本例結合本博另一篇文章 MHA-大殺器實現MYSQL單主宕機時,VIP漂移實現高可用，環境請參考上一篇文章
本例使用keepalived-1.1.20.tar.gz版本
[root@node1 keepalived-1.1.20]# ./configure --sysconf=/etc/ #此項表示設置keepalived的根目錄
如果編譯報錯“configure: error: Popt libraries is required” 則yum -y install popt-devel
[root@node1 keepalived-1.1.20]# make && make install
[root@node1 keepalived]# cp /usr/local/sbin/keepalived /usr/sbin/ #拷貝命令文件
#####################主機的配置文件設置####################
[root@node1 keepalived]# cat /etc/keepalived/keepalived.conf
! Configuration File for keepalived
global_defs {#全局配置項
router_id node1#本機標識號，設置全局唯一
}
vrrp_script check_run {#設置檢測服務的腳本
script "/root/check_mysql.sh"
interval 1#每隔一秒檢測一次
}
vrrp_instance mha { #定義keepalived組
state MASTER#定義本機的keepalived狀態
interface eth0#設置監聽網卡
virtual_router_id 100#設置虛擬router-id，所有keepalived的機器都得是一樣
priority 100 #設置權值，master必須高於backup
advert_int 5#定義master與backup之間同步檢查間隔，單位是秒
track_interface {#設置監聽網卡組
eth0
eth1
}
authentication {#設置keepalived組的驗證類型和密碼，所有主機須一樣
auth_type PASS
auth_pass 1111
}
track_script {#設置服務監測腳本運行
check_run
}
virtual_ipaddress {#設置VIP
10.2.16.250/24
}
}
#####################備機的配置文件設置####################
[root@node2 ~]# cat /etc/keepalived/keepalived.conf
! Configuration File for keepalived
global_defs {
router_id node2
}
vrrp_script check_run {
script "/root/check_mysql.sh"
interval 1
}
vrrp_instance mha {
state BACKUP
interface eth0
virtual_router_id 100
priority 99
advert_int 5
track_interface {
eth0
eth1
}
track_interface {
eth0
eth1
}
track_script {
check_run
}
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
10.2.16.250/24
}
}
mysql服務檢測腳本內容，按需修改：
[root@node2 ~]# cat check_mysql.sh
#!/bin/bash
MYSQL=/usr/bin/mysql
MYSQL_HOST=127.0.0.1
MYSQL_USER=root
MYSQL_PASSWORD=123456
CHECK_TIME=3
#mysql is working MYSQL_OK is 1 , mysql down MYSQL_OK is 0
MYSQL_OK=1
function check_mysql_helth (){
$MYSQL -h $MYSQL_HOST -u $MYSQL_USER -e "show status;" --password=123456 >/dev/null 2>&1
if [ $? = 0 ] ;then
MYSQL_OK=1
else
MYSQL_OK=0
fi
return $MYSQL_OK
}
while [ $CHECK_TIME -ne 0 ]
do
let "CHECK_TIME -= 1"
check_mysql_helth
if [ $MYSQL_OK = 1 ] ; then
CHECK_TIME=0
exit 0
fi
if [ $MYSQL_OK -eq 0 ] && [ $CHECK_TIME -eq 0 ]
then
pkill keepalived
exit 1
fi
sleep 1
done
啓動Keepalived：
[root@node1 ~]#service keepalived start
啓動日誌：
[root@node1 ~]# tail -f /var/log/messages
Jun 27 11:48:47 node2 Keepalived: Starting VRRP child process, pid=21545
Jun 27 11:48:47 node2 Keepalived_vrrp: Registering Kernel netlink reflector
Jun 27 11:48:47 node2 Keepalived_vrrp: Registering Kernel netlink command channel
Jun 27 11:48:47 node2 Keepalived_vrrp: Registering gratutious ARP shared channel
Jun 27 11:48:47 node2 Keepalived_vrrp: Opening file '/etc/keepalived/keepalived.conf'.
Jun 27 11:48:47 node2 Keepalived_vrrp: Configuration is using : 65535 Bytes
Jun 27 11:48:47 node2 Keepalived_vrrp: Using LinkWatch kernel netlink reflector...
Jun 27 11:48:47 node2 Keepalived_vrrp: VRRP_Instance(mha) Entering BACKUP STATE
Jun 27 11:48:47 node2 Keepalived_vrrp: VRRP sockpool: [ifindex(2), proto(112), fd(9,10)]
Jun 27 11:48:47 node2 Keepalived_vrrp: VRRP_Script(check_run) succeeded
Jun 27 11:49:03 node2 Keepalived_vrrp: VRRP_Instance(mha) Transition to MASTER STATE
Jun 27 11:49:08 node2 Keepalived_vrrp: VRRP_Instance(mha) Entering MASTER STATE
Jun 27 11:49:08 node2 Keepalived_vrrp: VRRP_Instance(mha) setting protocol VIPs.
Jun 27 11:49:08 node2 Keepalived_vrrp: VRRP_Instance(mha) Sending gratuitous ARPs on eth0 for 10.2.16.250
Jun 27 11:49:13 node2 Keepalived_vrrp: VRRP_Instance(mha) Sending gratuitous ARPs on eth0 for 10.2.16.250
查看 VIP:
[root@node1 ~]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:c5:22:55 brd ff:ff:ff:ff:ff:ff
inet 10.2.16.253/24 brd 10.2.16.255 scope global eth0
inet 10.2.16.250/24 brd 10.2.16.255 scope global eth0
inet6 fe80::20c:29ff:fec5:2255/64 scope link
valid_lft forever preferred_lft forever
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:c5:22:5f brd ff:ff:ff:ff:ff:ff
inet 10.0.0.1/24 brd 10.0.0.255 scope global eth1
inet6 fe80::20c:29ff:fec5:225f/64 scope link
valid_lft forever preferred_lft forever
此VIP用ifconfig看不到，但是真實存在的。
##########給keepalived 生成的VIP進行mysql登錄授權###########
注意：兩臺機器都需要執行：
mysql> grant all on *.* to 'root'@'10.2.16.250' identified by '123456';
####check_mysql.sh 需要在兩臺機器上都存在
###############keepalived+mha+check_mysql運行原理#################
實現VIP飄移，需要將node1的Mysql服務停掉，此時keepalived中運行的腳本會檢測mysql，檢測到mysql掛了之後，會pkill keepalived
當BACKUP-keepalived檢測到主keepalived掛掉之後，會自動啓動VIP，而MHA檢測到master的mysql掛掉之後，也會將Master移至BAKCUP上
此時所有的權限都移至Mysql-backup上，並且他已經成爲MASTER，接替宕機的master所有工作，從而實現高可用。