延时复制 deleyed replication
Posted
tags:
篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了延时复制 deleyed replication相关的知识,希望对你有一定的参考价值。
mysql5.6开始支持延时复制,默认master_delay为0秒,
CHANGE MASTER TO MASTER_DELAY = N;
表示延时N秒
原理:延时复制的本质是sql_thread需要等待延时时间之后才能执行。
延时复制适用场景:
(1)防止主库误操作,在复制同步之前,可以停止同步
(2)用作测试
(3)用来检查数据库以前的数据
reset slave会把master_delay清零;
创建一个延时复制: slave: [email protected] [testdb]>stop slave; [email protected] [testdb]>change master to master_delay=60; [email protected] [testdb]>start slave; [email protected] [testdb]>show slave status\G ...... SQL_Delay: 60 --延时时间 SQL_Remaining_Delay: 56 --剩余时间 Slave_SQL_Running_State: Waiting until MASTER_DELAY seconds after master executed event --等待延时 ...... master: [email protected] [testdb]>delete from t1 where c1=4; slave: [email protected] [testdb]>select * from t1; +----+------+ | c1 | c2 | +----+------+ | 1 | aaa | | 2 | bbb | | 3 | ccc | | 4 | ddd | +----+------+ [email protected] [testdb]>show processlist; +----+-------------+-----------+--------+---------+------+----------------------------------------------------------------+------------------+ | Id | User | Host | db | Command | Time | State | Info | +----+-------------+-----------+--------+---------+------+----------------------------------------------------------------+------------------+ | 26 | root | localhost | testdb | Query | 0 | starting | show processlist | | 27 | system user | | NULL | Connect | 345 | Waiting for master to send event | NULL | | 28 | system user | | NULL | Connect | 10 | Waiting until MASTER_DELAY seconds after master executed event | NULL | +----+-------------+-----------+--------+---------+------+----------------------------------------------------------------+------------------+ #在没有达到60秒之前查看relay-log日志,发现已经写入relay-lo中,说明延时是阻塞SQL_thread线程 [[email protected] data]# mysqlbinlog -vv --base64-output=decode-rows relay-bin.000003 BEGIN /*!*/; # at 452 #170409 22:12:27 server id 330622 end_log_pos 5624 CRC32 0x86f7edf4 Table_map: `testdb`.`t1` mapped to number 147 # at 502 #170409 22:12:27 server id 330622 end_log_pos 5668 CRC32 0x697c52ed Delete_rows: table id 147 flags: STMT_END_F ### DELETE FROM `testdb`.`t1` ### WHERE ### @1=3 /* INT meta=0 nullable=0 is_null=0 */ ### @2=‘ccc‘ /* VARSTRING(30) meta=30 nullable=1 is_null=0 */ [email protected] [testdb]>select * from t1; +----+------+ | c1 | c2 | +----+------+ | 1 | aaa | | 2 | bbb | | 3 | ccc | +----+------+
本文出自 “10979687” 博客,请务必保留此出处http://10989687.blog.51cto.com/10979687/1921702
以上是关于延时复制 deleyed replication的主要内容,如果未能解决你的问题,请参考以下文章
MySQL性能调优与架构设计——第13章 可扩展性设计之 MySQL Replication
kafkakafka.admin.AdminOperationException: replication factor: 1 larger than available brokers: 0((代
JUC并发编程 共享模式之工具 ThreadPoolExecutor -- 任务调度线程池 定时任务 / 延时执行(ScheduledThreadPoolExecutor 延时执行 / 定时执行)(代