On 19/09/2007, Eelco Hillenius <[EMAIL PROTECTED]> wrote: > > > Anyone tried something like this before? Any pointers at why this > > > doesn't work/ what I can do to make it work? I'm wild guessing this > > > might be because the test reuses the threads from the thread pool > > > rather than recreating them. > > > > Depends on the web application. If it has a login screen, then make > > sure the test script logs out. > > It has a login screen and relies on jsessionid/ Servlet API's httpsessions. > > The thing is though, that I'd like to test new sessions being built up > without removing the old ones. So I would like to do e.g. 150 > iterations of 100 threads (/users) to end up with 15,000 logged in > users. > > The reason I want to do that is because I want to measure memory > consumption (which is always a bit tricky to extrapolate with Java, so > nothing beats actual logins). > > I hoped that this would work as I'm clearing my cookies, thus forcing > my app server (Jetty 6 btw) to create new ones. But that doesn't seem > to work. I tried with turning off keep-alive as well. Any idea if I'm > just doing something stupid? :-)
Try checking the jsession ids after login and on subsequent samples to see if they change. Are you using different login ids for each user? > Cheers, > > Eelco > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
