Well as I said, I first wanted to have a robust scenario to start from and to further elaborate it. So, in order to give you feedback, I made 2 tests : the first with a pool of 1 connection and a reuse factor of 1. The second is with a pool of 10 and a reuse of 1. Those values have been defined in a "Database Connection Pool defaults" entry. I guess it does not represent something "real" in terms of database access but it does not crash anymore, which is the primary condition to enable further investigation. Cheers,
Serge On Sunday 10 August 2003 04:54 pm, [EMAIL PROTECTED] wrote: > That's interesting - I'm curious, what numbers did you choose for number of > connections in the pool and number of re-uses allowed? > > -Mike > > On 10 Aug 2003 at 16:45, serge van Thiel wrote: > > Hi Mike and Jeremy, > > > > Thank you very much for your replies and sorry for being late in my > > reaction. Here are the results of my tests : > > Rel1.9 standard keeps crashing for the same reason on my very simple one > > thread SELECT. > > Rel1.9 with the 2 new jars from Jeremy does not crash at all and is, in > > that oversimplified test, even 50 times faster than any previous result I > > have been recording. > > The logfile (jmeter.log) does not show any abnormal behaviour as far as I > > can > > > see. It even includes much more details. > > Given the few and non exhaustive tests I have made, It seems that you > > made a > > > great job. > > > > Thanks again, > > > > Serge van Thiel > > Ph.:+32 2 3752277 > > Mob.:+32 477 414543 > > email: [EMAIL PROTECTED] > > > > --------------------------------------------------------------------- > > 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]
