一简单介绍
Keepalived+lvs+mysql主主复制是比較经常使用的一种Mysql高可用方案，当中lvs 提供读负载均衡，Keepalived通过虚拟vip漂移实现故障自己主动转移，而Mysql尽管配置成主主复制。可是一般都是採用单点写入来保证数据一致性。
整个方案实现很easy，仅需两台server就可以，安装配置也相对简单，但须要自己写脚本实现keepalived服务的开启和关闭，适用于仅仅有两台数据库server而且未实现读写分离的场景。
系统总体结构框图例如以下
二环境搭建
2.1 环境配置例如以下
|
Keepalived Server |
LVS Server |
LVS Realserver |
MySQL Master-Master |
Write VIP |
Read VIP |
OS |
MySQL |
|
Master 10.0.11.243 |
10.0.11.243 |
10.0.11.243 |
10.0.11.243 |
10.0.11.190 |
10.0.11.191 |
Centos6 |
Percona5.6 |
|
Backup 10.0.11.244 |
10.0.11.244 |
10.0.11.244 |
10.0.11.244 |
10.0.11.190 |
10.0.11.191 |
Centos6 |
Percona5.6 |
2.2 mysql master-master复制配置
主主复制配置本质上就是配置两台server互相的主从，实现比較简单，大致过程例如以下
2.2.1 在Master上创建一个复制账户
mysql>grantreplication slave on *.* to rep71@10.0.11.244identified by‘123456’;
mysql>flushprivileges;
2.2.2 改动两台mysql配置文件my.cnf例如以下
##Replication General Config (both master and slave)
server-id= 101
log-bin= mysql-log-bin
auto-increment-increment= 2
auto-increment-offset= 1
binlog_format= row
relay_log= mysql-relay-bin
2.2.3 在Masterserver上运行show master status，记录下File和Position
mysql>show master statusG;
***************************1. row ***************************
File:mysql-log-bin.000005
Position:327933
Binlog_Do_DB:
Binlog_Ignore_DB:test,mysql,information_schema,performance_schema
2.2.4 在Slave服务上运行
mysql> slave stop;
Query OK, 0 rows affected, 1 warning (0.00 sec)
mysql> change master to
-> master_host='10.0.11.243',
-> master_user='rep71',
-> master_password='123456',
-> master_log_file=' mysql-log-bin.000005',
-> master_log_pos=327933;
Query OK, 0 rows affected (0.01 sec)
mysql> start slave;
2.2.5反复以上四个步骤。在Slaveserver上创建账户rep72，show master status。在Master上start slave，这样mysql的主主同步就配置好了，分别在两台server上执行show slave status，假设结果都显示Slave_IO_Running: Yes，Slave_SQL_Running: Yes。则证明同步正常。
2.3 Keepalived的安装
2.3.1 在 master、backup server都进行安装：
wget http://www.linuxvirtualserver.org/software/kernel-2.6/ipvsadm-1.24.tar.gz
sudotar zxvf ipvsadm-1.24.tar.gz
cdipvsadm-1.24
sudoln -s /usr/src/kernels/2.6.32-358.el6.x86_64/ /usr/src/linux
make
sudomake install
2.4 Keepalived 的安装
2.4.1在 master、backup server都进行安装：
wget http://www.keepalived.org/software/keepalived-1.1.19.tar.gz
tarzxvf keepalived-1.1.19.tar.gz
cdkeepalived-1.1.19
yuminstall -y openssl openssl-devel
sudoyum install popt-devel –y
./configure--sysconf=/etc/ --with-kernel-dir=/usr/src/kernels/2.6.18-308.el5-x86_64/
make
sudomake install
ln -s /usr/local/sbin/keepalived /sbin/
2.4.2 Master上的Keepalived 的配置
global_defs{
router_id mysqlmha1
}
vrrp_scriptcheck_run {
script"/etc/keepalived/check_mysqll.sh"
interval 1
}
vrrp_instanceVI_1 {
state MASTER
interface em1
virtual_router_id 51
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
track_script {
check_run
}
virtual_ipaddress {
10.0.11.190 dev em1 label em1:1
10.0.11.191 dev em1 label em1:2
}
}
virtual_server10.0.11.190 3306 {
delay_loop6
lb_algorr
lb_kindDR
persistence_timeout20
protocolTCP
sorry_server10.0.11.244 3306
real_server10.0.11.243 3306 {
weight3
TCP_CHECK{
connect_timeout3
nb_get_retry3
delay_before_retry3
connect_port3306
}
}
}
virtual_server10.0.11.191 3306 {
delay_loop6
lb_algorr
lb_kindDR
#persistence_timeout20
protocolTCP
real_server10.0.11.243 3306 {
weight3
TCP_CHECK{
connect_timeout3
nb_get_retry3
delay_before_retry3
connect_port3306
}
}
real_server10.0.11.244 3306 {
weight3
TCP_CHECK{
connect_timeout3
nb_get_retry3
delay_before_retry3
connect_port3306
}
}
}
2.4.3 Backup 的 keepalived 的配置
global_defs {
router_id mysqlmha1
}
vrrp_script check_run {
script "/etc/keepalived/check_mysqll.sh"
interval 1
}
vrrp_instance VI_1 {
state BACKUP
interface em1
virtual_router_id 51
priority 90
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
track_script {
check_run
}
virtual_ipaddress {
10.0.11.190 dev em1 label em1:1
10.0.11.191 dev em1 label em1:2
}
}
virtual_server 10.0.11.190 3306 {
delay_loop 6
lb_algo rr
lb_kind DR
persistence_timeout 20
protocol TCP
sorry_server 10.0.11.244 3306
real_server 10.0.11.243 3306 {
weight 3
TCP_CHECK {
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
connect_port 3306
}
}
}
virtual_server 10.0.11.191 3306 {
delay_loop 6
lb_algo rr
lb_kind DR
#persistence_timeout 20
protocol TCP
real_server 10.0.11.243 3306 {
weight 3
TCP_CHECK {
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
connect_port 3306
}
}
real_server 10.0.11.244 3306 {
weight 3
TCP_CHECK {
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
connect_port 3306
}
}
}
2.5 Master 和 backup 的 realserver 的配置
对于 realserver 的配置 master 和 backup 是一致的，脚本内容例如以下：
#!/bin/bash
#description: Config realserver lo and apply noarp
SNS_VIP=10.1.11.190
SNS_VIP2=10.0.11.191
/etc/rc.d/init.d/functions
case"$1" in
start)
ifconfiglo:0 $SNS_VIP netmask 255.255.255.255 broadcast $SNS_VIP
ifconfiglo:1 $SNS_VIP2 netmask 255.255.255.255 broadcast $SNS_VIP2
/sbin/routeadd -host $SNS_VIP dev lo:0
/sbin/routeadd -host $SNS_VIP2 dev lo:1
echo"1" >/proc/sys/net/ipv4/conf/lo/arp_ignore
echo"2" >/proc/sys/net/ipv4/conf/lo/arp_announce
echo"1" >/proc/sys/net/ipv4/conf/all/arp_ignore
echo"2" >/proc/sys/net/ipv4/conf/all/arp_announce
echo"1" >/proc/sys/net/ipv4/conf/em1/arp_ignore
echo"2" >/proc/sys/net/ipv4/conf/em1/arp_announce
sysctl-p >/dev/null 2>&1
echo"RealServer Start OK"
;;
stop)
ifconfiglo:0 down
ifconfiglo:1 down
routedel $SNS_VIP >/dev/null 2>&1
routedel $SNS_VIP2 >/dev/null 2>&1
echo"0" >/proc/sys/net/ipv4/conf/lo/arp_ignore
echo"0" >/proc/sys/net/ipv4/conf/lo/arp_announce
echo"0" >/proc/sys/net/ipv4/conf/all/arp_ignore
echo"0" >/proc/sys/net/ipv4/conf/all/arp_announce
echo"0" >/proc/sys/net/ipv4/conf/em1/arp_ignore
echo"0" >/proc/sys/net/ipv4/conf/em1/arp_announce
echo"RealServer Stoped"
;;
*)
echo"Usage: $0 {start|stop}"
exit1
esac
exit0
2.6 编辑脚本控制Keepalived的开启和停止
大体意思是仅仅要检測到mysql服务停止keepalived服务也停止 ，由于keepalived是通过组播方式告诉本网段自己还活着当mysql服务停止后keepalived还依旧执行 这时就须要停止keepalived让还有一个主机获得虚拟IP，能够在后台执行这个脚本 也能够在keepalived配置文件增加这个脚本，master和slave节点都须要，仅仅是数据库username和password有差别
[root@244 ~]#more /etc/keepalived/check_mysql.sh
#20140722
#!/bin/bash
MYSQL=/usr/bin/mysql
MYSQL_HOST=10.0.11.243
MYSQL_USER=root
MYSQL_PASSWORD=mysql
CHECK_TIME=3
#mysql isworking MYSQL_OK is 1 , mysql down MYSQL_OK is 0
MYSQL_OK=1
function check_mysql_helth (){
$MYSQL -h $MYSQL_HOST -u $MYSQL_USER -p$MYSQL_PASSWORD -e "showstatus;" >/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
该脚本须要有运行权限。通过下面命令加上运行权限
# chmod +x check_mysql.sh
2.7 測试Keepalived自己主动故障转移
Master和slave上运行sudo servicekeepalived start，在master上运行ip a
Slave上运行 ip a
从上能够看到masteMySQL上虚拟IP绑定成功，backup MySQL上keepalived服务正常执行，但无虚拟IP(这属于正常)
client能通过vip连接上mysql
停止master上的mysql服务，可发现Keepalived服务也会停止
这时候slave执行ip a发现绑定了虚拟ip，而且从库mysql执行正常，而client依然能通过vip连上数据库
重新启动master的mysql和Keepalived以后，并在slave库执行start slave，master执行ip a发现恢复之前的服务
2.8 測试lvs的负载均衡效果
启动 master 和 backup 的 mysql 以后，再在 master 和 backup 运行例如以下命令启
动 keepalived 和 realserver 脚本：
/etc/rc.d/init.d/realserver.sh start
/etc/rc.d/init.d/keepalived start
执行master和backup上的sudo ipvsadm -ln可看到例如以下信息
在client通过read vip连接到mysql并运行几次select,可考到inaction发生了变化。证明实现了负载均衡
