I've been checking this a bit... originally my custom java sampler was causing errors because I was improperly initializing and using the SampleResult which I needed to return.
I fixed those errors, ran the test again (20 threads with 100 iterations this time) and got 1545 actual executions (again, lines in my separate log file) out of 2000 specified, and now in jmeter.log there is only one section which reports errors, an HTTP sampler got an IOException somewhere about halfway through the test. It doesn't appear related to the dropped executions. All the threads terminate at approximately the same time at the end of the test. "Thread Thread Group1-13 is done", etc appears for each thread. I checked through the bugs on the website and didn't see any that dealt with this sort of behavior. The memory and CPU usage on my machine never hit any kind of critical level at all (1 gig of RAM, P4 1600MHz). I wish I had something concrete to hand you and say "see?" but I don't =( Let me know if you would like to take a look at anything (jmeter.log, etc). It's not going to totally break my tests to have the threads terminate early, but it's one of those "grain of salt" issues I need to bring up when I present my findings/results from this test project. Thanks, Brad Denkmann [EMAIL PROTECTED] -----Original Message----- From: Michael Stover [mailto:[EMAIL PROTECTED] Sent: Wednesday, August 11, 2004 4:53 PM To: JMeter Users List Subject: Re: Jmeter not executing all specified threads/loops...? My first guess would be that a sampler in some thread is failing and crashing the thread. Maybe you could make your java sampler logger also record the thread number so you can see if certain threads are stopping at some point. JMeter's log file might have errors in it for you to look at. -Mike On Wed, 2004-08-11 at 17:13, Brad Denkmann wrote: > Hi, > > For a given test plan, I have 30 threads in a thread group, set to run > for 100 iterations. > > I noticed a little while ago that the threads do not seem to be > executing as many times as they should be... out of 3000 total > executions specified above, I get just under 1600 actual executions. > > I measure this by having a simple java request at the beginning of the > test, which simply logs the thread number to a file. So I should see > 3000 entries in the file after the test... however, as I said, I only > get ~1600. > > Any thoughts on this? I have no ideas as to why this would be happening > (or not happening, as the case may be). > > > > Brad Denkmann > [EMAIL PROTECTED] > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] -- Michael Stover <[EMAIL PROTECTED]> Apache Software Foundation --------------------------------------------------------------------- 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]
