利用keepalived构建高可用MySQL High Application
关于MySQL-HA，目前有多种解决方案，比如heartbeat、drbd、mmm、共享存储，但是它们各有优缺点。heartbeat、 drbd配置较为复杂，需要自己写脚本才能实现MySQL自动切换；对于mmm，生产环境中很少有人 用，且mmm 管理端需要单独运行一台服务器上，要是想实现高可用，就得对mmm管理端做HA，这样无疑又增加了硬件开支；对于共享存储，个人觉得MySQL数据还是放 在本地较为安全，存储设备毕竟存在单点隐患。
使用MySQL双master+keepalived是一种非常好的解决方案，在MySQL-HA环境中，MySQL互为主从关系，这样就保证 了两台MySQL数据的一致性，然后用keepalived实现虚拟IP，通过keepalived自带的服务监控功能来实现MySQL故障时自动切换。
下面是MySQL-HA是实现方式：
一、环境
MySQL-VIP：10.0.5.76
MySQL-master1：10.0.5.41
MySQL-master2：10.0.5.75
OS版本：ubuntu 14.04 LTS
MySQL版本：5.5.43
Keepalived版本：v1.2.7 (08/14,2013)
二、Mysql Master to Master 配置
1、配置MySQL-master1：10.0.5.41
a) 修改mysql配置文件
vim /etc/mysql/my.cnf
user = mysql
log-bin=mysql-bin
server-id=1
binlog-ignore-db=mysql
replicate-ignore-db=mysql,information_schema
sync_binlog=1
auto_increment_offset=1
auto_increment_increment=2
slave-skip-errors=all
b) 新建授权用户
/etc/init.d/mysql restart
mysql>CREATE USER 'replication'IDENTIFIED BY '123456';
mysql>grant replication slave on *.* to 'replication'@'10.0.5.75' identified by '123456' with grant option;
mysql> show master status \G
*************************** 1. row ***************************
File: mysql-bin.000032
Position: 107
Binlog_Do_DB:
Binlog_Ignore_DB: mysql
1 row in set (0.00 sec)
c) 将10.0.5.75设置为10.0.5.41的主服务器
mysql>stop slave;
mysql>change master to master_host='10.0.5.75',master_user='replication',master_password='123456', master_log_file='mysql-bin.000006',master_log_pos=107;
mysql>start slave;
mysql> show slave status\G
*************************** 1. row ***************************
Slave_IO_State: Waiting for master to send event
Master_Host: 10.0.5.75
Master_User: replication
Master_Port: 3306
Connect_Retry: 60
Master_Log_File: mysql-bin.000019
Read_Master_Log_Pos: 107
Relay_Log_File: mysqld-relay-bin.000002
Relay_Log_Pos: 333
Relay_Master_Log_File: mysql-bin.000006
Slave_IO_Running: Yes
Slave_SQL_Running: Yes
Replicate_Do_DB:
Replicate_Ignore_DB: mysql,information_schema
Replicate_Do_Table:
Replicate_Ignore_Table:
Replicate_Wild_Do_Table:
Replicate_Wild_Ignore_Table:
Last_Errno: 0
Last_Error:
Skip_Counter: 0
Exec_Master_Log_Pos: 187
Relay_Log_Space: 490
Until_Condition: None
Until_Log_File:
Until_Log_Pos: 0
Master_SSL_Allowed: No
Master_SSL_CA_File:
Master_SSL_CA_Path:
Master_SSL_Cert:
Master_SSL_Cipher:
Master_SSL_Key:
Seconds_Behind_Master: 0
Master_SSL_Verify_Server_Cert: No
Last_IO_Errno: 0
Last_IO_Error:
Last_SQL_Errno: 0
Last_SQL_Error:
Replicate_Ignore_Server_Ids:
Master_Server_Id: 1
2、配置MySQL-master2：10.0.5.75
a) 修改配置文件
vim /etc/mysql/my.cnf
user = mysql
log-bin=mysql-bin
server-id=2
binlog-ignore-db=mysql
replicate-ignore-db=mysql,information_schema
sync_binlog=1
auto_increment_offset=2
auto_increment_increment=2
slave-skip-errors=all
b) 新建授权用户
/etc/init.d/mysql restart
mysql>CREATE USER 'replication'IDENTIFIED BY '123456';
mysql>grant replication slave on *.* to 'replication'@'10.0.5.41' identified by '123456' with grant option;
mysql> show master status\G
*************************** 1. row ***************************
File: mysql-bin.000019
Position: 107
Binlog_Do_DB:
Binlog_Ignore_DB: mysql
1 row in set (0.00 sec)
c) 将10.0.5.41设置为10.0.5.75的主服务器
mysql>stop slave;
mysql>change master to master_host='10.0.5.41',master_user='replication',master_password='123456', master_log_file='mysql-bin.000006',master_log_pos=107;
mysql>start slave;
mysql> show slave status\G
*************************** 1. row ***************************
Slave_IO_State: Waiting for master to send event
Master_Host: 10.0.5.41
Master_User: replication
Master_Port: 3306
Connect_Retry: 60
Master_Log_File: mysql-bin.000032
Read_Master_Log_Pos: 107
Relay_Log_File: mysqld-relay-bin.000002
Relay_Log_Pos: 333
Relay_Master_Log_File: mysql-bin.000010
Slave_IO_Running: Yes
Slave_SQL_Running: Yes
Replicate_Do_DB: chris,ian
Replicate_Ignore_DB: mysql,information_schema
Replicate_Do_Table:
Replicate_Ignore_Table:
Replicate_Wild_Do_Table:
Replicate_Wild_Ignore_Table:
Last_Errno: 0
Last_Error:
Skip_Counter: 0Exec_Master_Log_Pos: 187
Relay_Log_Space: 490
Until_Condition: None
Until_Log_File:
Until_Log_Pos: 0
Master_SSL_Allowed: No
Master_SSL_CA_File:
Master_SSL_CA_Path:
Master_SSL_Cert:
Master_SSL_Cipher:
Master_SSL_Key:
Seconds_Behind_Master: 0
Master_SSL_Verify_Server_Cert: No
Last_IO_Errno: 0
Last_IO_Error:
Last_SQL_Errno: 0
Last_SQL_Error:
Replicate_Ignore_Server_Ids:
Master_Server_Id: 2
3、MySQL同步测试
如上述均正确配置，现在任何一台MySQL上更新数据都会同步到另一台MySQL。
三、配置keepalived
1、在10.0.5.41上配置keepalived
apt-get install keepalived
vim /etc/keepalived/keepalived.conf
# Global Configuration
global_defs {
router_id MySQL-ha
}
# VRRP Configuration
vrrp_instance VI_1 {
state BACKUP
interface eth0
virtual_router_id 51
priority 90
advert_int 1
authentication {
auth_type PASS
auth_pass root
}
virtual_ipaddress {
10.0.5.76 # * VIP
}
}
# Virtual Server Configuration
virtual_server 10.0.5.76 3306 {
delay_loop 2
lb_algo wrr
lb_kind DR
persistence_timeout 60
protocol TCP
# Real Server 1 configuration
real_server 10.0.5.41 3306 {
weight 3
notify_down /etc/keepalived/kill_keepalived.sh
TCP_CHECK {
connection_timeout 10
nb_get_retry 3
delay_before_retry 3
connect_port=3306
}
}
}
编写检测服务down后所要执行的脚本
vim /etc/keepalived/kill_keepalived.sh
#!/bin/sh
pkill keepalived
chmod +x /etc/keepalived/kill_keepalived.sh
/etc/init.d/keepalived start
注：此脚本是上面配置文件notify_down选项所用到的，keepalived使用notify_down选项来检查real_server的服务状态，当发现real_server服务故障时，便触发此脚本；我们可以看到，脚本就一个命令，通过pkill keepalived强制杀死keepalived进程，从而实现了MySQL故障自动转移。另外，我们不用担心两个MySQL会同时提供数据更新操作， 因为每台MySQL上的keepalived的配置里面只有本机MySQL的IP+VIP，而不是两台MySQL的IP+VIP。
2、在10.0.5.75上配置keepalived
apt-get install keepalived
vim /etc/keepalived/keepalived.conf
# Global Configuration
global_defs {
router_id MySQL-ha
}
# VRRP Configuration
vrrp_instance VI_1 {
state MASTER
interface eth0
virtual_router_id 51
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass root
}
virtual_ipaddress {
10.0.5.76 # * VIP
}
}
# Virtual Server Configuration
virtual_server 10.0.5.76 3306 {
delay_loop 2
lb_algo wrr
lb_kind DR
persistence_timeout 60
protocol TCP
# Real Server 1 configuration
real_server 10.0.5.75 3306 {
weight 3
notify_down /etc/keepalived/kill_keepalived.sh
TCP_CHECK {
connection_timeout 10
nb_get_retry 3
delay_before_retry 3
connect_port=3306
}
}
}
编写检测服务down后所要执行的脚本
vim /etc/keepalived/kill_keepalived.sh
#!/bin/sh
pkill keepalived
chmod +x /etc/keepalived/kill_keepalived.sh
/etc/init.d/keepalived start
停止MySQL服务，看keepalived健康检查程序是否会触发我们编写的脚本。
四、测试
MySQL远程登录测试，选择一台内网机器10.0.5.74远程登录VIP，看是否能登录，在登录之前两台MySQL服务器都要授权允许从远程登录。
MySQL> grant all privileges on *.* to 'root'@'%' identified by 'root';
MySQL> flush privileges;
远程登录测试
mysql -uroot -proot -h 10.0.5.76
Welcome to the MySQL monitor. Commands end with ; or \g.
Your MySQL connection id is 10818
Server version: 5.5.43-0ubuntu0.14.04.1-log (Ubuntu)
Copyright (c) 2000, 2015, Oracle and/or its affiliates. All rights reserved.
Oracle is a registered trademark of Oracle Corporation and/or its
affiliates. Other names may be trademarks of their respective
owners.
Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.
mysql>
keepalived故障转移测试
从10.0.5.74一直去ping VIP，然后关闭10.0.5.41上的keepalived，正常情况下VIP就会切换到10.0.5.75上面去
开启10.0.5.75上的keepalived，关闭10.0.5.41上的keepalived，看是否能自动切换，正常情况下VIP又会属于10.0.5.41
MySQL故障转移测试
在10.0.5.41上关闭MySQL服务，看VIP是否会切换到10.0.5.75上
开启10.0.5.41上的MySQL和keepalived，然后关闭10.0.5.75上的MySQL，看VIP是否会切换到10.0.5.41上
版权所有@chrisDuan 博客地址http://www.cnblogs.com/chrisDuan 可以转载，注明出处.
