Date: Wed, 30 May 2012 17:53:32 +0200
From: Corinna Vinschen <corinna-cygwin@cygwin.com>
To: cygwin@cygwin.com
Subject: Re: procps and pgrep hang
Message-ID: <20120530155332.GC26411@calimero.vinschen.de>
Reply-To: cygwin@cygwin.com
Mail-Followup-To: cygwin@cygwin.com
References: <201205161650.q4GGoiv3000900@localhost.localdomain> <201205170146.q4H1k9AJ001926@localhost.localdomain> <20120517020441.GA10314@ednor.casa.cgf.cx> <201205211640.q4LGeucD023330@localhost.localdomain> <20120523155327.GJ9200@calimero.vinschen.de> <201205301312.q4UDCcIO004057@localhost.localdomain>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <201205301312.q4UDCcIO004057@localhost.localdomain>
User-Agent: Mutt/1.5.21 (2010-09-15)
Mailing-List: contact cygwin-help@cygwin.com; run by ezmlm
Precedence: bulk
Sender: cygwin-owner@cygwin.com
Mail-Followup-To: cygwin@cygwin.com

On May 30 08:12, Tom Rodman wrote:
> On Wed 5/23/12 17:53 +0200 Corinna wrote:
> > On May 21 11:40, Tom Rodman wrote:
> > > On Wed 5/16/12 22:04 EDT Christopher Faylor wrote:
> > > > On Wed, May 16, 2012 at 08:46:09PM -0500, Tom Rodman wrote:
> > > > >  $ pgrep bash
> > > --snip
> > > > >   $ procps -o pid,args
> > > > >   PID COMMAND
> > > > >  4580 procps -o pid,args
> > > > >  --note: above command hangs, ended up killing it
> > > >
> > > > Try a snapshot.
> > > 
> > > Thanks.  It seems better, but still hangs, on this box (and 1 other).
> > > 
> > > Examples ran tonight w/20120517 snapshot:
> > > 
> > >   http://pastebin.com/xQbtChsJ
> > 
> > Looking into your strace I'm wondering if you really mean hanging in
> > the sense of "hanging indefinitely until killed", 
> 
> That's what I mean - I have to kill it, it hangs forever.
> 
> > or if you mean it
> > hangs for about half a second.  Since that's what the strace shows.
> > For some of the information about a process under /proc, the requesting
> > process has to open a pipe and then, by using a special signal, informs
> > that process that it requests information which in turn is send over
> > the pipe back to the requester.  The timeout for each of the read 
> > operations on the pipe is restricted to 500ms to avoid indefinite
> > hangs.
> > 
> > Unfortunately I can't reproduce this effect, neither on XP nor on W7.
> > 
> > Hmm, puzzeling.  I'm wondering though, if this may be a result of using
> > message type pipes.  Nothing else comes to mind which has changed in
> > this piece of code lately.  For testing you could set the environment
> > variable CYGWIN to "pipe_byte" and try again.
> 
> On my laptop, only the pgrep bug shows up, here is the 'forever hang'
> strace w/CYGWIN set to pipe_byte:

A signal delivery hangs as if the receiver never reads its signal pipe.
I still can't reproduce the problem, though.  Can you try to find out
which snapshot introduced the problem?


Thanks,
Corinna

-- 
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader          cygwin AT cygwin DOT com
Red Hat

--
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