| Bug #21952 | SLOW LOG KILLING MYSQLD PROCESS | ||
|---|---|---|---|
| Submitted: | 31 Aug 2006 18:46 | Modified: | 29 Mar 2011 18:43 |
| Reporter: | Roberto Spadim (Basic Quality Contributor) | Email Updates: | |
| Status: | Closed | Impact on me: | |
| Category: | MySQL Server | Severity: | S3 (Non-critical) |
| Version: | 5.1.11-beta | OS: | Linux (LINUX 2.6.10) |
| Assigned to: | CPU Architecture: | Any | |
[31 Aug 2006 18:46]
Roberto Spadim
[31 Aug 2006 20:13]
Sveta Smirnova
Thank you for the report. Could you please upload corrupted general_log.CSM, general_log.CSV, general_log.frm, slow_log.CSM, slow_log.CSV, slow_log.frm you can find in mysql directory inside your datadir?
[31 Aug 2006 20:25]
Roberto Spadim
o can't cause i clean to start mysql again, next time it get error i will send (no disk full error ocurring)
[5 Sep 2006 8:12]
Sveta Smirnova
OK, Roberto, will wait feedback from you. Also is interesting how you got corrupt these tables. You may set --log-output=TABLE,FILE to have copy of queries in old format logs to indicate what was cause of the problem.
[5 Sep 2006 11:56]
Roberto Spadim
hello i will put err.log and mysql folder without user table ;)
[5 Sep 2006 11:58]
Roberto Spadim
sorry i will put log-output=TABLE,FILE now
[5 Sep 2006 16:14]
Sveta Smirnova
I checked your archive and it does not contain broken tables, isn't it? Please, upload broken tables to we can examine what was broken.
[5 Sep 2006 19:26]
Roberto Spadim
i saw that but if i don't delete the files and touch again mysql server don't start i'm using mysqld_safe script to load mysql server, so the server could never stop just if script die and after server, but it don't occur in my machine cause just i use it any idea of what is going on?
[5 Sep 2006 19:33]
Roberto Spadim
i changed to noncritical cause i don't undestand why mysql is been killed
[5 Sep 2006 19:37]
Sveta Smirnova
Roberto, do not delete broken tables. Simple move them out from mysql directory and send us.
[14 Sep 2006 2:33]
Roberto Spadim
file with error data dir uploaded
[14 Sep 2006 19:48]
Sveta Smirnova
I examined again the error log and found what mysqld stops not immediately after start, but after any hours in some cases. Are you sure corrupted log tables is cause of the crash? Anyway current mysql 5.1.12-BK repaired log tables after start as you can see from the test error log: 060914 21:24:08 InnoDB: Started; log sequence number 0 53481 libexec/mysqld: Table 'general_log' is marked as crashed and should be repaired libexec/mysqld: Table 'slow_log' is marked as crashed and should be repaired 060914 21:24:08 [Note] libexec/mysqld: ready for connections. Version: '5.1.12-beta-debug-log' socket: '/tmp/mysql.sock' port: 0 Source distribution 060914 21:24:08 [Note] SCHEDULER: Manager thread booting 060914 21:24:08 [Note] SCHEDULER: Loaded 0 events 060914 21:24:08 [Note] SCHEDULER: Suspending operations 060914 21:43:34 [Note] Got signal 15 to shutdown mysqld 060914 21:43:34 [Note] libexec/mysqld: Normal shutdown 060914 21:43:34 [Note] SCHEDULER: Killing manager thread 1 060914 21:43:34 [Note] SCHEDULER: Waiting the manager thread to reply 060914 21:43:34 [Note] SCHEDULER: Resuming operations 060914 21:43:34 [Note] SCHEDULER: Manager thread started with id 1 060914 21:43:34 [Note] SCHEDULER: Shutting down 060914 21:43:34 [Note] SCHEDULER: Waiting for worker threads to finish 060914 21:43:34 [Note] SCHEDULER: Emptying the queue 060914 21:43:34 [Note] SCHEDULER: Stopped 060914 21:43:35 InnoDB: Starting shutdown... 060914 21:43:38 InnoDB: Shutdown completed; log sequence number 0 61483 060914 21:43:38 [Note] libexec/mysqld: Shutdown complete 060914 21:43:45 InnoDB: Started; log sequence number 0 61483 060914 21:43:45 [Note] libexec/mysqld: ready for connections. Version: '5.1.12-beta-debug-log' socket: '/tmp/mysql.sock' port: 0 Source distribution 060914 21:43:45 [Note] SCHEDULER: Manager thread booting 060914 21:43:45 [Note] SCHEDULER: Loaded 0 events 060914 21:43:45 [Note] SCHEDULER: Suspending operations 060914 21:44:23 [Note] Got signal 15 to shutdown mysqld 060914 21:44:23 [Note] libexec/mysqld: Normal shutdown 060914 21:44:23 [Note] SCHEDULER: Killing manager thread 1 060914 21:44:23 [Note] SCHEDULER: Waiting the manager thread to reply 060914 21:44:23 [Note] SCHEDULER: Resuming operations 060914 21:44:23 [Note] SCHEDULER: Manager thread started with id 1 060914 21:44:23 [Note] SCHEDULER: Shutting down 060914 21:44:23 [Note] SCHEDULER: Waiting for worker threads to finish 060914 21:44:23 [Note] SCHEDULER: Emptying the queue 060914 21:44:23 [Note] SCHEDULER: Stopped 060914 21:44:23 InnoDB: Starting shutdown... 060914 21:44:25 InnoDB: Shutdown completed; log sequence number 0 61483 060914 21:44:25 [Note] libexec/mysqld: Shutdown complete
[14 Sep 2006 19:57]
Roberto Spadim
i'm not sure cause if mysqld crash mysqld_safe should make it run again, but when it crash and i run ps ax | grep mysql no process is showed can i run mysql in production use with some error log to crash and log what make it crash? i can run mysql debug? the server is big not very, but have some tables (5 or 6) that have +- 300000 rows and they are used can i run mysql with debug option? how can i do it?
[21 Sep 2006 6:36]
USAF LEE
Description: 060920 15:31:00 mysqld restarted InnoDB: Error: log file ./ib_logfile0 is of different size 0 5242880 bytes InnoDB: than specified in the .cnf file 0 268435456 bytes! 060920 15:31:04 [Note] Starting MySQL Cluster Binlog Thread /usr/sbin/mysqld: Table 'general_log' is marked as crashed and should be repaired /usr/sbin/mysqld: Woops, blew up opening a mapped file /usr/sbin/mysqld: Out of memory; check if mysqld or some other process uses all available memory; if not, you may have to use 'ulimit' to allow mysqld to use more memory or you can add more swap space 060920 15:31:04 [ERROR] Failed to initialize log tables. Falling back to the old-fashioned logs 060920 15:31:05 [Note] /usr/sbin/mysqld: ready for connections. Version: '5.1.11-beta-log' socket: '/var/lib/mysql/mysql.sock' port: 3306 MySQL Community Server (GPL) 060920 15:31:05 [Note] SCHEDULER: Manager thread booting 060920 15:31:05 [Note] SCHEDULER: Loaded 0 events 060920 15:31:05 [Note] SCHEDULER: Suspending operations 060920 15:31:06 [ERROR] /usr/sbin/mysqld: Table './cluster/binlog_index' is marked as crashed and should be repaired 060920 15:31:06 [Warning] Checking table: './cluster/binlog_index' 060920 15:31:06 [ERROR] 1 client is using or hasn't closed the table properly 060920 15:31:08 [ERROR] NDB Binlog: logging of table ./promisedata2006/company_setting with BLOB attribute and no PK is not supported 060920 15:31:08 [ERROR] NDB Binlog: FAILED CREATE (DISCOVER) TABLE Event: REPL$promisedata2006/company_setting What is the error,how to resolve?
[29 Mar 2011 18:43]
Roberto Spadim
working without problems now, using mysql 5.5
