> The problem within the current Apache+Cygwin (normal setup) is within > the accept() serialization, which means the mechanism that the core > parent httpd prozess deals with incoming TCP connect and spreads the > load across the idle child httpd within the scoreboard file.
Stipe, have you tested with the latest or semi-latest snapshot? Corinna made a change to the socket code that fixed the socket state to being set to CONNECTED...just a thought ;-) Would try myself but am behind bars infront of a vax brrrrrrrrrrrrrr. Regards, Elfyn McBratney [EMAIL PROTECTED] www.exposure.org.uk -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Bug reporting: http://cygwin.com/bugs.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/