| Bug #46338 | Unknow to crash MySQL | ||
|---|---|---|---|
| Submitted: | 22 Jul 2009 10:49 | Modified: | 14 Sep 2009 19:56 |
| Reporter: | Steven Yeung | Email Updates: | |
| Status: | Not a Bug | Impact on me: | |
| Category: | MySQL Server | Severity: | S2 (Serious) |
| Version: | 5.1.34-community | OS: | Linux (RHEL 5) |
| Assigned to: | CPU Architecture: | Any | |
[22 Jul 2009 10:49]
Steven Yeung
[22 Jul 2009 21:16]
Sveta Smirnova
Thank you for the report. We have 2 similar, but not exactly same bugs: bug #34895 and bug #41726. Did you do upgrade? Did you run mysql_upgrade? If you have general and/or binary query log turned on please send them to us, so we can guess which queries were running same time. Please also try current version 5.1.36 to check if problem fixed occasionally.
[23 Jul 2009 7:21]
Steven Yeung
Yes. I have exceuted mysql_upgrade command before. And I have added below's setting in my.cnf two weeks ago. max_sp_recursion_depth = 255 thread_stack = 393216
[27 Jul 2009 20:47]
Sveta Smirnova
Thank you for the feedback. I can not repeat described behavior using load only with statements from the binary log provided (update, insert, delete). Please try to gather more information about load from your application: probably some statements affect this. You can temporarily turn on general query log (as it is online since version 5.1) or use logging from your application or MySQL Proxy. Also, please, provide your configuration file and output of SHOW GRANTS for user you run SHOW PROCEDURE STATUS as.
[28 Jul 2009 1:29]
Steven Yeung
Actually, I try to comment the new configration. Then, the crash doesn't occur any more. But I need stored procedure to support recursion here. So, I hope belows information can help yours to find out the problem. My Server information: CPU: Intel(R) Xeon(R) CPU E5335 @ 2.00GHz RAM: 2G RAM OS: RHEL 5 My configration as below: [mysqld] #default#datadir=/var/lib/mysql #default#socket=/var/lib/mysql/mysql.sock # Default to using old password format for compatibility with mysql 3.x # clients (those using the mysqlclient10 compatibility package). #default#old_passwords=1 server-id=1 max_connections=2048 default-storage-engine=MYISAM ##log log-bin=lx18-bin #log-slow-queries #--log-bin=lx18-bin #--log-bin=/var/lib/mysql/lx18-bin (not worked) character-set-server=big5 key_buffer = 384M read_buffer_size = 2M sort_buffer_size = 2M myisam_sort_buffer_size = 64M table_cache = 512 thread_cache_size = 8 query_cache_size = 32M # Try number of CPU's*2 for thread_concurrency thread_concurrency = 8 max_sp_recursion_depth = 255 #thread stack in 384KB thread_stack = 393216 [mysql.server] #default#user=mysql #default#basedir=/var/lib [client] ##default-character-set=big5 [mysqld_safe] #default#err-log=/var/mysql/log/mysqld.log #default#pid-file=/var/run/mysqld/mysqld.pid ##server-id=1 ##max_connections=2048 ##log ##log-bin ##log-slow-queries
[28 Jul 2009 5:52]
Sveta Smirnova
Thank you for the feedback. Please also provide output of SHOW GRANTS for user you run SHOW PROCEDURE STATUS as.
[29 Jul 2009 1:06]
Steven Yeung
There are two users: GRANT ALL PRIVILEGES ON *.* TO 'admin'@'%' IDENTIFIED BY PASSWORD <<PASSWORD>> WITH GRANT OPTION GRANT SELECT, INSERT, UPDATE, DELETE, CREATE, DROP, RELOAD, SHUTDOWN, FILE, REFERENCES, INDEX, ALTER, SHOW DATABASES, CREATE TEMPORARY TABLES, LOCK TABLES, EXECUTE, REPLICATION SLAVE, REPLICATION CLIENT, CREATE VIEW, SHOW VIEW, CREATE ROUTINE, ALTER ROUTINE ON *.* TO 'keith'@'%' IDENTIFIED BY PASSWORD <<PASSWORD>>
[4 Sep 2009 7:46]
Sveta Smirnova
Thank you for the feedback. I still can not repeat described behavior. Do you see crash every time when issue SHOW PROCEDURE STATUS? Please try current version 5.1.38 and if problem still exists provide dump of mysql.proc table.
[14 Sep 2009 15:14]
Steven Yeung
I guess the problem was caused by old stored procedure (from old MySQL version). All old stored procedure show in mysql.proc column 'db' is empty. After I delete those stored procedure, it doesn't occurs any more.
[14 Sep 2009 19:56]
Sveta Smirnova
Thank you for the feedback. Closed as "Not a Bug", because last comment: empty record in mysql.proc can lead to unpredictable behavior.
