From owner-freebsd-bugs  Thu Jun 12 09:51:07 1997
Return-Path: <owner-bugs>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id JAA07048
          for bugs-outgoing; Thu, 12 Jun 1997 09:51:07 -0700 (PDT)
Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2])
          by hub.freebsd.org (8.8.5/8.8.5) with SMTP id JAA07022
          for <freebsd-bugs@hub.freebsd.org>; Thu, 12 Jun 1997 09:50:57 -0700 (PDT)
Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02)
	id AA13287; Thu, 12 Jun 1997 12:50:22 -0400
Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Thu, 12 Jun 1997 12:50 EDT
Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.5/8.7.3) with ESMTP id IAA19885; Thu, 12 Jun 1997 08:51:19 -0400 (EDT)
Received: (from rivers@localhost) by lakes.water.net (8.8.5/8.6.9) id IAA01024; Thu, 12 Jun 1997 08:59:10 -0400 (EDT)
Date: Thu, 12 Jun 1997 08:59:10 -0400 (EDT)
From: Thomas David Rivers <ponds!rivers@dg-rtp.dg.com>
Message-Id: <199706121259.IAA01024@lakes.water.net>
To: dfr@nlsystems.com, ponds!polstra.com!jdp
Subject: Re: kern/3771 NFS hangs when writing to local FS re-mounted via NFS
Cc: ponds!hub.freebsd.org!freebsd-bugs
Content-Type: text
Sender: owner-bugs@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

> 
> On Wed, 4 Jun 1997, John Polstra wrote:
> 
> > The following reply was made to PR kern/3771; it has been noted by GNATS.
> > 
> > From: John Polstra <jdp@polstra.com>
> > To: freebsd-gnats-submit@freebsd.org
> > Cc: Chris Timmons <skynyrd@opus.cts.cwu.edu>
> > Subject: Re: kern/3771 NFS hangs when writing to local FS re-mounted via NFS
> > Date: Wed, 04 Jun 1997 07:40:47 -0700
> > 
> >  > mount localhost:/ mnt
> >  > cp /mnt/tmp/junkfile /tmp/junkfile.copy         [this works ok
> >  > cp /tmp/junkfile /mnt/tmp/junkfile.dead         [this hangs the system]
> >  
> >  Chris Timmons has pointed out in private mail that this looks like
> >  it might be caused by a more general problem with the loopback
> >  interface that I and others have noticed.  Could you please try
> >  the same test again, but using your real host name instead of
> >  "localhost" for the mount?
> 
> There is a deadly embrace style deadlock when you loopback mount a
> filesystem using NFSv3.  I don't plan to try and fix this since I don't
> think loopback mounts are useful other than for trivial testing.
> 

 Ahh... but if you were doing something like "alex" - say to get
to RCS/CVS files via a locally mounted NFS (user-mode) server.... then,
this could be important.

 I just wanted to point out that the issue isn't as straight-forward
as simply reading local files, there may be some name -> file translation
going on in an NFS server...

	- Dave Rivers -