在之前一篇使用nginx搭建高可用的解决方案的时候，很多同学会问，如果nginx挂掉怎么办，比如下面这张图：
你可以清楚的看到，如果192.168.2.100这台机器挂掉了，那么整个集群就下线了，这个问题该怎么解决呢？？？ 简单的想想确实不大好处理，因为你
的webBrowser总得要访问一个ip地址，对吧。。这个问题怎么破呢？
一：问题分析
如果你有一些网络底子的话，就会明白，你给一个不在本网段的机器发送请求的话，这个请求会先经过你的网关IP，然后通过网关IP传给对方的网关IP，然
后网关IP会将请求转给它所在局域网的主机，当然我的网关IP和对方的网关IP之间可能有很多跳的路由地址，大概的流程就是下面这样：
如果你不信的话，可以用tracert 看看你到www.ctrip.com的路由总过程。
从上图中可以看到，从我当前主机到ctrip.com一共有20跳，第一条是192.168.2.1，这个就是我的路由器IP，也就是我的网关IP。
二：虚拟路由冗余协议
好了，说了这么多有什么用呢？其实大家仔细观察这张图，你会想我能不能在网关IP上做一些手脚呢？ 可喜的是如今的路由器基本上都支持一个叫做
VRRP（虚拟路由冗余协议），这一协议的作用你可以理解成把网关IP虚拟化成一个网关IP集群，就好像兽族剑圣的镜像技能，这里面有master，也有slave，
然后局域网内的主机设置的都是虚拟的masterIP（VIP），刚好keepealived就是一个实现VRRP的一款应用程序，你需要，我专业，大家就这样走到一块了。
三：keepalived搭建一览
1. 下载：从官网上找到当前最新的版本1.4.2。http://www.keepalived.org/software/keepalived-1.4.2.tar.gz。
配置机器： 192.168.23.156 【centos】
192.168.23.157 【centos】
[root@localhost app]# wget http://www.keepalived.org/software/keepalived-1.4.2.tar.gz
---- ::-- http://www.keepalived.org/software/keepalived-1.4.2.tar.gz
Resolving www.keepalived.org (www.keepalived.org)... 37.59.63.157, :41d0::7a9d::
Connecting to www.keepalived.org (www.keepalived.org)|37.59.63.157|:... connected.
HTTP request sent, awaiting response... OK
Length: (721K) [application/x-gzip]
Saving to: ‘keepalived-1.4..tar.gz’ %[==================================================================>] , .24KB/s in 4m 44s
2. 然后把相关依赖装起来：yum install -y openssl openssl-devel。
[root@localhost app]# yum install -y openssl openssl-devel
Loaded plugins: fastestmirror, langpacks
Could not retrieve mirrorlist http://mirrorlist.centos.org/?release=7&arch=x86_64&repo=os&infra=stock error was
: curl# - "Empty reply from server"
base | 3.6 kB ::
epel/x86_64/metalink | 7.8 kB ::
Could not retrieve mirrorlist http://mirrorlist.centos.org/?release=7&arch=x86_64&repo=extras&infra=stock error was
: curl# - "Empty reply from server"
extras | 3.4 kB ::
updates | 3.4 kB ::
updates//x86_64/primary_db | 6.9 MB ::
3. 接下来继续解压，最后安装三板斧： ./configure --prefix=/usr/app/keepalived && make && make install。
[root@localhost app]# tar -zxvf keepalived-1.4..tar.gz
[root@localhost app]# ls
keepalived-1.4. keepalived-1.4..tar.gz
[root@localhost app]# cd keepalived-1.4.
[root@localhost keepalived-1.4.]# ls
aclocal.m4 bin_install configure COPYING genhash keepalived Makefile.am README.md
ar-lib ChangeLog configure.ac depcomp INSTALL keepalived.spec.in Makefile.in snap
AUTHOR compile CONTRIBUTORS doc install-sh lib missing TODO
[root@localhost keepalived-1.4.]#
[root@localhost keepalived-1.4.2]# ./configure --prefix=/usr/app/keepalived && make && make install
安装好了之后，你就会看到如下的内容，那就恭喜你，安装成功了。
Keepalived configuration
------------------------
Keepalived version : 1.4.
Compiler : gcc
Preprocessor flags :
Compiler flags : -Wall -Wunused -Wstrict-prototypes -Wextra -g -O2 -fPIE -D_GNU_SOURCE
Linker flags : -pie
Extra Lib : -lcrypto -lssl
Use IPVS Framework : Yes
IPVS use libnl : No
IPVS syncd attributes : No
IPVS bit stats : No
fwmark socket support : Yes
Use VRRP Framework : Yes
Use VRRP VMAC : Yes
Use VRRP authentication : Yes
With ip rules/routes : Yes
SNMP vrrp support : No
SNMP checker support : No
SNMP RFCv2 support : No
SNMP RFCv3 support : No
DBUS support : No
SHA1 support : No
Use Debug flags : No
smtp-alert debugging : No
Use Json output : No
Stacktrace support : No
Memory alloc check : No
libnl version : None
Use IPv4 devconf : No
Use libiptc : No
Use libipset : No
init type : systemd
Build genhash : Yes
Build documentation : No
4. 安装好了之后，在/usr/app/keepalived/etc/keepalived目录下有一个keepalived.conf文件，现在你要做的事情就是
将它copy到/etc/keepalived文件夹下就可以了。
[root@localhost keepalived]# ls
keepalived.conf samples
[root@localhost keepalived]# pwd
/usr/app/keepalived/etc/keepalived
[root@localhost keepalived]# mkdir -p /etc/keepalived
[root@localhost keepalived]# cp ./keepalived.conf /etc/keepalived/keepalived.conf
5. 接下来我们改一下配置文件。
在192.168.23.156机器中的配置文件，修改如下：
【原来】
global_defs {
notification_email {
acassen@firewall.loc
failover@firewall.loc
sysadmin@firewall.loc
}
notification_email_from Alexandre.Cassen@firewall.loc
smtp_server 192.168.200.1
smtp_connect_timeout
router_id LVS_DEVEL
vrrp_skip_check_adv_addr
vrrp_strict
vrrp_garp_interval
vrrp_gna_interval
}
vrrp_instance VI_1 {
state MASTER
interface eth0
virtual_router_id
priority
advert_int
authentication {
auth_type PASS
auth_pass
}
virtual_ipaddress {
192.168.200.16
192.168.200.17
192.168.200.18
}
}
【修改】
global_defs {
notification_email {
acassen@firewall.loc
failover@firewall.loc
sysadmin@firewall.loc
}
notification_email_from Alexandre.Cassen@firewall.loc
smtp_server 192.168.200.1
smtp_connect_timeout
router_id NodeA
vrrp_skip_check_adv_addr
vrrp_strict
vrrp_garp_interval
vrrp_gna_interval
}
vrrp_instance VI_1 {
state MASTER
interface ens33
virtual_router_id
priority 150
advert_int
authentication {
auth_type PASS
auth_pass
}
virtual_ipaddress {
192.168.2.200
}
}
其中要注意的就是：
《1》priority 150
节点的优先级，master要比slave高。
《2》interface ens33
ens33大家可以通过ipconfig查看一下自己的网卡。
[root@localhost ~]# ifconfig
br-11757db6abf5: flags=<UP,BROADCAST,MULTICAST> mtu
inet 172.22.0.1 netmask 255.255.0.0 broadcast 0.0.0.0
ether ::c2:e0:: txqueuelen (Ethernet)
RX packets bytes (0.0 B)
RX errors dropped overruns frame
TX packets bytes (0.0 B)
TX errors dropped overruns carrier collisions br-875e3c64ec79: flags=<UP,BROADCAST,MULTICAST> mtu
inet 172.23.0.1 netmask 255.255.0.0 broadcast 0.0.0.0
ether ::4e::5b:a0 txqueuelen (Ethernet)
RX packets bytes (0.0 B)
RX errors dropped overruns frame
TX packets bytes (0.0 B)
TX errors dropped overruns carrier collisions br-904f2c62861e: flags=<UP,BROADCAST,MULTICAST> mtu
inet 172.19.0.1 netmask 255.255.0.0 broadcast 0.0.0.0
ether ::6d::: txqueuelen (Ethernet)
RX packets bytes (0.0 B)
RX errors dropped overruns frame
TX packets bytes (0.0 B)
TX errors dropped overruns carrier collisions br-b0028a425959: flags=<UP,BROADCAST,MULTICAST> mtu
inet 172.21.0.1 netmask 255.255.0.0 broadcast 0.0.0.0
ether ::::: txqueuelen (Ethernet)
RX packets bytes (1.4 KiB)
RX errors dropped overruns frame
TX packets bytes (3.8 KiB)
TX errors dropped overruns carrier collisions br-c4a09a75fc67: flags=<UP,BROADCAST,MULTICAST> mtu
inet 172.18.0.1 netmask 255.255.0.0 broadcast 0.0.0.0
ether :::f9:2d: txqueuelen (Ethernet)
RX packets bytes (340.0 B)
RX errors dropped overruns frame
TX packets bytes (340.0 B)
TX errors dropped overruns carrier collisions br-f0fb207788a0: flags=<UP,BROADCAST,RUNNING,MULTICAST> mtu
inet 172.20.0.1 netmask 255.255.0.0 broadcast 0.0.0.0
inet6 fe80:::86ff:fe1e:c970 prefixlen scopeid 0x20<link>
ether :::1e:c9: txqueuelen (Ethernet)
RX packets bytes (0.0 B)
RX errors dropped overruns frame
TX packets bytes (0.0 B)
TX errors dropped overruns carrier collisions docker0: flags=<UP,BROADCAST,MULTICAST> mtu
inet 172.17.0.1 netmask 255.255.0.0 broadcast 0.0.0.0
ether ::8f:8c:a9:a7 txqueuelen (Ethernet)
RX packets bytes (0.0 B)
RX errors dropped overruns frame
TX packets bytes (0.0 B)
TX errors dropped overruns carrier collisions ens33: flags=<UP,BROADCAST,RUNNING,MULTICAST> mtu
inet 192.168.23.157 netmask 255.255.255.0 broadcast 192.168.23.255
inet6 fe80::20c:29ff:fe54:4f5a prefixlen scopeid 0x20<link>
ether :0c:::4f:5a txqueuelen (Ethernet)
RX packets bytes (10.8 MiB)
RX errors dropped overruns frame
TX packets bytes (585.6 KiB)
TX errors dropped overruns carrier collisions lo: flags=<UP,LOOPBACK,RUNNING> mtu
inet 127.0.0.1 netmask 255.0.0.0
inet6 :: prefixlen scopeid 0x10<host>
loop txqueuelen (Local Loopback)
RX packets bytes (340.0 B)
RX errors dropped overruns frame
TX packets bytes (340.0 B)
TX errors dropped overruns carrier collisions veth4d72ad4: flags=<UP,BROADCAST,RUNNING,MULTICAST> mtu
inet6 fe80::d4fb:fcff:feeb:cd7c prefixlen scopeid 0x20<link>
ether d6:fb:fc:eb:cd:7c txqueuelen (Ethernet)
RX packets bytes (1.2 KiB)
RX errors dropped overruns frame
TX packets bytes (4.0 KiB)
TX errors dropped overruns carrier collisions vethe634b1c: flags=<UP,BROADCAST,RUNNING,MULTICAST> mtu
inet6 fe80::e0c6:88ff:fe1c:f4a1 prefixlen scopeid 0x20<link>
ether e2:c6::1c:f4:a1 txqueuelen (Ethernet)
RX packets bytes (1.4 KiB)
RX errors dropped overruns frame
TX packets bytes (3.8 KiB)
TX errors dropped overruns carrier collisions virbr0: flags=<UP,BROADCAST,MULTICAST> mtu
inet 192.168.122.1 netmask 255.255.255.0 broadcast 192.168.122.255
ether ::::: txqueuelen (Ethernet)
RX packets bytes (0.0 B)
RX errors dropped overruns frame
TX packets bytes (0.0 B)
TX errors dropped overruns carrier collisions [root@localhost ~]#
《3》virtual_ipaddress 192.168.23.200
设置好虚拟IP（VIP）为：192.168.23.200
同样的道理，在192.168.23.157设置如下：
global_defs {
notification_email {
acassen@firewall.loc
failover@firewall.loc
sysadmin@firewall.loc
}
notification_email_from Alexandre.Cassen@firewall.loc
smtp_server 192.168.200.1
smtp_connect_timeout
router_id NodeB
vrrp_skip_check_adv_addr
vrrp_strict
vrrp_garp_interval
vrrp_gna_interval
}
vrrp_instance VI_1 {
state BACKUP
interface ens33
virtual_router_id
priority
advert_int
authentication {
auth_type PASS
auth_pass
}
virtual_ipaddress {
192.168.23.200
}
}
7. 接下来就可以启动keepalived了。
[root@localhost sbin]# ./keepalived -D
[root@localhost sbin]# ps -ef | grep keepalived
root : ? :: ./keepalived -D
root : ? :: ./keepalived -D
root : ? :: ./keepalived -D
root : pts/ :: grep --color=auto keepalived
五：检测
1. 通过ip a 看看当前ens33网卡上是否绑定了192.168.23.200虚拟IP。
2. 然后通过arp -a 查看当前的vip映射到的物理(mac)地址，可以看到当前的vip映射到的是192.168.23.156上面。
C:\Users\hxc>arp -a 接口: 192.168.23.1 --- 0x6
Internet 地址 物理地址 类型
192.168.23.156 00-0c-29-75-7e-20 动态
192.168.23.157 -0c---4f-5a 动态
192.168.23.200 00-0c-29-75-7e-20 动态
192.168.23.255 ff-ff-ff-ff-ff-ff 静态
224.0.0.22 --5e--- 静态
224.0.0.251 --5e---fb 静态
224.0.0.252 --5e---fc 静态
239.11.20.1 --5e-0b-- 静态
239.255.255.250 --5e-7f-ff-fa 静态
255.255.255.255 ff-ff-ff-ff-ff-ff 静态
3. 然后我们把156这台机器关闭了，可以看到当前的vip已经漂移到了157这台机器上了。
C:\Users\hxc>arp -a 接口: 192.168.23.1 --- 0x6
Internet 地址 物理地址 类型
192.168.23.156 -0c---7e- 动态
192.168.23.157 00-0c-29-54-4f-5a 动态
192.168.23.200 00-0c-29-54-4f-5a 动态
192.168.23.255 ff-ff-ff-ff-ff-ff 静态
224.0.0.22 --5e--- 静态
224.0.0.251 --5e---fb 静态
224.0.0.252 --5e---fc 静态
239.11.20.1 --5e-0b-- 静态
239.255.255.250 --5e-7f-ff-fa 静态
255.255.255.255 ff-ff-ff-ff-ff-ff 静态
好了，这个就是本篇所说的所有内容，希望对您有帮助。