Mail Archives: cygwin/2002/04/10/11:26:02
On Wed, Apr 10, 2002 at 10:26:09AM -0400, Corinna Vinschen wrote:
> On Wed, Apr 10, 2002 at 12:46:50PM +0200, Corinna Vinschen wrote:
> >On Wed, Apr 10, 2002 at 04:59:51AM -0400, Steve Chew wrote:
> >> Thank you for the swift response! Are you saying that the
> >> problem is the the Windows 98 Winsock code? Lovely... How would you
> >
> >It's not restricted to 9x/Me since XP (and W2K) show the same - even
> >if slightly different - phenomenon. Looking into the strace shows
> >that the socket is nonblocking. I could imagine that a non-blocking
> >socket is "listening" when read() is called on it for the first time
> >or something like that. Who knows how that's implemented in Winsock?
>
> It's probably the listen in start_thread_socket in select().
>
> The linger value for the dummy socket that gets opened is supposed to
> be set to zero, though, to eliminate this kind of thing.
The weird thing is, that these listening sockets never go away as
long as the application is running, not after 2MSL (4 minutes in
Windows) nor after 20 minutes.
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Developer mailto:cygwin AT cygwin DOT com
Red Hat, Inc.
--
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/
- Raw text -