Mailing-List:	contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm
List-Subscribe:	<mailto:cygwin-subscribe AT sources DOT redhat DOT com>
List-Archive:	<http://sources.redhat.com/ml/cygwin/>
List-Post:	<mailto:cygwin AT sources DOT redhat DOT com>
List-Help:	<mailto:cygwin-help AT sources DOT redhat DOT com>, <http://sources.redhat.com/ml/#faqs>
Sender:	cygwin-owner AT sources DOT redhat DOT com
Delivered-To:	mailing list cygwin AT sources DOT redhat DOT com
Date:	Sun, 15 Apr 2001 17:34:15 -0700
From:	Enoch Wu <enochw AT scn DOT org>
To:	cygwin AT cygwin DOT com
Subject:	Re: cvs broken on win2k after recent cygwin upgrade: "The descriptor is a file, not a socket"
Message-ID:	<20010415173415.A4109115@MELON>
Reply-To:	Enoch Wu <enochw AT scn DOT org>
Mail-Followup-To:	Enoch Wu <enochw AT scn DOT org>, cygwin AT cygwin DOT com
References:	<Pine DOT GSO DOT 4 DOT 21 DOT 0103210129220 DOT 24576-100000 AT sunshine DOT cs DOT uu DOT nl> <002e01c0b29d$c880d5b0$0401a8c0 AT trinity> <f5by9twn4mm DOT fsf AT cogsci DOT ed DOT ac DOT uk> <010e01c0bafa$f04d9320$2b904f86 AT trinity> <f5bofu2cm9s DOT fsf AT cogsci DOT ed DOT ac DOT uk> <07fd01c0c3dc$cf05f730$0200a8c0 AT voyager> <f5bhezs4e2c DOT fsf AT cogsci DOT ed DOT ac DOT uk> <20010414161237 DOT A4149051 AT MELON> <f5b8zl23731 DOT fsf AT cogsci DOT ed DOT ac DOT uk>
Mime-Version:	1.0
User-Agent:	Mutt/1.2.5i
In-Reply-To:	<f5b8zl23731.fsf@cogsci.ed.ac.uk>; from ht@cogsci.ed.ac.uk on Sun, Apr 15, 2001 at 09:22:58AM +0100
X-Mailer:	Mutt 1.2.5i <http://www.mutt.org>
X-Operating-System:	CYGWIN_98-4.10 1.1.8(0.34/3/2) i586<http://www.cygwin.com>
X-Sender:	ewu AT eskimo DOT com

Henry,

On Sun, Apr 15, 2001 at 09:22:58AM +0100, Henry S. Thompson wrote:
> Thanks for the observations.  Seems to me to confirm Corinna's
> original belief that somehow Cygwin is getting the wrong WINSOCK
> library, even though the right one is available.

Yes. It seems that Cygwin is getting the wrong file handle from winsock. 
Why would Cygwin get a file when it requests a socket from MS winsock? 
That is beyond my understanding. I can dig a little deeper:

The message "The descriptor is a file. not a socket" is a Cygwin error 
message from src/winsup/cygwin/errno.cc as far as I can tell. 
The evidence is a code fragment from that routine:

case ENOTSOCK:
      error = "The descriptor is a file, not a socket";
      break;

Cygwin itself detected a non-socket and then printed that error message.
Winsock did not print that error message on your screen. 

It appears that a Cygwin function just returned a file descriptor instead 
of a socket, or that it mistook the socket as a file descriptor. Could it
be possible that winsock actually returned a socket but the socket was
mistaken as a file descriptor? That is beyond my understanding. 
However, this anomaly manifested itself only when a non-MS dialup networking
program was running (in my case). The non-MS dialup software was causing
problem to Cygwin only --- Windows telnet/ftp was not being affected at
all.

The non-MS dialup software has since been removed from my computer. I can no
longer reproduce that same problem. 

If you know C and C++, would you be able to debug it? (since you can reproduce
the problem).  You are the only person that can provide more information.

Enoch










--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple

- Raw text -