if you killed the query and was the wrong one you should have to look better next time. I don't see probable that query ids had to be reused (unless we are talking here from a really uncommon scenario where you make queries fast enough to overflow and get old ids again...)

I recommend you to use innotop [1] or mtop [2] to monitor your long-running queries and kill the right one. Also, there is other kind of information like user and host that you can use to "identify" the query.

Remember, this is not MS Windows, so you won't be asked twice :D

[1] http://code.google.com/p/innotop/
[2] http://mtop.sourceforge.net/

On Fri, Aug 16, 2013 at 4:48 PM, Roberto Spadim <roberto@spadim.com.br> wrote:

hi guys, i hit a problem in this exact second...
some programs are running and i want to kill only one big query

well i executed
kill 17143

ok.. but when i execute this command, the query was done, and starting a second query... in other words... i killed the wrong query

could we add a hash information to kill, to make it more secure and precise?
for example...

kill 17143 md5 "123456789......."
where md5 = id + start time of query + others informations to make it relative unique

maybe it exists and i don't know... any other idea how to solve this?

thanks

--
Roberto Spadim

_______________________________________________
Mailing list: https://launchpad.net/~maria-discuss
Post to : maria-discuss@lists.launchpad.net
Unsubscribe : https://launchpad.net/~maria-discuss
More help : https://help.launchpad.net/ListHelp

--
Gabriel Sosa
Sometimes the questions are complicated and the answers are simple. -- Dr. Seuss