From owner-freebsd-chat@FreeBSD.ORG Wed Jul 23 17:20:14 2003 Return-Path: Delivered-To: freebsd-chat@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 3157637B401 for ; Wed, 23 Jul 2003 17:20:14 -0700 (PDT) Received: from HAL9000.homeunix.com (ip114.bella-vista.sfo.interquest.net [66.199.86.114]) by mx1.FreeBSD.org (Postfix) with ESMTP id 8FFB643FBF for ; Wed, 23 Jul 2003 17:20:13 -0700 (PDT) (envelope-from das@FreeBSD.ORG) Received: from HAL9000.homeunix.com (localhost [127.0.0.1]) by HAL9000.homeunix.com (8.12.9/8.12.9) with ESMTP id h6O0KBhC016428; Wed, 23 Jul 2003 17:20:11 -0700 (PDT) (envelope-from das@FreeBSD.ORG) Received: (from das@localhost) by HAL9000.homeunix.com (8.12.9/8.12.9/Submit) id h6O0K9WQ016427; Wed, 23 Jul 2003 17:20:09 -0700 (PDT) (envelope-from das@FreeBSD.ORG) Date: Wed, 23 Jul 2003 17:20:09 -0700 From: David Schultz To: Brad Knowles Message-ID: <20030724002009.GA16322@HAL9000.homeunix.com> Mail-Followup-To: Brad Knowles , Attila Nagy , FreeBSD Chat Mailing List References: <3F1E6456.9090400@fsn.hu> <20030723173242.GC14408@HAL9000.homeunix.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: cc: Attila Nagy cc: FreeBSD Chat Mailing List Subject: Re: maildir with softupdates X-BeenThere: freebsd-chat@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Non technical items related to the community List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 24 Jul 2003 00:20:14 -0000 On Wed, Jul 23, 2003, Brad Knowles wrote: > > The statement is FUD; this is a topic that mailer people love to > > complain about. It's only true if your MTA doesn't call fsync() > > when it wants to guarantee that the file it just wrote is on > > stable storage. > > The MTA does not know anything about maildir. This would be a > local delivery agent (LDA) issue, not an MTA issue. Brain fart. Thanks for the correction. > Moreover, the software not only needs to issue an fsync() on the > file, it also needs to issue an fsync() on the directory, in order to > have reasonable guarantees that the date has been safely written. Softupdates automatically syncs the parent directory (and its parent, etc) when you fsync() the file. The bug that I mentioned ext3 had was that it didn't do this. Actually, it may be too harsh to call it a bug, given that POSIX has rather lax requirements regarding what fsync() is required to do, but most applications I've looked at in regards to this issue assume the behavior that softupdates has.