X-Recipient: archive-cygwin AT delorie DOT com
X-SWARE-Spam-Status: No, hits=-0.7 required=5.0 	tests=BAYES_20,SPF_PASS
X-Spam-Check-By: sourceware.org
Date: Thu, 30 Jul 2009 13:09:21 +0000 (UTC)
From: Eric Blake <ericblake AT comcast DOT net>
To: David Antliff <david DOT antliff AT gmail DOT com>
Cc: cygwin AT cygwin DOT com
Message-ID: <770992004.6687041248959361064.JavaMail.root@sz0059a.emeryville.ca.mail.comcast.net>
In-Reply-To: <db95995b0907291911x6fb8a8c9lf91b84417e298f7f@mail.gmail.com>
Subject: Re: bash pipeline exit code - possible race condition?
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
Precedence: bulk
List-Id: <cygwin.cygwin.com>
List-Unsubscribe: <mailto:cygwin-unsubscribe-archive-cygwin=delorie DOT com AT cygwin DOT 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

> >>I've noticed a strange problem with bash pipelines in Cygwin that
> >>might indicate some sort of race condition. I cannot reproduce the
> >>problem on a Linux system, but it seems easy to reproduce in
> Cygwin.
> >
> > This doesn't appear to be a cygwin problem. =C2=A0I get occasional
> errors
> > on linux as well.
>=20
> I tried to reproduce on Linux but without success - did you simply
> run
> my script or perhaps something else? Do you see error "141" as well?

Status 141 is from SIGPIPE.  Basically, it happens when the read end of the
pipe exits before the writer.  It is not specific to cygwin, but may occur
more frequently on cygwin since cygwin has different scheduling than
Linux.  It is not an error in the OS - if you are going to worry about
the exit status of all processes in a pipeline, then you must be prepared
to deal with the earlier processes ending abruptly because the later
processes aren't interested in any more data.  There's nothing we can
do to change this race, and complaining to the bash folks won't change
it either.  About the only thing you could do by complaining to the bash
folks is to get another shell option that ignores sigpipe failures when
you are paying attention to all statuses within a pipeline, rather than
the normal behavior of just the last status.

--=20
Eric Blake

--
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