keepalive配置mysql自动故障转移

本文先配置了一个双master环境，互为主从，然后通过Keepalive配置了一个虚拟IP，客户端通过虚拟IP连接master1,当master1宕机，自动切换到master2。一次只能连接其中一个master进行读写，所以是active-passive模式。

一 Mysql主主复制搭建

1.1 实验环境

两台机器事先都已经装好了mysql单实例。 

二者的端口号需要保持一致，否则在最后用vip连接的时候，不能使用相同端口号连接。 

1.2 实验步骤

1.2.1 修改配置文件

修改master1：

在[mysqld]下面添加：

server-id =  1

relay-log=/data/server/mysql_3307/binlog/ZabbixServer-relay-bin

relay-log-index=/data/server/mysql_3307/binlog/ZabbixServer-relay-bin.index

auto-increment-offset = 1    

auto-increment-increment = 2 

log-slave-updates=true

修改master2：

在[mysqld]下面添加：

server-id =  3

relay-log =/data/server/mysql/binlog/single-relay-bin

relay-log-index=/data/server/mysql/binlog/single-relay-bin.index

auto-increment-offset = 2   

auto-increment-increment = 2 

log-slave-updates=true

添加auto-increment-offset那两项,是为了避免在MySQL INSERT时主键冲突。

修改完后记得重启mysql

1.2.2 建复制用户

分别在两台mysql上执行

GRANT REPLICATION SLAVE ON *.* TO 'RepUser'@'%'identified by 'beijing';

1.2.3 指向master

两台服务器均为新建立，且无其它写入操作，各服务器只需记录当前自己二进制日志文件及事件位置，以之作为另外的服务器复制起始位置即可。否则，需要先备份主库，在备库进行恢复，从而保持数据一致，然后再指向master。

Master1:

mysql> show master status;

+------------------+----------+--------------+------------------+-------------------+

| File            | Position | Binlog_Do_DB | Binlog_Ignore_DB | Executed_Gtid_Set |

+------------------+----------+--------------+------------------+-------------------+

| mysql-bin.000001 |      302|              |                  |                   |

+------------------+----------+--------------+------------------+-------------------+

1 row in set (0.00 sec)

Master2:

mysql> show master status;

+------------------+----------+--------------+------------------+-------------------+

| File            | Position | Binlog_Do_DB | Binlog_Ignore_DB | Executed_Gtid_Set |

+------------------+----------+--------------+------------------+-------------------+

| mysql-bin.000001 |      120 |              |                  |                   |

+------------------+----------+--------------+------------------+-------------------+

1 row in set (0.00 sec)

#Master1指向Master2

 

CHANGE MASTER TO MASTER_USER='RepUser',MASTER_HOST='192.168.1.21',MASTER_PASSWORD='beijing',MASTER_PORT=3307,MASTER_LOG_FILE='mysql-bin.000001',MASTER_LOG_POS=120;

#Master2指向Master1

 

CHANGE MASTER TO MASTER_USER='RepUser',MASTER_HOST='192.168.1.22',MASTER_PASSWORD='beijing', MASTER_PORT=3307,MASTER_LOG_FILE='mysql-bin.000001',MASTER_LOG_POS=302;

1.2.4 分别启动slave

start slave ;

确保show slave status

Slave_IO_Running: Yes

Slave_SQL_Running: Yes

测试两边是否同步，略。

 

二：配置keepalived

1 keepalive安装

分别在master1，master2上安装keepalive

 

yum install -y popt-devel
cd /usr/local/src
wget http://www.keepalived.org/software/keepalived-1.2.2.tar.gz
tar zxvf keepalived-1.2.2.tar.gz
cd keepalived-1.2.2
./configure --prefix=/
make
make install

#假如在执行./configure --prefix=/时报错： OpenSSL is not properly installed on your system !!!Can notinclude OpenSSL headers files，则yum install openssl-devel -y

2 分别在master1，master2上新建检查mysql脚本

vi /root/check_mysql.sh

内容如下

 

[plain] view plain copy

 

MYSQL=/usr/local/mysql/bin/mysql  
MYSQL_HOST=localhost  
MYSQL_USER=root  
MYSQL_PASSWORD=cc.123  
  
  
$MYSQL -h $MYSQL_HOST -u $MYSQL_USER -p$MYSQL_PASSWORD -e "show status;" >/dev/null 2>&1  
#$mysqlclient --host=$host --port=$port --user=$user --password=$password  -e "show databases;" > /dev/null 2>&1  
if [ $? == 0 ]  
then  
    echo " $host mysql login successfully "  
    exit 0  
else  
    #echo " $host mysql login faild"  
    /etc/init.d/keepalived stop  
    exit 2  
fi 

　　

chmod +x /root/check_mysql.sh

3 修改配置文件

vi /etc/keepalived/keepalived.conf

master1和master2配置文件内容相同。

内容：

 

global_defs {
     router_id HA_MySQL
}

vrrp_script check_mysql {                   ######定义监控mysql的脚本  
     script "/root/check_mysql.sh"  
     interval 2                             ######监控时间间隔  
     weight 2                               ######负载参数  
     }  
vrrp_instance VI_1 {
     state BACKUP
     interface eth0
     virtual_router_id 51
     priority 100
     advert_int 1
     nopreempt
     authentication {
         auth_type PASS
         auth_pass 1111
     }
     virtual_ipaddress {
            192.168.60.251
     }
}

track_script {                              ######执行监控mysql进程的脚本  
     check_mysql  
}  

virtual_server 192.168.60.251 3306 {
     delay_loop 2
     lb_algo wrr
     lb_kind DR
     persistence_timeout 60
     protocol TCP
     real_server 192.168.60.168 3306 {
         weight 3
         notify_down /root/shutdown.sh
         TCP_CHECK {
             connect_timeout 10
             nb_get_retry 3
             delay_before_retry 3
             connect_port 3306
         } 
     }
}

　　

这里state不配置MASTER，且优先级一样，是期望在MASTER1宕机后再恢复时，不主动将MASTER状态抢过来，避免MySQL服务的波动。

由于不存在使用lvs进行负载均衡，不需要配置虚拟服务器virtual server，下同。 

4 vi /etc/sysconfig/iptables

#注意，在两台机器上都要修改。

添加：

-A INPUT -d 192.168.1.60/32 -j ACCEPT

-A INPUT -d 224.0.0.18 -j ACCEPT #添加VRRP通讯支持

注意：第一行中的192.168.1.60需要改成你自己的vip。

service iptables restart

5 启动keepalived

在master1、master2上分别启动：
service keepalived start

分别执行ip addr命令，可以在其中一台机器上看到虚拟IP.如：

[root@slave1 keepalived]# 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_faststate UP qlen 1000
    link/ether 08:00:27:04:05:16 brd ff:ff:ff:ff:ff:ff
    inet 192.168.1.22/24 brd 192.168.1.255 scope global eth0
    inet 192.168.1.60/32 scope global eth0
    inet6 fe80::a00:27ff:fe04:516/64 scope link tentativedadfailed 
       valid_lft forever preferred_lft forever

说明虚拟vip连在了master1这台机器上。

如果自动只连接到了master2,关闭master2的keepalived，再启动，自动就连接到master1了。

现在都可以ping通虚拟ip了。

 

 

6 测试

停止master1服务器keepalived，检查VIP是否切换到master2服务器（用ip addr命令验证即可）；
 

三 测试高可用环境是否配置成功

3.1 建允许远程访问的用户

在master1,master2创建允许远程访问的用户：

grant select,update,delete,insert on *.* to 'dandan' identified by 'dandan';

 

3.2 访问虚拟IP

用一台同网段的机器访问通过vip访问数据库：

mysql -u dandan-pdandan -h 192.168.1.60 --port 3307

停止master1服务器的mysql，VIP切换到了master2服务器。

在master2上查看：

 

mysql> showprocesslist;
+----+-------------+--------------------+------+---------+------+-----------------------------------------------------------------------------+------------------+
| Id | User        | Host               | db   | Command | Time | State                                                                      | Info             |
+----+-------------+--------------------+------+---------+------+-----------------------------------------------------------------------------+------------------+
|  3 | root        | localhost          | dba | Query   |    0 | init                                                                       | show processlist |
| 14 | systemuser |                    | NULL |Connect |  247 | Reconnecting after afailed master event read                               | NULL             |
| 15 | systemuser |                    | NULL |Connect |  207 | Slave has read all relaylog; waiting for the slave I/O thread to update it | NULL             |
| 90 |dandan      | 192.168.1.60:39995 |dba  | Sleep   |    8|                                                                            | NULL             |
+----+-------------+--------------------+------+---------+------+-----------------------------------------------------------------------------+------------------+

看到了dandan的连接信息。