详解MySQL主从复制实战 - 基于GTID的复制

基于GTID的复制

简介

基于GTID的复制是MySQL 5.6后新增的复制方式.

GTID (global transaction identifier) 即全局事务ID, 保证了在每个在主库上提交的事务在集群中有一个唯一的ID.

在原来基于日志的复制中, 从库需要告知主库要从哪个偏移量进行增量同步, 如果指定错误会造成数据的遗漏, 从而造成数据的不一致.

而基于GTID的复制中, 从库会告知主库已经执行的事务的GTID的值, 然后主库会将所有未执行的事务的GTID的列表返回给从库. 并且可以保证同一个事务只在指定的从库执行一次.

实战

1、在主库上建立复制账户并授予权限

基于GTID的复制会自动地将没有在从库执行的事务重放, 所以不要在其他从库上建立相同的账号. 如果建立了相同的账户, 有可能造成复制链路的错误.

mysql> create user 'repl'@'172.%' identified by '123456';

注意在生产上的密码必须依照相关规范以达到一定的密码强度, 并且规定在从库上的特定网段上才能访问主库.

mysql> grant replication slave on *.* to 'repl'@'172.%';

查看用户

mysql> select user, host from mysql.user;
+-----------+-----------+
| user  | host  |
+-----------+-----------+
| prontera | %   |
| root  | %   |
| mysql.sys | localhost |
| root  | localhost |
+-----------+-----------+
4 rows in set (0.00 sec)

查看授权

mysql> show grants for repl@'172.%';
+--------------------------------------------------+
| Grants for repl@172.%       |
+--------------------------------------------------+
| GRANT REPLICATION SLAVE ON *.* TO 'repl'@'172.%' |
+--------------------------------------------------+
1 row in set (0.00 sec)

2、配置主库服务器

[mysqld]
log_bin = /var/log/mysql/mysql-bin
log_bin_index = /var/log/mysql/mysql-bin.index
binlog_format = row
server_id = 101
gtid_mode = ON
enforce_gtid_consistency = ON
#log_slave_updates = ON

NOTE: 把日志与数据分开是个好习惯, 最好能放到不同的数据分区

enforce_gtid_consistency 强制GTID一致性, 启用后以下命令无法再使用

create table ... select ...

mysql> create table dept select * from departments;
ERROR 1786 (HY000): Statement violates GTID consistency: CREATE TABLE ... SELECT.

因为实际上是两个独立事件, 所以只能将其拆分先建立表, 然后再把数据插入到表中

create temporary table

事务内部不能创建临时表

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

mysql> create temporary table dept(id int);
ERROR 1787 (HY000): Statement violates GTID consistency: CREATE TEMPORARY TABLE and DROP TEMPORARY TABLE can only be executed outside transactional context. These statements are also not allowed in a function or trigger because functions and triggers are also considered to be multi-statement transactions.

同一事务中更新事务表与非事务表(MyISAM)

mysql> CREATE TABLE `dept_innodb` (id INT(11) UNSIGNED NOT NULL PRIMARY KEY AUTO_INCREMENT);
Query OK, 0 rows affected (0.04 sec)

mysql> CREATE TABLE `dept_myisam` (id INT(11) UNSIGNED NOT NULL PRIMARY KEY AUTO_INCREMENT) ENGINE = `MyISAM`;
Query OK, 0 rows affected (0.03 sec)

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

mysql> insert into dept_innodb(id) value(1);
Query OK, 1 row affected (0.00 sec)

mysql> insert into dept_myisam(id) value(1);
ERROR 1785 (HY000): Statement violates GTID consistency: Updates to non-transactional tables can only be done in either autocommitted statements or single-statement transactions, and never in the same statement as updates to transactional tables.

所以建议选择Innodb作为默认的数据库引擎.

log_slave_updates 该选项在MySQL 5.6版本时基于GTID的复制是必须的, 但是其增大了从服务器的IO负载, 而在MySQL 5.7中该选项已经不是必须项

3、配置从库服务器

master_info_repository 与relay_log_info_repository

在MySQL 5.6.2之前, slave记录的master信息以及slave应用binlog的信息存放在文件中, 即master.info与relay-log.info. 在5.6.2版本之后, 允许记录到table中. 对应的表分别为mysql.slave_master_info与mysql.slave_relay_log_info, 且这两个表均为innodb引擎表.

[mysqld]
log_bin = /var/log/mysql/mysql-bin
log_bin_index = /var/log/mysql/mysql-bin.index
server_id = 102
# slaves
relay_log  = /var/log/mysql/relay-bin
relay_log_index = /var/log/mysql/relay-bin.index
relay_log_info_file = /var/log/mysql/relay-bin.info
enforce_gtid_consistency = ON
log_slave_updates = ON
read_only = ON
master_info_repository = TABLE
relay_log_info_repository = TABLE

4、从库数据初始化 - [optional]

先在主库上备份数据

复制代码代码如下:
mysqldump --single-transaction --master-data=2 --triggers --routines --all-databases --events -u root -p > backup.sql

—master-data=2 该选项将当前服务器的binlog的位置和文件名追加到输出文件中(show master status). 如果为1, 将偏移量拼接到CHANGE MASTER 命令. 如果为2, 输出的偏移量信息将会被注释。

--all-databases 因为基于GTID的复制会记录全部的事务, 所以要构建一个完整的dump这个选项是推荐的

常见错误

当从库导入SQL的时候出现

复制代码代码如下:
ERROR 1840 (HY000) at line 24: @@GLOBAL.GTID_PURGED can only be set when @@GLOBAL.GTID_EXECUTED is empty.

此时进入从库的MySQL Command Line, 使用reset master即可

5、启动基于GTID的复制

现有master@172.20.0.2和slave@172.20.0.3, 并且已经通过mysqldump将数据同步至从库slave中. 现在在从服务器slave上配置复制链路

mysql> change master to master_host='master', master_user='repl', master_password='123456', master_auto_position=1;
Query OK, 0 rows affected, 2 warnings (0.06 sec)

启动复制

mysql> start slave;

启动成功后查看slave的状态

mysql> show slave status\G
*************************** 1. row ***************************
    Slave_IO_State: Queueing master event to the relay log
     Master_Host: master
     Master_User: repl
     Master_Port: 3306
    Connect_Retry: 60
    Master_Log_File: mysql-bin.000002
   Read_Master_Log_Pos: 12793692
    Relay_Log_File: relay-bin.000002
    Relay_Log_Pos: 1027
  Relay_Master_Log_File: mysql-bin.000002
    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: 814
    Relay_Log_Space: 12794106
    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: 5096
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: 101
     Master_UUID: a9fd4765-ec70-11e6-b543-0242ac140002
    Master_Info_File: mysql.slave_master_info
     SQL_Delay: 0
   SQL_Remaining_Delay: NULL
  Slave_SQL_Running_State: Reading event from the relay log
   Master_Retry_Count: 86400
     Master_Bind:
  Last_IO_Error_Timestamp:
  Last_SQL_Error_Timestamp:
    Master_SSL_Crl:
   Master_SSL_Crlpath:
   Retrieved_Gtid_Set: a9fd4765-ec70-11e6-b543-0242ac140002:1-39
   Executed_Gtid_Set: a9fd4765-ec70-11e6-b543-0242ac140002:1-4
    Auto_Position: 1
   Replicate_Rewrite_DB:
     Channel_Name:
   Master_TLS_Version:
1 row in set (0.00 sec)

当Slave_IO_Running, Slave_SQL_Running为YES,

且Slave_SQL_Running_State 为Slave has read all relay log; waiting for more updates时表示成功构建复制链路

6、总结

优点