X-Spam-Check-By: sourceware.org
Date: Wed, 11 Oct 2006 10:14:22 -0400
From: Christopher Faylor <cgf-no-personal-reply-please AT cygwin DOT com>
To: cygwin AT cygwin DOT com
Subject: Re: fork.cc (frok::parent) problems after 3-13-06 snapshot
Message-ID: <20061011141422.GB24362@trixie.casa.cgf.cx>
Reply-To: cygwin AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
References: <loom DOT 20061011T071953-203 AT post DOT gmane DOT org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <loom.20061011T071953-203@post.gmane.org>
User-Agent: Mutt/1.5.11
Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
Precedence: bulk
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

On Wed, Oct 11, 2006 at 05:54:04AM +0000, Steve wrote:
>Hello Folks,
>
>I origianly posted this problem on the group on Oct.21 2006 under title "1.5.21
>Forked background processes". The problem involves forking processes in the
>background using `backticks` from a sh script. When doing this, the parent and
>child (backticked) processes will both die shortly and without error, although
>ps will still show the child process as alive. You can reproduce this problem
>using the example I added in my original post. Other folks I know and on the
>group were able to reproduce this problem with little effort.
>
>Since then I nailed down the culprit to a change that happened in method
>frok::parent in file fork.cc. The change to this method first appreared in the
>snapshot from 3-13-06. Here is the change log entry for this method:
>
>(frok::parent): Reorganize to allow retry of failed child creation if
>	child signalled that it was ok to do so.
>
>Reverting this change (to the 3-9-06 snapshot), as well as the change to fork.cc
>(resume_child) in the same snapshot, fixed the problem. Reverting this, of
>course, invalidates the fork_retry (now proc_retry) CYGWIN environment option,
>which was one of the reasons this method was reorginized to accomidate. 
>
>I can't quite nail down the problem within this method however, but I'm guessing
>it has something to do with the retry iteration that's present. I consider this
>a rather nasty bug. 

If you've narrowed it down this far then it should not be much more work to
determine what *specifically* is causing the problem.

cgf

--
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/