X-Recipient: archive-cygwin AT delorie DOT com
X-SWARE-Spam-Status: No, hits=-2.6 required=5.0 	tests=BAYES_00
X-Spam-Check-By: sourceware.org
Message-ID: <4AFD49EA.5020804@cs.umass.edu>
Date: Fri, 13 Nov 2009 06:58:34 -0500
From: Eliot Moss <moss AT cs DOT umass DOT edu>
User-Agent: Thunderbird 2.0.0.23 (Windows/20090812)
MIME-Version: 1.0
To: cygwin AT cygwin DOT com, cygwin AT cygwin DOT com
Subject: Re: Problem with rsync 3.0.6-1 [and 3.0.5] under 1.7.0-62 and 63       [and 64]
References: <4AF58831 DOT 3080306 AT cs DOT umass DOT edu>  <4AFC2A57 DOT 60909 AT cs DOT umass DOT edu>  <20091112154701 DOT GA21323 AT calimero DOT vinschen DOT de>  <4AFC3204 DOT 8080505 AT cs DOT umass DOT edu>  <4AFC8AF3 DOT 2040009 AT cs DOT umass DOT edu> <20091113112405 DOT GO26238 AT calimero DOT vinschen DOT de>
In-Reply-To: <20091113112405.GO26238@calimero.vinschen.de>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-IsSubscribed: yes
Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Id: <cygwin.cygwin.com>
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sourceware.org/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sourceware.org/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com

Corinna Vinschen wrote:
> On Nov 12 17:23, Eliot Moss wrote:
>> I went ahead and wrote a little program that narrows
>> down the rsync problem to a dup2 call. The program:
>> creates two pipes (for talking to a child process),
>> forks the child, and the child tries to dup2 the
>> pipe fds to its stdin and stdout. If it wins (which
>> it doesn't), it will then run sleep for 5 seconds
>> and quit. The parent closes some fds it doesn't need
>> and waits for the child, then quits. I attach the
>> program in question, and strace output.
>>
>> Cheers -- Eliot Moss
> 
> Thanks for the testcase.  However, the problem is that I can't reproduce
> any problem using your testcase.  I ran it on Windows XP SP3, Server 2008
> SP1, and Windows 7.  The result is this, just with changing PIDs:
> 
> $ ./a_test
> pid = 3272, fin = 5, fout = 4
> wait got 0
> 
> I assume you see some error, but you didn't explicitely write what error
> you see.  The strace output shows an error when closing a socket, which
> is that dreaded Winsock error 10038, "socket operation on non-socket".
> 
> The reason why this operation fails must be something on your machine.
> I'm just not sure how we can find out what that is, and how to avoid the
> error then.  Simply ignoring error 10038 in close() doesn't sound like
> a terribly good idea...

Thank you for looking more closely!

The actual failure is earlier in the strace output, I think. If you
look for dup2 you see that the first dup2 call fails on my machine.
Can this fail because of some permissions thing? I have UAC set to
"Never notify". I am logged in as me (Eliot), and I have Administrator
rights. I get the same result from a_test if I run bash as Administrator
and run a_test.

Here is sample program output for me:

pid = 9064, fin = 5, fout = 4
error: first dup2 -1 9
wait got 65280

If I understand correctly, the wait code says the child terminated
with result -1, which corresponds with what the code does when that
dup2 fails.

Any suggestions for the next step in debugging this? I can certainly
accept that it's something about my machine, but I am rather at a loss
of where to look ...

Regards -- Eliot

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple