this one is slightly off topic but I like to know if ICS could be 
affected by this phenomen as well (didn't yet find the time to test is):

I'm doing serial communications and have a timeout mechanism via timer.
If the other device doesn't answer within the timer interval I send out 
the same command again. This is also fully event driven (like ICS).

Now I found out that sometimes I get two answers even if the device only 
sends one. Further research showed me that there exist situations where 
windows doesn't raise the data received event but the timeout timer 
happens. Then I sent the same command again because I think (have to 
think) that a real timeout occured and get shortly after when windows 
does handle these data received events again two answers in the same 
event: the original answer and the one to my timeout retry.

One thing to force it happen is to switch betweeen 2 pcs via KVM switch 
where a USB keyboard is plugged in. The USB enumeration of the keyboard 
seems to disrupt/interrupt windows so it's not telling me that my 
expected answer has arrived.

Anyone noticed similar things yet?



To unsubscribe or change your settings for TWSocket mailing list
please goto http://lists.elists.org/cgi-bin/mailman/listinfo/twsocket
Visit our website at http://www.overbyte.be

Reply via email to