Re: shutting down mysqld

[David Rhodus <drhodus@xxxxxxxxx>]

On Nov 10, 2003, at 5:50 AM, Erik Paulsen Skaalerud wrote:

> Looks like dragonfly are having a bad time shutting down services 
> nicely
> when you hit "reboot".
>
> ###From mysql error-log:
> 031109 23:49:21  mysqld started
> InnoDB: Error: log file ./ib_logfile0 is of different size 8 5242880 
> bytes
> InnoDB: than specified in the .cnf file 0 5242880 bytes!
> 031109 23:49:21  Can't init databases
> 031109 23:49:21  Aborting
>
> 031109 23:49:21  InnoDB: Warning: shutting down a not properly started
>                  InnoDB: or created database!
> 031109 23:49:21  /usr/local/libexec/mysqld: Shutdown Complete
>
> 031109 23:49:21  mysqld ended
> ###End of log
>
> I've never ever killed any daemons before I restart, cause "reboot" 
> would
> shut them all down nicely for me.
>
> I got a hint from #dragonflybsd on EFNet that this could be a mtime 
> problem,
> but I'm no coder so I have no idea where to start.


Well, if you ran with --log, take a look at the details of what thread 
0 did
to cause the crash. In most cases its just bad corruption in the 
database files.
Though its very very unlikely that just killing mysqld would cause this 
to happen.
You should take a look at http://www.mysql.com/doc/en/Crashing.html to 
help
you figure out what the problem is.

-DR