Actually, the O_NONBLOCK thing is new behaviour introduced in 2.0.49 with the "multi-listen DoS" security fix... so maybe this is a recent regression, and it really counts as an httpd bug. (or have you been getting these similar failure reports on Solaris/... for ages?)
joe --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
