基于GTID的replication

4020阅读 0评论2018-06-06 zhangshengdong
分类:Mysql/postgreSQL

步骤:
1)两个MySQL实例:
      IP地址:10.0.0.201  端口:3306  (MySQL的Master)
      IP地址:10.0.0.201  端口:3307 (MySQL的Slave)

2)两台数据库实例,需要开启log-bin、GTID、和级联复制的选线log_slave_updates
      my.cnf配置参数如下:
  master(3306)
    server-id = 1203306
    log-bin = /data/mysqldata/3306/binlog/mysql-bin

    #*********** GTID settting*******************
    gtid_mode=ON
    enforce-gtid-consistency=true
    binlog_format= row
    skip-slave-start=1
    log_slave_updates = 1

  slave(3307)
    server-id = 1203307
    log-bin = /data/mysqldata/3307/binlog/mysql-bin

    #*********** GTID settting*******************
    gtid_mode=ON
    enforce-gtid-consistency=true
    binlog_format= row
    skip-slave-start=1
    log_slave_updates = 1


3)为Replication创建一个用户
  在master上,配置一个新用户,可以让slave服务器能读取master服务器
mysql> CREATE USER IDENTIFIED BY ;
mysql> GRANT REPLICATION SLAVE ON *.* TO ;


4)Master保证数据一致性,设置为只读
mysql> SET @@global.read_only = ON;
() [(none)]> show master status\G;
*************************** 1. row ***************************
             File: mysql-bin.000006
         Position: 737
     Binlog_Do_DB:
 Binlog_Ignore_DB:
Executed_Gtid_Set: 4160e9b3-58d9-11e8-b174-005056af6f24:1-21
1 row in set (0.00 sec)

********备份******************
/usr/local/mysql/bin/mysqldump -uroot -S /data/mysqldata/3306/mysql.sock -A --flush-logs --single-transaction -e | gzip > mysql_3306_full_backup_`date +%F`.sql.gz
-A参数                        参数解释:--all-databases, -A。备份所有数据库的所有表。
--flush-logs参数              参数解释:在数据dump之前,刷新数据库的日志。
--single-transaction参数      参数解释:服务器dump之前,发起一个BEGIN SQL的语句, START TRANSACTION 放进一个事务里面。
-e参数                        参数解释:--extended-insert, -e 产生一个小的dump文件,加速插入的速度

master导出备份
/usr/local/mysql/bin/mysqldump -uroot -S /data/mysqldata/3306/mysql.sock --all-databases --triggers --routines --events --flush-logs --single-transaction -e --default-character-set=utf8 | gzip > mysql_3306_full_backup_`date +%F`.sql.gz

slave导入备份
/usr/local/mysql/bin/mysql -uroot -S /data/mysqldata/3307/mysql.sock --default-character-set=utf8 < mysql_3306_full_backup_2018-06-01.sql


5)SLAVE追日志
------------------抽象语句-----------------------------
mysql> CHANGE MASTER TO
     >     MASTER_HOST = host,
     >     MASTER_PORT = port,
     >     MASTER_USER = user,
     >     MASTER_PASSWORD = password,
     >     MASTER_AUTO_POSITION = 1;
-----------------实战语句------------------------------

CHANGE MASTER TO MASTER_HOST = '10.0.0.201',MASTER_PORT = 3306,MASTER_USER = 'repl',MASTER_PASSWORD = = 1;


6)启动slave
mysql> START SLAVE;

7)Master服务,关闭只读
mysql> SET @@global.read_only = OFF;


完全删除replication的拓扑结构

() [(none)]> stop slave;
() [(none)]> RESET SLAVE  ALL;

上一篇:解决MySQL Error (2013):waiting for initial communication packet
下一篇:在Oracle11gR2下基于RMAN恢复schema的数据