Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
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
Mail-Followup-To: cygwin AT cygwin DOT com
To: cygwin AT cygwin DOT com
From: =?ISO-8859-1?Q?J=F6rg?= Schaible <joerg DOT schaible AT gmx DOT de>
Subject: Re: scp exits often with -1
Date: Thu, 18 Nov 2004 21:00:38 +0100
Lines: 41
Message-ID: <cniv16$534$2@sea.gmane.org>
References: <20041118155449 DOT GG10795 AT trixie DOT casa DOT cgf DOT cx> <20041118172729 DOT D51B6E523 AT wildcard DOT curl DOT com> <20041118185309 DOT GA246157 AT Worldnet>
Reply-To: joerg DOT schaible AT gmx DOT de
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8Bit
X-Complaints-To: usenet AT sea DOT gmane DOT org
X-Gmane-NNTP-Posting-Host: pd9eb8f53.dip0.t-ipconnect.de
Mail-Copies-To: never
User-Agent: KNode/0.8.1
X-IsSubscribed: yes

Pierre A. Humblet wrote:

> On Thu, Nov 18, 2004 at 12:27:29PM -0500, Bob Byrnes wrote:
>> > If this is a problem with the new pipe code then maybe Bob Byrnes could
>> > offer some insight.
>> 
>> Offhand, I can't think of any way that the new pipe stuff could cause
>> this behavior, but I'll add this to my list of (potential) pipe-related
>> things to investigate and think about.
> 
> From what I can see the failure is related to pipes:
> 
>    41  673146 [main] ssh 4392 cygwin_select: 7, 0x100EEBF8, 0x100EEB70,
>    0x0, 0x0
>    65  673211 [main] ssh 4392 dtable::select_read:  fd 3
>   662  673873 [main] ssh 4392 dtable::select_read:  fd 4
>    33  673906 [main] ssh 4392 cygwin_select: to NULL, ms FFFFFFFF
>    34  673940 [main] ssh 4392 cygwin_select: sel.always_ready 0
>   307  674247 [main] ssh 4392 start_thread_socket: Handle 0x634
>    33  674280 [main] ssh 4392 start_thread_socket: Added to readfds
>    34  674314 [main] ssh 4392 start_thread_socket: exitsock 0x620
>    33  674347 [main] ssh 4392 start_thread_socket: stuff_start 0x22EBE4
>    98  674445 [main] ssh 4392 select_stuff::wait: m 3, ms 4294967295
>    43  674488 [select_socket] ssh 4392 thread_socket: stuff_start
>    0x101196BC
>    71  674559 [select_pipe] ssh 4392 peek_pipe: , PeekNamedPipe failed,
>    Win32 error 109
> 
> 109 is "The pipe has been ended". Isn't that strange when reading a pipe?
> Unfortunately fd 4 was opened in a part of the trace that we don't have.

I'll have a look tomorrow in the office, where fd 4 came from (thought I
posted all related stuff, but trace is hard). Still have the full protocol
though.

> Sending a url to the full trace is more useful (and less controversial)
> than sending parts.

You're right. Will try to find a place online tomorrow.

- Jörg


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