Linux+mysql5.6配置主从

1.1、配置主数据库：

vim /etc/my.cnf

[mysqld]
server-id=1
log-bin=master-bin
log-bin-index=master-bin.index
binlog_do_db=test_slave  ##要同步的数据库的名称
#不同步的数据库，可设置多个
binlog_ignore_db=mysql
binlog-ignore-db=information_schema

1.2、重启mysql：service mysqld restart

1.3、查看主服务器状态：mysql> show master status;

注意：记录好File和Position，后面要用

 

2.1、配置从数据库：

vim /etc/my.cnf

[mysqld]
server-id=2
log-bin=master-bin
relay-log=slave-relay-bin
relay-log-index=slave-relay-bin.index
replicate-do-db=test_slave  ##要同步的数据库的名称，与主库一只
replicate-ignore-db=mysql
log-slave-updates
slave-skip-errors=all

2.2、重启数据库；

2.3、登录数据库：（复制上面1.3的信息）

mysql> stop slave;  #设置连接主库信息
mysql> change master to master_host='主库IP',,master_port=3306,master_user='主库用户',master_password='主库密码',master_log_file='主库log_file',master_log_pos=主库POS;
mysql> start slave;  #开启Slave

2.4、查看信息：mysql> show slave status\G;

主要是看：这两个都要是YES。

Slave_IO_Running: Yes
Slave_SQL_Running: Yes

mysql> show slave status\G;
*************************** 1. row ***************************
               Slave_IO_State: Waiting for master to send event
                  Master_Host: 主库IP
                  Master_User: 主库用户
                  Master_Port: 3306
                Connect_Retry: 60
              Master_Log_File: master-bin.000006
          Read_Master_Log_Pos: 1491
               Relay_Log_File: slave-relay-bin.000024
                Relay_Log_Pos: 284
        Relay_Master_Log_File: master-bin.000006
             Slave_IO_Running: Yes
            Slave_SQL_Running: Yes
              Replicate_Do_DB: test_slave
          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: 1491
              Relay_Log_Space: 621
              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
                  Master_UUID: ae0dd5de-3916-11e7-8c75-00163e13f64e
             Master_Info_File: /www/server/data/master.info
                    SQL_Delay: 0
          SQL_Remaining_Delay: NULL
      Slave_SQL_Running_State: Slave has read all relay log; waiting for the slave I/O thread to update it
           Master_Retry_Count: 86400
                  Master_Bind: 
      Last_IO_Error_Timestamp: 
     Last_SQL_Error_Timestamp: 
               Master_SSL_Crl: 
           Master_SSL_Crlpath: 
           Retrieved_Gtid_Set: 
            Executed_Gtid_Set: 
                Auto_Position: 0
1 row in set (0.00 sec)

2.5、碰到问题：mysql> show variables like 'log_error%';   可查看mysql-error.log，以排查问题

Fatal error: The slave I/O thread stops because master and slave have equal MySQL server UUIDs; these UUIDs must be different for replication to work.

原因分析：

mysql 5.6的复制引入了uuid的概念，各个复制结构中的server_uuid得保证不一样，但是查看到直接copy  data文件夹后server_uuid是相同的，show variables like '%server_uuid%';

解决方法：

找到/www/server/data文件夹下的auto.cnf文件，修改里面的uuid值，保证各个db的uuid不一样，重启db即可