Mailing-List: contact cygwin-developers-help AT sourceware DOT cygnus DOT com; run by ezmlm
Sender: cygwin-developers-owner AT sourceware DOT cygnus DOT com
List-Unsubscribe: <mailto:cygwin-developers-unsubscribe-archive-cygwin-developers=delorie DOT com AT sourceware DOT cygnus DOT com>
List-Archive: <http://sourceware.cygnus.com/ml/cygwin-developers/>
List-Help: <mailto:cygwin-developers-help AT sourceware DOT cygnus DOT com>,
	<http://sourceware.cygnus.com/ml/#faqs>
Delivered-To: mailing list cygwin-developers AT sourceware DOT cygnus DOT com
From: Chris Faylor <cgf AT cygnus DOT com>
Date: Fri, 30 Jul 1999 10:07:23 -0400
To: Earnie Boyd <earnie_boyd AT yahoo DOT com>
Cc: cygwin developers <cygwin-developers AT sourceware DOT cygnus DOT com>
Subject: Re: problems with the latest snapshot
Message-ID: <19990730100723.A713@cygnus.com>
References: <19990730123410 DOT 17983 DOT rocketmail AT web105 DOT yahoomail DOT com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 0.95.6i
In-Reply-To: <19990730123410.17983.rocketmail@web105.yahoomail.com>; from Earnie Boyd on Fri, Jul 30, 1999 at 05:34:10AM -0700

On Fri, Jul 30, 1999 at 05:34:10AM -0700, Earnie Boyd wrote:
>> >  it seems that there're some problems with signal handling in the
>> >latest snapshots the first one is that bash freezes at startup if it's
>> >being started from sh. It seems that this error prevents sshd from
>> >running user shell.
>> 
>> I think that this is really an ASH bug.  I have been playing with a
>> new version of ASH for some time.  If anyone wants to try it, go to
>> 
>> ftp.cygnus.com:/private/home/cgf
>> 
>> and retrieve
>> 
>> sh.exe.bz2	# the executable
>> 
>> and
>> 
>> ash.tar.bz2	# the source
>
>Actually, I saw this behaviour a couple of nights ago when I was working with
>Corinna on her security patch.  By the way, I think she has that problem
>resolved.  I sent her a strace of this bash freezing when started from sh
>problem.  It doesn't just freeze, it goes into an endless loop which is really
>noticable in the strace file.  This endless loop is IMO not a problem with ash
>but a runtime problem.

The bash hanging behavior that Egor Duda reported is easy to reproduce
and I know exactly why it happened.  I can trace it back to some early
snapshots where it only occurred if you set CYGWIN=tty.  As I have been
consolidating the functionality of ttys and console devices, I ended up
"breaking" this behavior under normal console I/O, too.  It is
definitely an ash problem.  You'll notice the same behavior if you port
ash to linux and run bash under it.

This is probably unrelated to the problems that you were experiencing.

cgf