On 9 February 2011 12:00, SanderW <[email protected]> wrote: > > Thanks for your response, sebb > > > sebb-2-2 wrote: >> >> Latency is time to *first* response, so it does not make sense to sum it. >> > But it it correct to show a latency of zero when in fact both individual > http requests have latency values?
Would have to look at the code - maybe it was decided that latency made no sense for multiple requests, or maybe it was just overlooked. In fact what would make sense as latency here? - minimum of all - maximum of all - average of all - start of first sample to first response of last sample? > The guys from the JMeterPlugins created a LatenciesOverTime listener, but > this apparantly does not work if you use transaction controllers. > > > sebb-2-2 wrote: >> >> Load time includes any timers or delays between samples. >> >> Have a look at the start and end times of the individual samples and >> the elapsed should be equal to (last end - first start) >> > > Ah yes, I see that between the start of the transaction controller and the > start of the first http request already is difference of 2ms. > > The (start time of the last embedded resource + load time of the last > embedded resource) - start time of the transaction controller is indeed the > load time of the transaction controller. Good - so the elapsed time is the overall time for all the child samples. > -- > View this message in context: > http://jmeter.512774.n5.nabble.com/Load-time-and-latency-in-transaction-controller-tp3377271p3377450.html > Sent from the JMeter - User mailing list archive at Nabble.com. > > --------------------------------------------------------------------- > 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]
