背景

搭建了mysql的主从环境，然后在从上面kill -9杀死了mysql进程。之后在slave上面重启mysql服务，端口起来后无法登陆数据库。过了好一会才能登陆上去。

查看mysql.err错误日志发现如下信息

2020-04-22T10:27:16.091165Z 0 [Note] InnoDB: Doing recovery: scanned up to log sequence number 78607991370

之后打印

[Note] InnoDB: Database was not shutdown normally!

[Note] InnoDB: Starting crash recovery.

提示说mysql没有正常关闭。那么怎么才算正常关闭呢，正常关闭和非正常关闭有啥区别呢？

带着这些疑问，去查一下资料。

资料

（1）mysql正常关闭方法： bin/mysqladmin -u ${user} -p ${password} -S ${mysql.sock} shutdown

举例：

/kingdb/usr/local/nestdb_master/bin/mysqladmin -uroot -p123456 -S /kingdb/usr/local/nestdb_master/tmp/mysql.sock shutdown

查看mysql.err日志，发现记录了shutdown之后都干了什么事：

[Note] Giving 0 client threads a chance to die gracefully ---gracefully，应该就表示是温和不失礼貌的关闭方式

Shutting down slave threads

Forcefully disconnecting 0 remaining clients

Binlog end

Shutting down plugin 'ngram'

InnoDB: FTS optimize thread exiting.

InnoDB: Starting shutdown...

InnoDB: Dumping buffer pool(s) to /kingdb/usr/local/nestdb_master/var/ib_buffer_pool

InnoDB: Buffer pool(s) dump completed at 200422 20:59:40

InnoDB: shut down release done

InnoDB: shut down all bg threads

[Warning] InnoDB: Some (1) threads are still active

[Note] InnoDB: Shutdown completed; log sequence number 4522278

[Note] InnoDB: Removed temporary tablespace data file: "ibtmp1"

[Note] Shutting down plugin 'binlog'

[Note] /kingdb/usr/local/nestdb_master/bin/mysqld: Shutdown complete

上面的日志，大致看出流程为：关闭插件-->dumping buffer pool--->杀掉进程--->删掉了临时表空间文件ibtmp1---->关闭binlog插件等