From owner-freebsd-fs@FreeBSD.ORG Mon May 17 10:56:17 2004 Return-Path: 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 C6B6B16A4CE for ; Mon, 17 May 2004 10:56:17 -0700 (PDT) Received: from thebsh.namesys.com (thebsh.namesys.com [212.16.7.65]) by mx1.FreeBSD.org (Postfix) with SMTP id 809AA43D2F for ; Mon, 17 May 2004 10:56:16 -0700 (PDT) (envelope-from Nikita@Namesys.COM) Received: (qmail 24273 invoked from network); 17 May 2004 17:56:15 -0000 Received: from laputa.namesys.com (212.16.7.124) by thebsh.namesys.com with SMTP; 17 May 2004 17:56:15 -0000 Received: by laputa.namesys.com (Postfix on SuSE Linux 8.0 (i386), from userid 511) id 9942F12E68; Mon, 17 May 2004 21:56:09 +0400 (MSD) From: Nikita Danilov MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <16552.64697.572176.262372@laputa.namesys.com> Date: Mon, 17 May 2004 21:56:09 +0400 To: Xin LI In-Reply-To: <20040517174836.GA983@frontfree.net> References: <200405171318.15200@misha-mx.virtual-estates.net> <20040517174836.GA983@frontfree.net> X-Mailer: VM 7.17 under 21.5 (patch 17) "chayote" (+CVS-20040321) XEmacs Lucid cc: freebsd-fs@freebsd.org cc: Mikhail Teterin Subject: Re: QMail and SoftUpdates X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 17 May 2004 17:56:17 -0000 Xin LI writes: > On Mon, May 17, 2004 at 01:18:15PM -0400, Mikhail Teterin wrote: > > The link at > > > > http://cr.yp.to/qmail/faq/reliability.html#filesystems > > > > claims, using SoftUpdates for mailqueue is dangerous. Is that still > > true? Thanks! > > Yes, it is dangerous. Same is true for any journalling file systems, > which essentially does the same thing: delayed write of data/metadata. > > Delayed write will make it possible for the Operating System to group > several writes together and write them once, or at least, in a better > order in order to improve performance. However, for the mail case, once > it responds "250", then the remote peer is allowed to remove the message > from its queue. If the system crashes, and the data was not written into > disk, then your message is lost. Unless mail-server did fsync(2) which is guaranteed to return only after data reached stable storage. If file-system doesn't provide such guarantee it's broken, if mail server doesn't call fsync, or fdatasync---it is. Even without any journalling involved. Nikita.