On 2 December 2010 08:36, rkannan <[email protected]> wrote: > > Yes I'm running the tests in same machine. > > Results are same. But non-gui mode takes lot of time to finish the test? > > For example: > > If i'm running 2 users in GUI mode means it finish the test in 3 mins. If > i'm running 2 users in Non-GUI mode means it take more than 30 mins to > finish the test.
That does not sound right; the same code is used for sampling regardless of whether you are using GUI or non-GUI mode. The only difference I can think of is that the non-GUI mode will wait until all threads (non-daemon) have finished before exitting, whereas GUI mode can be exitted at any point. Perhaps there is a thread which is not shutting down properly. This would not necessarily be noticed in GUI mode. Can you create a Bugzilla issue, and attach the jmeter.log files from GUI and non-GUI runs? Also ideally attach the JTL and JMX files (but redact any sensitive information first - e.g. overwrite with xxx) > -- > View this message in context: > http://jmeter.512774.n5.nabble.com/Jmeter-Non-GUI-mode-tp3288983p3289046.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]
