205.20  (keepalived备负载服务器)
node3:192.168.205.30  (web01服务器)
node4:192.168.205.40  (web02服务器)

安装Keepalived
yum install keepalived -y
rpm -qa keepalived
已经安装好：
keepalived-1.3.5-6.el7.x86_64

node1 启动keepalived

如果启动不起来：
查看一下配置，确认一下网卡。
vim /etc/keepalived/keepalived.conf 

下面是成功的，默认是3个进程
[[email protected] vagrant]# ps -ef |grep keep|grep -v grep
root      5268     1  0 10:14 ?        00:00:00 /usr/sbin/keepalived -D
root      5269  5268  0 10:14 ?        00:00:00 /usr/sbin/keepalived -D
root      5270  5268  0 10:14 ?        00:00:00 /usr/sbin/keepalived -D

默认也是三个
[[email protected] vagrant]# ip add |grep 192.168
    inet 192.168.205.10/24 brd 192.168.205.255 scope global enp0s8
    inet 192.168.200.16/32 scope global enp0s8
    inet 192.168.200.17/32 scope global enp0s8
    inet 192.168.200.18/32 scope global enp0s8

node2也和上面的安装上

node1和node2 都停止keepalived
systemctl stop keepalived.service

编辑：node1

vim /etc/keepalived/keepalived.conf
! Configuration File for keepalived

global_defs {
   notification_email {
     [email protected]
     [email protected]
     [email protected]
   }
   notification_email_from [email protected]
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id node1               #节点名
   #vrrp_skip_check_adv_addr
   #vrrp_strict
   #vrrp_garp_interval 0
   #vrrp_gna_interval 0
}

vrrp_instance VI_1 {
    state MASTER
    interface enp0s8            #网卡名
    virtual_router_id 55        #ip主备一致
    priority 150                #优先级最高（主）
    advert_int 1          
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        192.168.205.15/24 dev enp0s8 node1 enp0s8:1    #虚拟地址，主备一致
    }
}

启动keepalived
systemctl start keepalived

[[email protected] vagrant]# ip a
3: enp0s8: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 08:00:27:06:eb:fa brd ff:ff:ff:ff:ff:ff
    inet 192.168.205.10/24 brd 192.168.205.255 scope global enp0s8
       valid_lft forever preferred_lft forever
    inet 192.168.205.15/24 scope global secondary enp0s8           #已经生效
       valid_lft forever preferred_lft forever
    inet6 fe80::a00:27ff:fe06:ebfa/64 scope link 
       valid_lft forever preferred_lft forever

编辑：node2
停止keepalived
systemctl stop keepalived.service

 vim /etc/keepalived/keepalived.conf

! Configuration File for keepalived

global_defs {
   notification_email {
     [email protected]
     [email protected]
     [email protected]
   }
   notification_email_from [email protected]
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id node2
   #vrrp_skip_check_adv_addr
   #vrrp_strict
   #vrrp_garp_interval 0
   #vrrp_gna_interval 0
}

vrrp_instance VI_1 {
    state BACKUP
    interface enp0s8
    virtual_router_id 55
    priority 100
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        192.168.205.15/24 dev enp0s8 node1 enp0s8:1
    }
}

启动keepalived 
systemctl start keepalived.service

查看是否生效ip a
enp0s8: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 08:00:27:62:54:9c brd ff:ff:ff:ff:ff:ff
    inet 192.168.205.20/24 brd 192.168.205.255 scope global enp0s8
       valid_lft forever preferred_lft forever
    inet6 fe80::a00:27ff:fe62:549c/64 scope link 
       valid_lft forever preferred_lft forever

注意：正常情况下是没有虚拟机ip的
只有node1 down后node2才会有
node1恢复正常后，node2的虚拟ip还会消失

如果node1正常启动下，node2的虚拟ip还在，检查一下：
1.防火墙是否关闭
2.192.168.205.15/24 dev enp0s8 node1 enp0s8:1    #虚拟地址，主备一致
3.virtual_router_id 55        #ip主备一致
4.priority 150                #优先级最高（主），BACKUP一定要低于MASTER

一般情况下是三项不同的
router id
state
priority

========================================================

开始测试环境试验:
架构：
node1:192.168.205.10  (keepalived主负载服务器)
node2:192.168.205.20  (keepalived备负载服务器)
node3:192.168.205.30  (web01服务器)
node4:192.168.205.40  (web02服务器)

node1和node2的虚拟ip是192.168.205.15
本地和主备host：didi.com  192.168.205.15

node1和node2都安装Nginx和keepalived
node1和node2都安装Nginx

下面四份配置文件完成后重启keepalived和nginx
systemctl start keepalived.service
systemctl start nginx.service

测试阶段：
测试keepalived的可用性
node1执行：ip a  应该是有192.168.205.15
node2执行：ip a  无有192.168.205.15

node1挂掉和关机后是可以将虚拟ip 192.168.205.15 到node2上
网站还是可以继续打开

注意
问题：虚拟ip：192.168.205.15在node1上时，node2是默认启动不了的，可以先让node2取得ip，然后启动nginx
 

Node1 keepalived配置：
! Configuration File for keepalived

global_defs {
   notification_email {
     [email protected]
     [email protected]
     [email protected]
   }
   notification_email_from [email protected]
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id node1
   #vrrp_skip_check_adv_addr
   #vrrp_strict
   #vrrp_garp_interval 0
   #vrrp_gna_interval 0
}

vrrp_instance VI_1 {
    state MASTER
    interface enp0s8
    virtual_router_id 55
    priority 150
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        192.168.205.15/24 dev enp0s8 node1 enp0s8:1
    }
}

Node1的Nginx配置：

worker_processes  1;
events {
    worker_connections  1024;
}
http {
    include       mime.types;
    default_type  application/octet-stream;

    sendfile        on;
    keepalive_timeout  65;
    upstream www_didi_pools {
        server 192.168.205.30:80 weight=1;
        server 192.168.205.40:80 weight=1;
    }
    server {
        listen 192.168.205.15:80;
        server_name  didi.com www.didi.com;
        location / {
            proxy_pass http://www_didi_pools;
            proxy_set_header Host $host;
            proxy_set_header X-Forwarded-For $remote_addr;
        }
    }
}

Node2 的keepalived配置：
! Configuration File for keepalived

global_defs {
   notification_email {
     [email protected]
     [email protected]
     [email protected]
   }
   notification_email_from [email protected]
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id node2
   #vrrp_skip_check_adv_addr
   #vrrp_strict
   #vrrp_garp_interval 0
   #vrrp_gna_interval 0
}

vrrp_instance VI_1 {
    state BACKUP
    interface enp0s8
    virtual_router_id 55
    priority 100
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        192.168.205.15/24 dev enp0s8 node1 enp0s8:1
    }
}

Node2的Nginx配置
worker_processes  1;
events {
    worker_connections  1024;
}
http {
    include       mime.types;
    default_type  application/octet-stream;

    sendfile        on;
    keepalive_timeout  65;
    upstream www_didi_pools {
        server 192.168.205.30:80 weight=1;
        server 192.168.205.40:80 weight=1;
    }
    server {
        listen 192.168.205.15:80;
        server_name  didi.com www.didi.com;
        location / {
            proxy_pass http://www_didi_pools;
            proxy_set_header Host $host;
            proxy_set_header X-Forwarded-For $remote_addr;
        }
    }
}