Sleep Question

[Guest]

On our production server, we have a great deal of data. To improve
performance, we give the user a "page" of data at a time and then spawn count
of the total number of products based on the query parameters on a separate
thread. The user can change their mind, of course, and change their query
before the count has returned for the previous query. We are trying to
simulate a lot of data on our developement server by putting a sleep command
(dbms_lock.sleep) in the stored procedure. I'm having problems terminating
the thread. I've used the bool variable as even abort, but it still takes the
full sleep time. Is the Oracle sleep command simular to the .Net command and
is not interruptable? It would be much easier to just load more data into teh
tables or at least do add some monster query into it for testing, but the DBA
says this should be just as good.

Thanks,
Susan

 

[Greg Young]

Your abort is killing the thread in your application but it sounds like your
command is still running on the database (is this how you are measuring that
it is still "alive"?)

Cheers,

Greg Young
MVP - C#
http://codebetter.com/blogs/gregyoung

 

[Nick Malik [Microsoft]]

your DBA is wrong. Use a good stress testing tool to generate the actual
volume rather than trying to simulate it with a hack.


--
--- Nick Malik [Microsoft]
MCSD, CFPS, Certified Scrummaster
http://blogs.msdn.com/nickmalik

Disclaimer: Opinions expressed in this forum are my own, and not
representative of my employer.
I do not answer questions on behalf of my employer. I'm just a
programmer helping programmers.

 

[John Timney \(MVP\)]

I can see what your DBA is trying to do, but its not a real measure of
stress testing as all your testing is the stored proc holding up the
conclusion of a request. In theory, all requests would take the duration of
the sleep timeout and thats not how systems work.

Get a proper stress testing tool and let your procedures work as they should
be in a live environment.

Regards

John Timney (MVP)

 

[Guest]

Actually, the thread running the process doesn't die till the stored
procedure comes back for some reason. I had tried the OracleCommand.Cancel
on the stored procedure and then ending the thread, but that seems to hang as
well. It also seems to prevent additional stored procs from running till it
is done.