From owner-freebsd-fs@FreeBSD.ORG  Thu Sep 25 05:20:06 2003
Return-Path: <owner-freebsd-fs@FreeBSD.ORG>
Delivered-To: freebsd-fs@freebsd.org
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP id 2863516A4B3
	for <fs@freebsd.org>; Thu, 25 Sep 2003 05:20:06 -0700 (PDT)
Received: from muse.csie.ntu.edu.tw (muse.csie.ntu.edu.tw [140.112.28.158])
	by mx1.FreeBSD.org (Postfix) with ESMTP id 80D1F4400E
	for <fs@freebsd.org>; Thu, 25 Sep 2003 05:20:05 -0700 (PDT)
	(envelope-from rafan@muse.csie.ntu.edu.tw)
Received: by muse.csie.ntu.edu.tw (Postfix, from userid 1001)
	id B867452EFF; Thu, 25 Sep 2003 20:20:04 +0800 (CST)
From: Rong-en Fan <rafan@infor.org>
To: bde@zeta.org.au (Bruce Evans)
In-Reply-To: <bkt80i$jou$1@FreeBSD.csie.NCTU.edu.tw>
X-Newsgroups: mailing.freebsd.fs
User-Agent: tin/1.6.1-20030810 ("Mingulay") (UNIX) (FreeBSD/4.8-RELEASE-p4
	(i386))
Message-Id: <20030925122004.B867452EFF@muse.csie.ntu.edu.tw>
Date: Thu, 25 Sep 2003 20:20:04 +0800 (CST)
cc: fs@freebsd.org
Subject: Re: NFS incorrectly update atime
X-BeenThere: freebsd-fs@freebsd.org
X-Mailman-Version: 2.1.1
Precedence: list
List-Id: Filesystems <freebsd-fs.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-fs>,
	<mailto:freebsd-fs-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-fs>
List-Post: <mailto:freebsd-fs@freebsd.org>
List-Help: <mailto:freebsd-fs-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-fs>,
	<mailto:freebsd-fs-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Thu, 25 Sep 2003 12:20:06 -0000

In article <bkt80i$jou$1@FreeBSD.csie.NCTU.edu.tw> you wrote:
>
>>   it only occur when file is not empty, also i've noticed that
> 
> I think the client just does a read as part of a read-modify-write.
> It works with blocks, so it must do a read before write in all cases where
> the write is not on a block boundary, and at least FreeBSD ffs servers
> have no way to avoid updating the atime on read short of a global mount
> -noatime.  Some of the reads, including ones from small test files in your
> example, may be from a client cache, so the atime update might be avoided
> accidentally, but a read will always be necessay if the cache is cold.

My application is NFS mail spool, so atime is important to me. 
I've tested it on Linux client, it's atime not updated in this case.
So, I don't know if we can fix this or any workaround?