On 26/10/2007, Christiaan Lamprecht <[EMAIL PROTECTED]> wrote: > > I suspect the larger size is caused by the error message when the > > timeout happens. > > > > But this shouldn't take up any bandwidth, I assume the error message > is not from the server, the client just times out and continues. The > test plan takes much longer when time-outs are enabled and I assumed > it was the bandwidth due to the 1815 file size. >
I don't understand why enabling timeouts should increase the run-time - it might decrease, but should not increase. If you don't use timeouts, do all the requests finish OK, or do some hang? I'm not clear why you wanted to enable timeouts... > > > It would help if the success and response code fields were included in > > the items logged. > > timeStamp,elapsed,label,responseCode,responseMessage,threadName,dataType,success,failureMessage,bytes,grpThreads,allThreads,URL,Filename,Latency,Encoding > > > 1193316042700,32,HTTP Request HTTPClient,200,OK,Thread Group > 1-1,text,true,,44,7,7,https://cala01.ac.uk:8083/index.html,,32,ISO-8859-1 > > > 1193316042484,203,HTTP Request HTTPClient,Non HTTP response code: > java.net.SocketTimeoutException,Non HTTP response message: Read timed > out,Thread Group > 1-4,text,false,,1815,7,7,https://cala01.ac.uk:8083/index.html,,0,ISO-8859-1 This shows that the longer elapsed time is associated with the timeout; I assume all the longer responses are also timeouts? > > --------------------------------------------------------------------- > 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]
