【 Keepalived 】Nginx or Http 主-主模式

上一篇：【 Keepalived 】Nginx or Http 主-备模式

在此基础上进行修改并实现 Keepalived主-主模式

首先，需要理解的是：主-备模式是一个VIP在工作，主-主模式则需要两个VIP来工作。一旦其中一台主机出现问题，则两个VIP都会绑定到同一台主机上，待到故障解除，重新启动httpd服务，恢复正常。

ka1 配置文件：

! Configuration File for keepalived

global_defs {
   notification_email {
     root@localhost
   }
   notification_email_from root@localhost
   smtp_server localhost
   smtp_connect_timeout
   router_id NodeA
}

vrrp_script check_nginx {
   script "/etc/keepalived/bash/check_nginx.sh"
   interval
   weight -
}

vrrp_instance VI_1 {
    state MASTER
    interface eth0
    virtual_router_id
    priority
    advert_int
    authentication {
        auth_type PASS
        auth_pass
    }
    track_script {
    check_nginx
    }
    virtual_ipaddress {
        192.168.2.200/
    }
}
vrrp_instance VI_2 {
    state BUCKUP
    interface eth0
    virtual_router_id
    priority
    advert_int
    authentication {
        auth_type PASS
        auth_pass
    }
    track_script {
        check_nginx
    }
    virtual_ipaddress {
        192.168.2.201/
    }
}

ka2 配置文件：

! Configuration File for keepalived

global_defs {
   notification_email {
     root@localhost
   }
   notification_email_from root@localhost
   smtp_server localhost
   smtp_connect_timeout
   router_id NodeB
}

vrrp_script check_nginx {
   script "/etc/keepalived/bash/check_nginx.sh"
   interval
   weight -
}

vrrp_instance VI_1 {
    state BACKUP
    interface eth1
    virtual_router_id
    priority
    advert_int
    authentication {
        auth_type PASS
        auth_pass
    }
    track_script {
    check_nginx
    }
    virtual_ipaddress {
        192.168.2.200/
    }
}
vrrp_instance VI_2 {
    state MASTER
    interface eth1
    virtual_router_id
    priority
    advert_int
    authentication {
        auth_type PASS
        auth_pass
    }
    track_script {
        check_nginx
    }
    virtual_ipaddress {
        192.168.2.201/
    }
}

测试：

正常情况下：

ka1:

[ka1 root@192.168.2.10 ~]#ip a
: lo: <LOOPBACK,UP,LOWER_UP> mtu  qdisc noqueue state UNKNOWN
    link/loopback ::::: brd :::::
    inet 127.0.0.1/ scope host lo
    inet6 ::/ scope host
       valid_lft forever preferred_lft forever
: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu  qdisc pfifo_fast state UP qlen
    link/ether :0c::::4f brd ff:ff:ff:ff:ff:ff
    inet 192.168.2.10/ brd 192.168.2.255 scope global eth0
    inet 192.168.2.200/ scope global secondary eth0
    inet6 fe80::20c:29ff:fe39:924f/ scope link
       valid_lft forever preferred_lft forever

ka2:

[ka2 root@192.168.2.11 ~]#ip a
: lo: <LOOPBACK,UP,LOWER_UP> mtu  qdisc noqueue state UNKNOWN
    link/loopback ::::: brd :::::
    inet 127.0.0.1/ scope host lo
    inet6 ::/ scope host
       valid_lft forever preferred_lft forever
: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu  qdisc pfifo_fast state UP qlen
    link/ether :0c::7b:9f:8c brd ff:ff:ff:ff:ff:ff
    inet 192.168.2.11/ brd 192.168.2.255 scope global eth1
    inet 192.168.2.201/ scope global secondary eth1
    inet6 fe80::20c:29ff:fe7b:9f8c/ scope link
       valid_lft forever preferred_lft forever

当ka1 httpd出现错误时：

ka1 VIP被转移

[ka1 root@192.168.2.10 ~]#ip  a
: lo: <LOOPBACK,UP,LOWER_UP> mtu  qdisc noqueue state UNKNOWN
    link/loopback ::::: brd :::::
    inet 127.0.0.1/ scope host lo
    inet6 ::/ scope host
       valid_lft forever preferred_lft forever
: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu  qdisc pfifo_fast state UP qlen
    link/ether :0c::::4f brd ff:ff:ff:ff:ff:ff
    inet 192.168.2.10/ brd 192.168.2.255 scope global eth0
    inet6 fe80::20c:29ff:fe39:924f/ scope link
       valid_lft forever preferred_lft forever

ka2 被绑定了两个VIP
[ka2 root@192.168.2.11 ~]#ip  a
: lo: <LOOPBACK,UP,LOWER_UP> mtu  qdisc noqueue state UNKNOWN
    link/loopback ::::: brd :::::
    inet 127.0.0.1/ scope host lo
    inet6 ::/ scope host
       valid_lft forever preferred_lft forever
: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu  qdisc pfifo_fast state UP qlen
    link/ether :0c::7b:9f:8c brd ff:ff:ff:ff:ff:ff
    inet 192.168.2.11/ brd 192.168.2.255 scope global eth1
    inet 192.168.2.201/ scope global secondary eth1
    inet 192.168.2.200/ scope global secondary eth1
    inet6 fe80::20c:29ff:fe7b:9f8c/ scope link
       valid_lft forever preferred_lft forever

再三强调：在keepalived使用中，脚本很重要很重要。要经过严格的测试调试，才能使用，我这里在贴一遍脚本，上个主-备中也有脚本，需要灵活修改的地方我红色标识出来：

#!/bin/bash

pidfile=/var/lock/subsys/`basename $`.pid
if [ -f $pidfile ] && [ -e /proc/`cat $pidfile` ] ; then
    exit
fi
trap "rm -rf $pidfile ; exit 0"
echo $$ > $pidfile
maxfails=
fails=
success=

while [  ]
do

    /usr/bin/wget --timeout= --tries= http://192.168.2.11/ -q -O /dev/null && ping -c1 192.168.2.1 &> /dev/null
    if [ $? -ne  ] ; then
        let fails=$[$fails+]
        success=
    else
        fails=
        let success=$[$success+]
    fi

    if [ $fails -ge $maxfails ] ; then
        fails=
        success=
        #check keepalived is running ? try to stop it
        /etc/init.d/keepalived status | grep 正在运行
        if [ $? -eq  ] ; then
            /usr/bin/logger -is "local service fails $maxfails times ... try to stop keepalived."
            /etc/init.d/keepalived stop >& | /usr/bin/logger
        fi

    fi

    if [ $success -gt $maxfails ] ; then
        #check keepalived is stopped ? try to start it
        /etc/init.d/keepalived status | grep 已停
        if [ $? -eq  ] ; then
            /usr/bin/logger -is "service changes normal, try to start keepalived ."
            /etc/init.d/keepalived start
        fi
        success=
    fi
    sleep 

done

注意：此脚本必须有访问默认页面支持。