keepalived绑定单播地址、非抢占模式及LVS的TCP模式的高可用

背景：keepalived默认是组播地址进行播放，且默认地址是224.0.0.18，如果配置多个keepalived主机，会导致虚拟IP地址存在冲突问题，这种问题怎么解决呢？

解决办法：就是将keepalived主机的多播地址修改为单播地址，绑定固定IP地址，避免在多播模式下，通过VRRP进行广播地址，造成IP地址地址冲突。

vrrp_strict #严格遵守VRRP协议,不允许状况，在配置单播IP地址时，此行需要删除或者注释掉即可。

1、没有VIP地址

2、单播邻居

3、在VRRP版本2中有IPv6地址

实验一：实现keepalived单播地址配置

1、在主节点配置keepalived文件

[root@centos_17~]#vim /etc/keepalived/keepalived.conf
global_defs {
   notification_email {
     root@localhost.com
   }
   notification_email_from root@localhost.com
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id ka1
   vrrp_skip_check_adv_addr
   vrrp_iptables
   vrrp_garp_interval 0
   vrrp_gna_interval 0
}
 
vrrp_instance VI_1 {
    state MASTER
    interface ens33
    virtual_router_id 50
    priority 100
    unicast_src_ip 192.168.37.17  配置源地址的IP地址
    unicast_peer {
       192.168.37.7  配置从节点的目标IP地址
    }
    advert_int 2
    authentication {
        auth_type pass
        auth_pass 123456
    }
    virtual_ipaddress {
        192.168.37.100 dev ens33 label ens33:1
    }
}

配置完成之后重启keepalived：

[root@centos_17~]#systemctl restart keepalived

2、在从节点配置keepalived文件　　

[root@centos7~]#vim /etc/keepalived/keepalived.conf
global_defs {
   notification_email {
     root@localhost.com
   }
   notification_email_from root@localhost.com
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id ka2
   vrrp_skip_check_adv_addr
   vrrp_iptables
   vrrp_strict
   vrrp_garp_interval 0
   vrrp_gna_interval 0
}
 
vrrp_instance VI_1 {
    state BACKUP
    interface ens33
    virtual_router_id 50
    priority 80
    unicast_src_ip 192.168.37.7
    unicast_peer {
       192.168.37.17
        }
    advert_int 2
    authentication {
        auth_type pass
        auth_pass 123456
    }
    virtual_ipaddress {
        192.168.37.100 dev ens33 label ens33:1
    }
}

　配置完成之后重启keepalived：　

[root@centos_17~]#systemctl restart keepalived

3、验证锁单播IP地址效果。

此时由于绑定了双方keepalived主机的IP地址，就只会对双方进行广播，避免多个keepalived的虚拟IP地址进行冲突。

实验二：实现非抢占模式漂移VIP

原理：关闭VIP抢占模式，需要VIP state都为BACKUP，此时哪个keepalived的优先级大，优先占用哪个keepalived服务器，当此占用的服务器宕机后，另一个BACKUP主机才会进行占用，就算优先级高的恢复了，也不能进行抢占，除非优先级低的服务器宕机后，才会继续占用到高优先级的keepalived服务器上。

1、在优先级高的keepalived主机进行配置

vim /etc/keepalived/keepalived.conf

global_defs {
   notification_email {
     root@localhost.com
   }
   notification_email_from root@localhost.com
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id ka1
   vrrp_skip_check_adv_addr
   vrrp_iptables
   vrrp_garp_interval 0
   vrrp_gna_interval 0
}
 
vrrp_instance VIP_1 {
    state BACKUP  角色必须是BACKUP
    interface ens33
    virtual_router_id 50
    priority 100
    unicast_src_ip 192.168.37.17
    unicast_peer {
       192.168.37.7
    }
    advert_int 2
    nopreempt  设置为非抢占模式
    authentication {
        auth_type PASS
        auth_pass 123456
    }
    virtual_ipaddress {
        192.168.37.100 dev ens33 label ens33:1
    }
}

2、在优先级低的keepalived主机配置

vim /etc/keepalived/keepalived.conf

global_defs {
   notification_email {
     root@localhost.com
   }
   notification_email_from root@localhost.com
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id ka2
   vrrp_skip_check_adv_addr
   vrrp_iptables
   vrrp_garp_interval 0
   vrrp_gna_interval 0
}
 
vrrp_instance VIP_1 {
    state BACKUP   角色必须是BACKUP
    interface ens33
    virtual_router_id 60
    priority 80
    unicast_src_ip 192.168.37.7
    unicast_peer {
       192.168.37.17
    }
    advert_int 2
    authentication {
        auth_type PASS
        auth_pass 123456
    }
    virtual_ipaddress {
        192.168.37.100 dev ens33 label ens33:1
    }
}

测试效果：

此时手动将keepalived主机优先级高的进行宕机，VIP就会漂移到优先级低的主机上。

[root@centos_17~]#systemctl stop keepalived

此时查看优先级低的主机IP地址情况：

此时就算高优先级的主机恢复，也无法抢占VIP地址，此时需要宕机优先级低的主机才会占用VIP地址。

启动优先级高的主机：systemctl start keepalived

停掉优先级低的主机：systemctl stop keepalived

实战三：实现两个以上的Keepalived主机

背景：当公司需求量较大时，两个keepalived已经不能满足公司需求，此时需要配置两台以上的keepalived，应该怎么配置？

实现方法如下：

1、在A主机配置keepalived

vim /etc/keepalived/keepalived.conf

global_defs {
   notification_email {
     root@localhost.com
   }
   notification_email_from root@localhost.com
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id ka1
   vrrp_skip_check_adv_addr
   vrrp_iptables
   vrrp_garp_interval 0
   vrrp_gna_interval 0
}
 
vrrp_instance VIP_1 {
    state MASTER  主节点服务器
    interface ens33
    virtual_router_id 60 
    priority 100   优先级为100
    unicast_src_ip 192.168.37.7  绑定单播地址，防止IP地址与其他keepalived地址冲突
    unicast_peer {
       192.168.37.17   目标keepalived主机IP地址
       192.168.37.27   目标keepalived主机IP地址                                                                                                                          
    }
    advert_int 2
    authentication {
        auth_type PASS
        auth_pass 123456
    }
    virtual_ipaddress {
        192.168.37.100 dev ens33 label ens33:1  添加VIP地址，也可以多添加几个地址
    }
}

2、在B主机配置keepalived　　

global_defs {
   notification_email {
     root@localhost.com
   }
   notification_email_from root@localhost.com
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id ka1
   vrrp_skip_check_adv_addr
   vrrp_iptables
   vrrp_garp_interval 0
   vrrp_gna_interval 0
}
 
vrrp_instance VIP_1 {
    state BACKUP  从节点服务器
    interface ens33
    virtual_router_id 50
    priority 80 优先级为80
    unicast_src_ip 192.168.37.17   绑定单播地址，源keepalived的IP地址
    unicast_peer {
       192.168.37.7   两个目标的keepalived的IP地址
       192.168.37.27
    }
    advert_int 2
    authentication {
        auth_type PASS
        auth_pass 123456
    }
    virtual_ipaddress {
        192.168.37.100 dev ens33 label ens33:1
    }
}

3、在C主机配置keepalived　　

global_defs {
   notification_email {
     root@localhost.com
   }
   notification_email_from root@localhost.com
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id ka1
   vrrp_skip_check_adv_addr
   vrrp_iptables
   vrrp_garp_interval 0
   vrrp_gna_interval 0
}
 
vrrp_instance VIP_1 {
    state BACKUP   从节点服务器
    interface ens33
    virtual_router_id 50
    priority 60  优先级为60，要比前两个的主机优先级都要低
    unicast_src_ip 192.168.37.27  绑定单播地址，源地址
    unicast_peer {
       192.168.37.7  两个keepalived的目标地址
       192.168.37.17
    }   
    advert_int 2
    authentication {                                                                                                                             
        auth_type PASS
        auth_pass 123456
    }
    virtual_ipaddress {
        192.168.37.100 dev ens33 label ens33:1
    }
}

测试效果：　　

将A主机的keepalived宕机，可以观察此时的B主机成为MASTER主机，VIP地址就会漂移到B主机上。　　

当B主机的keepalived也宕机之后，此时的VIP就会漂移到C主机从节点的keepalived服务器上

同理：当优先级高的keepalived服务器恢复后，VIP地址就又会漂移回去。

实战四：实现TCP模式keepalived及LVS-DR模式高可用（此用法多于HTTP用法）

参数含义：

delay_loop<INT>：检查后端服务器的时间间隔
lb_algorr|wrr|lc|wlc|lblc|sh|dh：定义调度方法
lb_kindNAT|DR|TUN：集群的类型
persistence_timeout<INT>：持久连接时长
protocol TCP|UDP|SCTP：指定服务协议
sorry_server<IPADDR> <PORT>：所有RS故障时，备用服务器地址
real_server<IPADDR> <PORT>
{
weight <INT> RS权重
notify_up<STRING>|<QUOTED-STRING> RS上线通知脚本
notify_down<STRING>|<QUOTED-STRING> RS下线通知脚本
HTTP_GET|SSL_GET|TCP_CHECK|SMTP_CHECK|MISC_CHEC K { ... }：定义当前主机的健康状态检测方法
}

HTTP监测含义

HTTP_GET|SSL_GET：应用层检测
HTTP_GET|SSL_GET {
url{
path <URL_PATH>：定义要监控的URL
status_code<INT>：判断上述检测机制为健康状态的响应码
}
connect_timeout<INTEGER>：连接请求的超时时长
nb_get_retry<INT>：重试次数
delay_before_retry<INT>：重试之前的延迟时长
connect_ip<IP ADDRESS>：向当前RS哪个IP地址发起健康状态检测请求
connect_port<PORT>：向当前RS的哪个PORT发起健康状态检测请求
bindto<IP ADDRESS>：发出健康状态检测请求时使用的源地址
bind_port<PORT>：发出健康状态检测请求时使用的源端口
}

TCP监测

传输层检测TCP_CHECK

TCP_CHECK {
connect_ip<IP ADDRESS>：向当前RS的哪个IP地址发起健康状态检测请求
connect_port<PORT>：向当前RS的哪个PORT发起健康状态检测请求
bindto<IP ADDRESS>：发出健康状态检测请求时使用的源地址
bind_port<PORT>：发出健康状态检测请求时使用的源端口
connect_timeout<INTEGER>：连接请求的超时时长
}

1、在A主机修改keepalived配置文件

vim /etc/keepalived/keepalived.conf

global_defs {
   notification_email {
     root@localhost.com
   }
   notification_email_from root@localhost.com
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id ka2
   vrrp_skip_check_adv_addr
   vrrp_iptables
   vrrp_garp_interval 0
   vrrp_gna_interval 0
}
 
vrrp_instance VIP_1 {
    state MASTER
    interface ens33
    virtual_router_id 50
    priority 100
    unicast_src_ip 192.168.37.7
    unicast_peer {
       192.168.37.17
    }
    advert_int 2
    authentication {
        auth_type PASS
        auth_pass 123456
    }
    virtual_ipaddress {
        192.168.37.100 dev ens33 label ens33:1
    }
}
 
include  /etc/keepalived/conf/*.conf   定义一个单独管理的目录，定义keepalived配置文件

在A主机定义LVS-DR模式配置文件

如果后期keepalived的配置文件修改过大，我们可以调用include，新建一个目录，并在此目录下进行存放配置文件。

[root@centos7keepalived]#mkdir conf
[root@centos7keepalived]#vim conf/tcp.conf 
virtual_server 192.168.37.100 80 {     VIP地址                                                                                                          
        delay_loop 6
        lb_algo wrr  权重轮询
        lb_kind DR   DR模式
        protocol TCP
        sorry_server 192.168.37.47  80  配置后端sorry服务器，当两个keepalived主机都宕机之后，就在此主机进行显示信息。
 
   real_server 192.168.37.27 80 {  后端RS1服务器IP地址
        weight 1
        TCP_CHECK {
        connect_timeout 5
        nb_get_retry 3
        delay_before_retry 3
        connect_port 80
        }
   }
 
   real_server 192.168.37.37 80 {  后端RS2服务器
        weight 1
        TCP_CHECK {
        connect_timeout 5
        nb_get_retry 3
        delay_before_retry 3
        connect_port 80
        }
    }
}

2、在B主机修改keepalived配置文件　

　vim /etc/keepalived/keepalived.conf

global_defs {
   notification_email {
     root@localhost.com
   }
   notification_email_from root@localhost.com
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id ka1
   vrrp_skip_check_adv_addr
   vrrp_iptables
   vrrp_garp_interval 0
   vrrp_gna_interval 0
}
 
vrrp_instance VIP_1 {
    state BACKUP
    interface ens33
    virtual_router_id 50
    priority 80
    unicast_src_ip 192.168.37.17
    unicast_peer {
       192.168.37.7
    }
    advert_int 2
    authentication {
        auth_type PASS
        auth_pass 123456
    }
    virtual_ipaddress {
        192.168.37.100 dev ens33 label ens33:1
    }
}
 
include  /etc/keepalived/conf/*.conf

在B主机定义LVS-DR模式的配置文件　

新建conf目录，并在此目录下新建一个配置文件

[root@centos_17keepalived]#mkdir conf
[root@centos_17keepalived]#vim conf/tcp.conf 
virtual_server 192.168.37.100 80 {
        delay_loop 6
        lb_algo wrr
        lb_kind DR
        protocol TCP
        sorry_server 192.168.37.47  80  定义sorry server的后端主机，当两个keepalived主机宕机后，就会提示此信息。
 
   real_server 192.168.37.27 80 {
        weight 1
        TCP_CHECK {
        connect_timeout 5
        nb_get_retry 3
        delay_before_retry 3
        connect_port 80
        }
   }
 
   real_server 192.168.37.37 80 {
        weight 1
        TCP_CHECK {
        connect_timeout 5
        nb_get_retry 3
        delay_before_retry 3
        connect_port 80
        }
    }
}

3、后端服务器RS1上修改配置　　

在RS1绑定VIP地址及lo回环网卡

[root@centos27~]#vim lvs_dr_rs.sh 
#!/bin/bash
#Author:wangxiaochun
#Date:2017-08-13                                                                                                                                 
vip=192.168.37.100   绑定VIP地址
mask='255.255.255.255'
dev=lo:1  绑定在lo回环网卡上
rpm -q httpd &> /dev/null || yum -y install httpd &>/dev/null
service httpd start &> /dev/null && echo "The httpd Server is Ready!"
echo "<h1>`hostname`</h1>" > /var/www/html/index.html
 
case $1 in
start)
    echo 1 > /proc/sys/net/ipv4/conf/all/arp_ignore
    echo 1 > /proc/sys/net/ipv4/conf/lo/arp_ignore
    echo 2 > /proc/sys/net/ipv4/conf/all/arp_announce
    echo 2 > /proc/sys/net/ipv4/conf/lo/arp_announce
    ifconfig $dev $vip netmask $mask #broadcast $vip up
    #route add -host $vip dev $dev
    echo "The RS Server is Ready!"
    ;;
stop)
    ifconfig $dev down
    echo 0 > /proc/sys/net/ipv4/conf/all/arp_ignore
    echo 0 > /proc/sys/net/ipv4/conf/lo/arp_ignore
    echo 0 > /proc/sys/net/ipv4/conf/all/arp_announce
    echo 0 > /proc/sys/net/ipv4/conf/lo/arp_announce
    echo "The RS Server is Canceled!"
    ;;
*)
    echo "Usage: $(basename $0) start|stop"
    exit 1
    ;;
esac

执行脚本：

[root@centos27~]#bash lvs_dr_rs.sh  start

4、在RS2后端服务器上修改配置　　

修改RS2的配置脚本，绑定VIP地址及lo回环网卡

#!/bin/bash
#Author:wangxiaochun
#Date:2017-08-13
vip=192.168.37.100                                                                                                                               
mask='255.255.255.255'
dev=lo:1
rpm -q httpd &> /dev/null || yum -y install httpd &>/dev/null
service httpd start &> /dev/null && echo "The httpd Server is Ready!"
echo "<h1>`hostname`</h1>" > /var/www/html/index.html
 
case $1 in
start)
    echo 1 > /proc/sys/net/ipv4/conf/all/arp_ignore
    echo 1 > /proc/sys/net/ipv4/conf/lo/arp_ignore
    echo 2 > /proc/sys/net/ipv4/conf/all/arp_announce
    echo 2 > /proc/sys/net/ipv4/conf/lo/arp_announce
    ifconfig $dev $vip netmask $mask #broadcast $vip up
    #route add -host $vip dev $dev
    echo "The RS Server is Ready!"
    ;;
stop)
    ifconfig $dev down
    echo 0 > /proc/sys/net/ipv4/conf/all/arp_ignore
    echo 0 > /proc/sys/net/ipv4/conf/lo/arp_ignore
    echo 0 > /proc/sys/net/ipv4/conf/all/arp_announce
    echo 0 > /proc/sys/net/ipv4/conf/lo/arp_announce
    echo "The RS Server is Canceled!"
    ;;
*)
    echo "Usage: $(basename $0) start|stop"
    exit 1
    ;;
esac

执行以上脚本

[root@centos37~]#bash lvs_dr_rs.sh  start

4、在sorry后端服务器上修改配置　　

修改sorry后端服务器配置脚本，绑定VIP地址及lo回环网卡

#!/bin/bash
#Author:wangxiaochun
#Date:2017-08-13
vip=192.168.37.100                                                                                                                               
mask='255.255.255.255'
dev=lo:1
rpm -q httpd &> /dev/null || yum -y install httpd &>/dev/null
service httpd start &> /dev/null && echo "The httpd Server is Ready!"
echo "<h1>`hostname`</h1>" > /var/www/html/index.html
 
case $1 in
start)
    echo 1 > /proc/sys/net/ipv4/conf/all/arp_ignore
    echo 1 > /proc/sys/net/ipv4/conf/lo/arp_ignore
    echo 2 > /proc/sys/net/ipv4/conf/all/arp_announce
    echo 2 > /proc/sys/net/ipv4/conf/lo/arp_announce
    ifconfig $dev $vip netmask $mask #broadcast $vip up
    #route add -host $vip dev $dev
    echo "The RS Server is Ready!"
    ;;
stop)
    ifconfig $dev down
    echo 0 > /proc/sys/net/ipv4/conf/all/arp_ignore
    echo 0 > /proc/sys/net/ipv4/conf/lo/arp_ignore
    echo 0 > /proc/sys/net/ipv4/conf/all/arp_announce
    echo 0 > /proc/sys/net/ipv4/conf/lo/arp_announce
    echo "The RS Server is Canceled!"
    ;;
*)
    echo "Usage: $(basename $0) start|stop"
    exit 1
    ;;
esac

执行以上脚本

[root@centos47~]#bash lvs_dr_rs.sh  start

5、测试效果：　

将RS1和RS2及sorry后端服务器添加测试页面，并启动httpd服务：systemctl start httpd

[root@cenots277~]#yum install httpd
[root@cenots27~]#cd /var/www/html
[root@cenots27html]#cat index.html 
<h1>cenots27</h1>
[root@cenots37html]#cat index.html 
<h1>cenots37</h1>
[root@cenots37html]#cat index.html 
sorry server !!

客户端访问LVS的VIP地址，此时LVS将调度到后端服务器，进行轮询访问。

当后端RS1和RS2后端服务器宕机后，此时就会显示sorry server信息。

　类: LVS负载均衡及Keepalived

https://www.cnblogs.com/struggle-1216/p/12098241.html