Note: This article is available in Chinese only. 本文暂无英文版本。
View original
其实 东西之前出现过…不过好像重启一下服务器就可以了？
这次比较麻烦。
一开始我是直接 google 了这条错误信息，结果答案五花八门，或者说…可能的原因非常多。
排查了几个，还是没有搞定。
突然想到，为何不直接看 log…我好傻啊。
12016-10-10 10:36:54 3755 [Note] IPv6 is not available.
22016-10-10 10:36:54 3755 [Note] - '0.0.0.0' resolves to '0.0.0.0';
32016-10-10 10:36:54 3755 [Note] Server socket created on IP: '0.0.0.0'.
42016-10-10 10:36:54 3755 [ERROR] /alidata/server/mysql/bin/mysqld: Table './mysql/user' is marked as crashed and last (automatic?) repair failed
52016-10-10 10:36:54 3755 [ERROR] Fatal error: Can't open and lock privilege tables: Table './mysql/user' is marked as crashed and last (automatic?) repair failed
6161010 10:36:54 mysqld_safe mysqld from pid file /alidata/server/mysql/data/iZ287lyogf4Z.pid ended
7161010 10:42:20 mysqld_safe Starting mysqld daemon with databases from /alidata/server/mysql/data
82016-10-10 10:42:20 0 [Warning] TIMESTAMP with implicit DEFAULT value is deprecated. Please use --explicit_defaults_for_timestamp server option (see documentation for more details).
92016-10-10 10:42:20 4118 [Note] Plugin 'FEDERATED' is disabled.
102016-10-10 10:42:20 4118 [Note] InnoDB: Using mutexes to ref count buffer pool pages
112016-10-10 10:42:20 4118 [Note] InnoDB: The InnoDB memory heap is disabled
122016-10-10 10:42:20 4118 [Note] InnoDB: Mutexes and rw_locks use InnoDB's own implementation
132016-10-10 10:42:20 4118 [Note] InnoDB: Memory barrier is not used
142016-10-10 10:42:20 4118 [Note] InnoDB: Compressed tables use zlib 1.2.3
152016-10-10 10:42:20 4118 [Note] InnoDB: Using Linux native AIO
162016-10-10 10:42:20 4118 [Note] InnoDB: Not using CPU crc32 instructions
172016-10-10 10:42:20 4118 [Note] InnoDB: Initializing buffer pool, size = 128.0M
182016-10-10 10:42:20 4118 [Note] InnoDB: Completed initialization of buffer pool
192016-10-10 10:42:20 4118 [Note] InnoDB: Highest supported file format is Barracuda.
202016-10-10 10:42:20 4118 [Note] InnoDB: The log sequence numbers 414871617 and 414871617 in ibdata files do not match the log sequence number 554212530 in the ib_logfiles!
212016-10-10 10:42:20 4118 [Note] InnoDB: Database was not shutdown normally!
222016-10-10 10:42:20 4118 [Note] InnoDB: Starting crash recovery.
232016-10-10 10:42:20 4118 [Note] InnoDB: Reading tablespace information from the .ibd files...
242016-10-10 10:42:20 4118 [Note] InnoDB: Restoring possible half-written data pages
252016-10-10 10:42:20 4118 [Note] InnoDB: from the doublewrite buffer...
26InnoDB: Last MySQL binlog file position 0 92722334, file name mysql-bin.000005
272016-10-10 10:42:20 4118 [Note] InnoDB: 128 rollback segment(s) are active.
282016-10-10 10:42:20 4118 [Note] InnoDB: Waiting for purge to start
292016-10-10 10:42:20 4118 [Note] InnoDB: 5.6.21 started; log sequence number 554212530
302016-10-10 10:42:20 4118 [Note] Recovering after a crash using mysql-bin
312016-10-10 10:42:20 4118 [Note] Starting crash recovery...
322016-10-10 10:42:20 4118 [Note] Crash recovery finished.
332016-10-10 10:42:20 4118 [Note] Server hostname (bind-address): '*'; port: 3306
342016-10-10 10:42:20 4118 [Note] IPv6 is not available.
352016-10-10 10:42:20 4118 [Note] - '0.0.0.0' resolves to '0.0.0.0';
362016-10-10 10:42:20 4118 [Note] Server socket created on IP: '0.0.0.0'.
372016-10-10 10:42:20 4118 [ERROR] /alidata/server/mysql/bin/mysqld: Table './mysql/user' is marked as crashed and last (automatic?) repair failed
382016-10-10 10:42:20 4118 [ERROR] Fatal error: Can't open and lock privilege tables: Table './mysql/user' is marked as crashed and last (automatic?) repair failed
39161010 10:42:20 mysqld_safe mysqld from pid file /alidata/server/mysql/data/iZ287lyogf4Z.pid ended发现是说一个表挂掉了。
然后在SO上看到这个stackoverflow
然后惊喜地发现完全不记得数据库密码是啥了…毕竟搭博客的时候我对这些完全没概念…
然后在my.cnf配置文件 加一行 skip-grant-tables 具体位置是在/etc/my.cnf
这样就可以跳过密码直接登录数据库。
查了下果然有问题。
然后用 repair table table_name; 命令修复。
注意 table_name 外面没有引号啊，中括号之类乱七八糟的 2333.
然后再check，发现少了条warning，就是日志里error的那条。
尝试了下访问博客，发现已经可以了，感动到哭。
感觉这次学到的最主要的事情是，出问题直接去找 log，不要从表面的症状来找问题。嗯，感动。