前言：Mysql在生产环境中的宕机率特别的高，一般我们在该机器上只安装mysql，不作他用。数据库本 身也是极易产生瓶颈的地方。Master：主（用于写入数据） Slave：从（用于读取数据），也是 实现读写分离的重要举措。国内很多的视频直播网站、bilibili等弹幕类的网站，在视频框架上 做了个透明的弹幕， 后台数据库每秒发生数以万计的读写操作，后台运维人员真是想 死的心都有。。。哈哈。。废话不多说，开始部署……
下面开始配置主从关系，配置完恢复快照到mysql编译好的环境继续主主关系的配置
准备前的步骤：关闭selinux，关闭防火墙、已编译好的mysql5.5
实验环境：Centos6.5 IP：192.168.1.1 Mysql主
Centos6.5 IP：192.168.1.2 Mysql从
配置主
[root@localhost ~]# mysql -u root -predhat //mysql登录用户及密码
mysql> show dabaseses; //显示数据库
mysql> use mysql; //使用mysql库
mysql> show tables; //显示表
mysql> desc user; //查看user表的结构
mysql> select Host,User,Password from mysql.user; //在user表中查询主机，用户密码字段
mysql> insert into mysql.user(Host,User,Password) values('localhost','baidu',password('123.com')); //插入一条数据，用户百度，密码123.com
mysql> grant replication slave on *.* to 'baidu'@'192.168.1.2' identified by '123.com' with grant option; //授权给192.168.1.2用户可以使用百度这个用户，密码123.com来进行复制操作
mysql> flush privileges; //刷新授权
mysql> create database testdb; //创建测试的库，库名为testdb
[root@localhost ~]# vim /etc/my.cnf
大约49行log-bin=mysql-bin处
添加：binlog-do-db=testdb //将test库写入二进制日志
binlog-ignore-db=mysql //忽视mysql库写入二进制日志
修改：server-id = 1 每台做主从或主主的server id的值应唯一。
[root@localhost ~]# /etc/init.d/mysqld restart
[root@localhost ~]# mysql -u root -predhat
mysql> show master status;
+------------------+----------+--------------+------------------+
| File | Position | Binlog_Do_DB | Binlog_Ignore_DB |
+------------------+----------+--------------+------------------+
| mysql-bin.000005 | 107 | testdb | mysql |
+------------------+----------+--------------+------------------+
1 row in set (0.00 sec) //上面两个值将在从Mysql中用到。。
下面配置从mysql
[root@localhost ~]# vim /etc/my.cnf
大约49行log-bin=mysql-bin处
添加：replicate-do-db=testdb //复制的库名testdb
replicate-ignore-db=mysql //不复制的库名mysql
修改：server-id = 2 值唯一
[root@localhost ~]# /etc/init.d/mysqld restart
[root@localhost ~]# mysql -u root -predhat
mysql> create database testdb; //创建和主上面一样的同步的库名
mysql> lave stop;
mysql> change master to master_host='192.168.1.1', master_user='baidu', master_password='123.com', master_log_file='mysql-bin.000005', master_log_pos=107;
//指定主的IP，可复制用户及密码。上面两个值在每次重启主Mysql都会变。如果重启，则需要重新change master to 了
mysql> slave start; //开启复制
mysql> show slave status\G
mysql> show slave status\G
*************************** 1. row ***************************
Slave_IO_State: Waiting for master to send event
Master_Host: 192.168.1.1
Master_User: baidu
Master_Port: 3306
Connect_Retry: 60
Master_Log_File: mysql-bin.000005
Read_Master_Log_Pos: 107
Relay_Log_File: localhost-relay-bin.000003
Relay_Log_Pos: 253
Relay_Master_Log_File: mysql-bin.000005
Slave_IO_Running: Yes
Slave_SQL_Running: Yes
Replicate_Do_DB: testdb
Replicate_Ignore_DB: mysql
两个Yes表示主从配置正确……
接下来我们到主mysql的test库上面创建一个表，再到从mysql上面看看是否同步过来了。
主mysql
mysql> use testdb;
Database changed
mysql> create table student(id int(11) not null, primary key(id));
Query OK, 0 rows affected (0.06 sec)
mysql> show tables;
+----------------+
| Tables_in_user |
+----------------+
| student |
+----------------+
1 row in set (0.00 sec)
从mysql
mysql> use testdb;
Database changed
mysql> show tables;
+----------------+
| Tables_in_user |
+----------------+
| student |
+----------------+
1 row in set (0.00 sec)
可以看到student表已经成功同步过来了，至此Mysql主从实验已完成。
PS：常见错误，防火墙阻塞了端口，授权密码填写错误 mysql-bin错误 每次重启服务会变更 change mater to 需要重新敲。
Mysql编译安装的都有一个小小的BUG，不管什么错误导致mysql重启失败都会报PID丢失，一般都是你的配置文件写错了，或者查找日志文件进行排错。。
下面开始Mysql的主主关系配置：会了上面的主从之后，就很简单了。实际上就是互相指定主就可以了，你是我的主，又是我的从。
实验环境：Centos6.5 192.168.1.1 Master1：授权给1.2的用户为baidu
Centos6.5 192.168.1.2 Master2：授权给1.1的用户为sina
1.1上面操作：↓
mysql> create database testdb;
Query OK, 1 row affected (0.00 sec)
mysql> insert into mysql.user(Host,User,Password) values('localhost','baidu',password('123.com'));
Query OK, 1 row affected, 3 warnings (0.01 sec)
mysql> grant replication slave on *.* to 'baidu'@'192.168.1.2' identified by '123.com' with grant option;
Query OK, 0 rows affected (0.00 sec)
mysql> flush privileges;
Query OK, 0 rows affected (0.00 sec)
[root@localhost ~]# vim /etc/my.cnf
大约49行log-bin=mysql-bin处
添加：binlog-do-db=testdb //开启testdb二进制日志
binlog-ignore-db=mysql //忽视开启mysql二进制日志
replicate-do-db=testdb //复制的表
replicate-ignore-db=mysql //不复制的表
auto-increment-increment=10 //自增的值
auto-increment-offset=1 //原值
log-slave-updates=on //开启从日志更新
修改：server-id = 1 //该值应唯一，与另一台不冲突
上面自增与原值的解释：为了防止mysql写入数据不冲突而设立，可以这样理解，1.1写入数据按照1，11，21，31，41，
1.2按照2，12，22，32，42 这样就不会产生冲突了
修改完配置文件要重启mysql服务
[root@localhost ~]# /etc/init.d/mysqld restart
1.2上面操作：↓
mysql> create database testdb;
Query OK, 1 row affected (0.00 sec)
mysql> insert into mysql.user(Host,User,Password) values('localhost','sina',password('123.com'));
Query OK, 1 row affected, 3 warnings (0.01 sec)
mysql> grant replication slave on *.* to 'sina'@'192.168.1.1' identified by '123.com' with grant option;
Query OK, 0 rows affected (0.00 sec)
mysql> flush privileges;
Query OK, 0 rows affected (0.00 sec)
[root@localhost ~]# vim /etc/my.cnf
大约49行log-bin=mysql-bin处
添加：binlog-do-db=testdb
binlog-ignore-db=mysql
replicate-do-db=testdb
replicate-ignore-db=mysql
auto-increment-increment=10
auto-increment-offset=2
log-slave-updates=on
修改：server-id = 2
修改完配置文件要重启mysql服务
[root@localhost ~]# /etc/init.d/mysqld restart
1.1上面
mysql> show master status;
+------------------+----------+--------------+------------------+
| File | Position | Binlog_Do_DB | Binlog_Ignore_DB |
+------------------+----------+--------------+------------------+
| mysql-bin.000005 | 107 | testdb | mysql |
+------------------+----------+--------------+------------------+
1 row in set (0.00 sec)
mysql> slave stop;
Query OK, 0 rows affected, 1 warning (0.00 sec)
1.2上面
mysql> show master status;
+------------------+----------+--------------+------------------+
| File | Position | Binlog_Do_DB | Binlog_Ignore_DB |
+------------------+----------+--------------+------------------+
| mysql-bin.000004 | 107 | testdb | mysql |
+------------------+----------+--------------+------------------+
1 row in set (0.00 sec)
mysql> slave stop;
Query OK, 0 rows affected, 1 warning (0.00 sec)
下面将是主主配置的最后一步！
1.1上面
mysql> change master to master_host='192.168.1.2', master_user='sina', master_password='123.com', master_log_file='mysql-bin.000004', master_log_pos=107;
Query OK, 0 rows affected (0.06 sec)
1.2上面
change master to master_host='192.168.1.1', master_user='baidu', master_password='123.com', master_log_file='mysql-bin.000005', master_log_pos=107;
Query OK, 0 rows affected (0.07 sec)
两台都打上这条命令，开启从复制
mysql> slave start;
Query OK, 0 rows affected (0.01 sec)
两台都show slave status一下
mysql> show slave status\G
*************************** 1. row ***************************
Slave_IO_State: Waiting for master to send event
Master_Host: 192.168.1.2
Master_User: sina
Master_Port: 3306
Connect_Retry: 60
Master_Log_File: mysql-bin.000004
Read_Master_Log_Pos: 107
Relay_Log_File: localhost-relay-bin.000002
Relay_Log_Pos: 253
Relay_Master_Log_File: mysql-bin.000004
Slave_IO_Running: Yes
Slave_SQL_Running: Yes
Replicate_Do_DB: testdb
Replicate_Ignore_DB: mysql
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: 107
Relay_Log_Space: 413
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
1 row in set (0.00 sec)
两个YES表示成功；
接下来测试；
1.1上面
mysql> use testdb;
Database changed
mysql> create table user(id int (16) not null, primary key (id));
Query OK, 0 rows affected (0.03 sec)
1.2上面use testdb；
show tables； 可以看到user表同步过来了，反之在1.2上面创建一个表，到1.1上面也能够看到同步成功
觉得博主写的好的收藏一下，手敲不容易。。