From owner-freebsd-hackers@FreeBSD.ORG  Tue Jun  8 13:13:50 2010
Return-Path: <owner-freebsd-hackers@FreeBSD.ORG>
Delivered-To: freebsd-hackers@freebsd.org
Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34])
	by hub.freebsd.org (Postfix) with ESMTP id E012A1065677
	for <freebsd-hackers@freebsd.org>; Tue,  8 Jun 2010 13:13:50 +0000 (UTC)
	(envelope-from tss@iki.fi)
Received: from dovecot.org (dovecot.org [82.118.211.50])
	by mx1.freebsd.org (Postfix) with ESMTP id 9FBF28FC16
	for <freebsd-hackers@freebsd.org>; Tue,  8 Jun 2010 13:13:50 +0000 (UTC)
Received: from [10.134.134.32] (pickles.tp.telepac.pt [194.65.5.228])
	by dovecot.org (Postfix) with ESMTP id 1E8ABFA89B4
	for <freebsd-hackers@freebsd.org>;
	Tue,  8 Jun 2010 15:51:49 +0300 (EEST)
From: Timo Sirainen <tss@iki.fi>
To: freebsd-hackers@freebsd.org
Content-Type: text/plain; charset="UTF-8"
Date: Tue, 08 Jun 2010 13:51:47 +0100
Message-ID: <1276001507.6691.388.camel@kurkku.sapo.corppt.com>
Mime-Version: 1.0
X-Mailer: Evolution 2.28.3 
Content-Transfer-Encoding: 7bit
Subject: close() failing with ECONNRESET
X-BeenThere: freebsd-hackers@freebsd.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: Technical Discussions relating to FreeBSD
	<freebsd-hackers.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-hackers>, 
	<mailto:freebsd-hackers-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-hackers>
List-Post: <mailto:freebsd-hackers@freebsd.org>
List-Help: <mailto:freebsd-hackers-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-hackers>,
	<mailto:freebsd-hackers-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Tue, 08 Jun 2010 13:13:51 -0000

I see that since FreeBSD 6.3 close() can fail with:

> [ECONNRESET]	The underlying object was a stream socket that was
> 		shut down by the peer before all pending data was
> 		delivered.

Could someone explain what this is useful for? I'm not aware of any
other OS that does this. Is this really something that many programs
care about? I'd think there are only very few, and those exceptions
could use some other syscall before close() to find out about it.
Instead now you're forcing everyone else to change their code from:

if (close(fd) < 0) log(..);

to 

if (close(fd) < 0 && errno != ECONNRESET) log(..);

or to write some wrapper to close().