From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: linux-nfs-owner@vger.kernel.org
Received: from fieldses.org ([174.143.236.118]:38609 "EHLO fieldses.org"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1751835Ab2AZUSW (ORCPT <rfc822;linux-nfs@vger.kernel.org>);
	Thu, 26 Jan 2012 15:18:22 -0500
Date: Thu, 26 Jan 2012 15:18:21 -0500
From: "J. Bruce Fields" <bfields@fieldses.org>
To: Jeff Layton <jlayton@redhat.com>
Cc: Steve Dickson <SteveD@redhat.com>, linux-nfs@vger.kernel.org
Subject: Re: [PATCH v4 09/11] nfsdcld: reopen pipe if it's deleted and
 recreated
Message-ID: <20120126201821.GC31531@fieldses.org>
References: <20120125152814.391ae064@tlielax.poochiereds.net>
 <4F207C7C.4090207@RedHat.com>
 <20120125183206.5efcd845@corrin.poochiereds.net>
 <4F214B77.8050300@RedHat.com>
 <20120126082830.23324255@corrin.poochiereds.net>
 <20120126093059.5c732d75@corrin.poochiereds.net>
 <4F2171B5.2030103@RedHat.com>
 <20120126104157.671a10c2@corrin.poochiereds.net>
 <20120126185859.GA31531@fieldses.org>
 <20120126143659.74a9de30@tlielax.poochiereds.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
In-Reply-To: <20120126143659.74a9de30@tlielax.poochiereds.net>
Sender: linux-nfs-owner@vger.kernel.org
List-ID: <linux-nfs.vger.kernel.org>

On Thu, Jan 26, 2012 at 02:36:59PM -0500, Jeff Layton wrote:
> On Thu, 26 Jan 2012 13:58:59 -0500
> "J. Bruce Fields" <bfields@fieldses.org> wrote:
> 
> > On Thu, Jan 26, 2012 at 10:41:57AM -0500, Jeff Layton wrote:
> > > Correct. When knfsd starts back up, a new pipe will be created. At that
> > > point, the daemon will try to open the new one and will then close the
> > > old if the open succeeds. It will only loop here if that open fails for
> > > some reason.
> > 
> > I'm a little confused--how could this happen?
> > 
> 
> Bugs?

If that's really the only case then "crash and log an error" might
really be more robust in the long term than spinning.

--b.