I have not looked at the IE9 info you have put together, but am very interested in a simple soulution to the problem. Do you have a solution?
Is IE delaying providing all the request data? Or is J waiting for data that isn't coming, then timesout, and carries on and things work out because the data that didn't come wasn't needed? I can see this as happening if IE gave an incorrect count in the header but this seems unlikely. You have dug so far, can you reach a conclusion? On Mon, Jan 23, 2012 at 9:39 AM, David Mitchell <[email protected]>wrote: > (correction) > > Sorry, I trimmed the prior message that explained how I modified the JHS > core.ijs timeouts to see if I could get more information about the delays. > The > timeout in question is now: > > PC_RECVTIMEOUT=: 3000 > > The IE9 'waterfall' display shows the timeouts as larger yellow horizontal > bars. > The total time taken for each transaction is in the 'Taken' column, which > shows 3+ seconds on some of the transactions. The ones that are less than > 3 > seconds are normal processing times without a timeout, as far as I can > tell. > > And the delays I see are randomly and occasionally 3+ seconds in IE9. > > On 1/23/2012 9:21, Raul Miller wrote: > > I do not see a 5 second delay in your IE9 image. > > > > Am I overlooking something obvious? (How do I read that display?) > > > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
