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
Message-ID: <3A9844BC.7485D813@ece.gatech.edu>
Date: Sat, 24 Feb 2001 18:33:16 -0500
From: "Charles S. Wilson" <cwilson AT ece DOT gatech DOT edu>
X-Mailer: Mozilla 4.75 [en] (WinNT; U)
X-Accept-Language: en
MIME-Version: 1.0
To: cygwin AT cygwin DOT com
CC: MADHU <madhu AT quickmonkey DOT com>
Subject: Re: cygwin with sockscap32
References: <200102222041 DOT f1MKfKj29110 AT quickmonkey DOT com> <20010224164002 DOT B6385 AT redhat DOT com> <200102242149 DOT f1OLns802613 AT quickmonkey DOT com> <20010224175137 DOT A6846 AT redhat DOT com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit



Chris Faylor wrote:

> However, the only way the problem is going to be narrowed down is if
> you, or someone, debugs sockscap.  The sources *must* be available given
> Cygwin's licensing agreement.

Not really -- see my other message.  Sockscap doesn't link or use cygwin
itself -- it modifies the networking layer of the machine on which it is
installed.  Actually, that means that on MADHU's machine, *cygwin* is
using *sockscap*, not the other way around.  So cygwin's license doesn't
apply.

> In my opinion, the facts that you have provided so far do not offer
> enough details to allow anyone to debug your problem.  That leaves the
> burden on you to either provide more facts or debug the problem yourself.

Yep.  I just spent some time researching sockscap and poring over the
cygwin mailing list -- and that's the limit of my interest and
involvement.  If someone wants to fix this, all well and good -- but it
ain't gonna be me.

The problem seems to be that sockscap makes assumptions about how
applications will access the network, which were valid for B20 apps but
not for V1.1 apps.  Perhaps some changes in cygwin1.dll's networking
code; I dunno.  The sockscap web page was NOT informative about exactly
WHAT those assumptions were, and they do NOT release their source code. 
So, even if we were inclined to change cygwin so that sockscap is happy,
we have no way of knowing exactly what calls to avoid or what
assumptions sockscap is making.

> Given that I haven't seen any sockscap experts stepping forward to offer
> in-depth insight, I fear that the burden of debugging this problem may
> be yours.

Looks that way to me.

--Chuck

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