mysql-5.6主从同步配置示例

本文环境   
主库：CentOS6.5 x64 192.168.0.65 mysql-5.6.29    
备库：CentOS6.5 x64 192.168.0.66 mysql-5.6.29

一、常规配置方式一

1. mysql主服务器配置

# vi /etc/my.cnf

[mysqld]   
log-bin                = master-bin    
log-bin-index          = master-bin.index    
binlog_format          = mixed    
server-id              = 1

# service mysqld restart

mysql> show master status;   
+-------------------+----------+--------------+------------------+-------------------+    
| File              | Position | Binlog_Do_DB | Binlog_Ignore_DB | Executed_Gtid_Set |    
+-------------------+----------+--------------+------------------+-------------------+    
| master-bin.000001 |      353 |              |                  |                   |    
+-------------------+----------+--------------+------------------+-------------------+    
1 row in set (0.00 sec)    
mysql>

2. 主服务器配置同步复制帐号

grant replication slave on *.* to 'repl'@'%' identified by '123456';   
flush privileges;

3. mysql从服务器配置

说明: 默认只要server-id不相同即可。

# vi /etc/my.cnf

[mysqld]   
log-bin                = mysql-bin    
binlog_format          = mixed    
server-id              = 11    
relay-log              = slave-relay-bin    
relay-log-index        = slave-relay-bin.index

配置说明: 不同步mysql库可以实现主从库有不同的帐号权限，经过测试，mysql5.6.29中只在从库中配置有效。

其它参数：   
binlog-do-db                = mydb    仅同步一个数据库    
#replicate-ignore-db        = mysql   忽略掉mysql库,该参数产生很多意外的同步问题，还是不使用。    
replicate_wild_ignore_table = mysql.% 忽略掉mysql库

# service mysqld restart

4. 测试示例

CREATE DATABASE `mydb`;

CREATE TABLE `user` (   
  `id` varchar(20) NOT NULL,    
  `username` varchar(20) NOT NULL,    
  `password` char(32) NOT NULL,    
  PRIMARY KEY (`id`)    
) ;

INSERT INTO user  VALUES ('1', 'koumm', '123456');   
INSERT INTO user  VALUES ('2', 'zhangsan', '123456');    
INSERT INTO user  VALUES ('3', 'lisi', '123456');     
INSERT INTO user  VALUES ('4', 'li2si', '123456');     
INSERT INTO user  VALUES ('5', 'abc', '123456');    
INSERT INTO user  VALUES ('6', 'tom', '123456');    
INSERT INTO user  VALUES ('7', 'jk', '123456');     
INSERT INTO user  VALUES ('8', 'xb', '123456');

5. 正常的主从配置过程

(1) 主库锁表

mysql> flush tables with read lock;   
Query OK, 0 rows affected (0.00 sec)

mysql> show master status;   
+-------------------+----------+--------------+------------------+-------------------+    
| File              | Position | Binlog_Do_DB | Binlog_Ignore_DB | Executed_Gtid_Set |    
+-------------------+----------+--------------+------------------+-------------------+    
| master-bin.000001 |      353 |              |                  |                   |    
+-------------------+----------+--------------+------------------+-------------------+    
1 row in set (0.00 sec)

mysql> show master logs;    
+-------------------+-----------+    
| Log_name          | File_size |    
+-------------------+-----------+    
| master-bin.000001 |       353 |    
+-------------------+-----------+    
1 row in set (0.00 sec)

mysql>

(2) 主库备份

[root@master ~]# mysqldump -uroot -p -B mydb > mydb.sql   
说明：-B参数有建库语句。

(3) 主库解开锁表功能

mysql> unlock tables;   
Query OK, 0 rows affected (0.00 sec)    
mysql>

6. 从库导入数据库

# mysql -uroot -padmin < mydb.sql

7. 从库配置同步过程

(1) 配置同步，手动执行同步参数，该配置会写入master.info文件中。

mysql >

CHANGE MASTER TO   
MASTER_HOST='192.168.0.65',    
MASTER_PORT=3306,    
MASTER_USER='repl',    
MASTER_PASSWORD='123456',    
MASTER_LOG_FILE='master-bin.000001',    
MASTER_LOG_POS=353;

(2) 启动从同步进程

mysql> start slave;   
Query OK, 0 rows affected (0.01 sec)

mysql> show slave status\G;   
*************************** 1. row ***************************    
               Slave_IO_State: Waiting for master to send event    
                  Master_Host: 192.168.0.65    
                  Master_User: repl    
                  Master_Port: 3306    
                Connect_Retry: 60    
              Master_Log_File: master-bin.000001    
          Read_Master_Log_Pos: 353    
               Relay_Log_File: slave-relay-bin.000002    
                Relay_Log_Pos: 284    
        Relay_Master_Log_File: master-bin.000001    
             Slave_IO_Running: Yes    
            Slave_SQL_Running: Yes    
              Replicate_Do_DB:     
          Replicate_Ignore_DB:     
           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: 353    
              Relay_Log_Space: 457    
              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: c8bb22a1-024e-11e6-a1e8-000c29225fa0    
             Master_Info_File: /usr/local/mysql/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)

ERROR:    
No query specified

mysql>

#查看如下两个参数为YES，说明从库运行正常。   
             Slave_IO_Running: Yes    
            Slave_SQL_Running: Yes

(3) 测试同步情况

# 主库插入一条记录

mysql> use mydb;   
Database changed    
mysql> select * from stu;    
+----+----------+----------+    
| id | username | password |    
+----+----------+----------+    
|  1 | koumm    | 123456   |    
|  2 | zhangsan | 123456   |    
|  3 | lisi     | 123456   |    
|  4 | wangwu   | 123456   |    
|  5 | zhaoliu  | 123456   |    
|  6 | zhouqi   | 123456   |    
+----+----------+----------+    
6 rows in set (0.00 sec)

mysql>    
mysql> INSERT INTO stu  VALUES ('7', 'tom', '123456');       
Query OK, 1 row affected (0.05 sec)

mysql> commit;   
Query OK, 0 rows affected (0.00 sec)

mysql>

# 从库查询同步情况

mysql> use mydb;   
Database changed    
mysql> select * from stu;    
+----+----------+----------+    
| id | username | password |    
+----+----------+----------+    
|  1 | koumm    | 123456   |    
|  2 | zhangsan | 123456   |    
|  3 | lisi     | 123456   |    
|  4 | wangwu   | 123456   |    
|  5 | zhaoliu  | 123456   |    
|  6 | zhouqi   | 123456   |    
|  7 | tom      | 123456   |    
+----+----------+----------+    
7 rows in set (0.00 sec)

mysql>

小结：

本文只是配置mysql主从的一个过程。还是有很多需要考虑与改进的地址。   
(1) 主从库同步用户权限的问题考虑，是直接从主库同步权限呢，还是单独考虑从库的权限。    
(2) mysql5.5有半自动主从同步复制    
(3) mysql5.6有基于GTID主从复制

二、快速配置从库方式二

1，主库备份(全库备份)

mysqldump -uroot -p -A -B --events --master-data=1 > mydb.sql

全库备份记录了数据库备份时的master-bin.000003', MASTER_LOG_POS=583位置，在配置从库时无需要在主库锁表，记录POS位置，解锁等配置。

[root@localhost ~]# egrep -v "#|\*|--|^$" mydb.sql |head

CHANGE MASTER TO MASTER_LOG_FILE='master-bin.000003', MASTER_LOG_POS=583;
USE `mydb`;
DROP TABLE IF EXISTS `user`;
CREATE TABLE `user` (
  `id` varchar(20) NOT NULL,
  `username` varchar(20) NOT NULL,
  `password` char(32) NOT NULL,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
LOCK TABLES `user` WRITE;

2，从库创建同步的数据库, 并导入数据

mysql -uroot -padmin
create database mydb;
mysql -uroot -padmin mydb < mydb.sql

3，配置主从同步

mysql> 
CHANGE MASTER TO
MASTER_HOST='192.168.0.65',
MASTER_PORT=3306,
MASTER_USER='repl',
MASTER_PASSWORD='123456';

mysql> start slave;
Query OK, 0 rows affected (0.02 sec)

mysql> show slave status \G;
*************************** 1. row ***************************
               Slave_IO_State: Waiting for master to send event
                  Master_Host: 192.168.0.65
                  Master_User: repl
                  Master_Port: 3306
                Connect_Retry: 60
              Master_Log_File: master-bin.000003
          Read_Master_Log_Pos: 583
               Relay_Log_File: testdb-relay-bin.000004
                Relay_Log_Pos: 747
        Relay_Master_Log_File: master-bin.000003
             Slave_IO_Running: Yes
            Slave_SQL_Running: Yes
              Replicate_Do_DB: 
          Replicate_Ignore_DB: 
           Replicate_Do_Table: 
       Replicate_Ignore_Table: 
      Replicate_Wild_Do_Table: 
  Replicate_Wild_Ignore_Table: mysql.%
                   Last_Errno: 0
                   Last_Error: 
                 Skip_Counter: 0
          Exec_Master_Log_Pos: 583
              Relay_Log_Space: 2536
              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: c8bb22a1-024e-11e6-a1e8-000c29225fa0
             Master_Info_File: /usr/local/mysql-5.6.29-linux-glibc2.5-x86_64/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)

三、主从同步简单管理

1. 停止MYSQL同步

STOP SLAVE IO_THREAD;     #停止IO进程
STOP SLAVE SQL_THREAD;    #停止SQL进程
STOP SLAVE;               #停止IO和SQL进程

2. 启动MYSQL同步

START SLAVE IO_THREAD;    #启动IO进程
START SLAVE SQL_THREAD;   #启动SQL进程
START SLAVE;              #启动IO和SQL进程

3. 重置MYSQL同步

RESET SLAVE;
清除主从同步参数，它会删除master.info和relay-log.info文件，以及所有的中继日志，并启动一个新的中继日志。
适用重新再次配置一次从库的情况。

4. 查看MYSQL同步状态

SHOW SLAVE STATUS;

5. 临时跳过MYSQL同步错误

确保数据一致的情况下临时的跳过这个错误操作如下，有可能要多次。

STOP SLAVE;
SET GLOBAL SQL_SLAVE_SKIP_COUNTER=1;
START SLAVE;

# vi /etc/my.cnf

[mysqld]
#slave-skip-errors=1062,1053,1146 #跳过指定error no类型的错误
#slave-skip-errors=all            #跳过所有错误

这里出现一次1146报错，是因为配置binlog-do-db过滤，出现该配置的BUG问题，所有要慎用binlog-do-db。

本文出自 “koumm的linux技术博客” 博客，请务必保留此出处http://koumm.blog.51cto.com/703525/1764093