From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: with ECARTIS (v1.0.0; list xfs); Mon, 08 Jan 2007 07:59:48 -0800 (PST) Received: from sumo.dreamhost.com (sumo.dreamhost.com [66.33.216.29]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id l08Fxgqw004776 for ; Mon, 8 Jan 2007 07:59:43 -0800 Received: from spaceymail-a1.dreamhost.com (sd-green-bigip-62.dreamhost.com [208.97.132.62]) by sumo.dreamhost.com (Postfix) with ESMTP id 1584C17E75A for ; Mon, 8 Jan 2007 07:35:38 -0800 (PST) Received: from jupiter.solar.net (cpe-24-27-90-21.houston.res.rr.com [24.27.90.21]) by spaceymail-a1.dreamhost.com (Postfix) with ESMTP id 308AC194F6C for ; Mon, 8 Jan 2007 07:35:35 -0800 (PST) From: Joe Bacom Reply-To: joe@docsimple.com Subject: Re: What's wrong with XFS? Date: Mon, 8 Jan 2007 09:35:36 -0600 References: <936386.57179.qm@web59111.mail.re1.yahoo.com> <1168267755.29690.13.camel@venus.local.navi.pl> In-Reply-To: <1168267755.29690.13.camel@venus.local.navi.pl> MIME-Version: 1.0 Content-Type: multipart/signed; boundary="nextPart1755204.IqRS27UC8K"; protocol="application/pgp-signature"; micalg=pgp-sha1 Content-Transfer-Encoding: 7bit Message-Id: <200701080935.36736.joe@docsimple.com> Sender: xfs-bounce@oss.sgi.com Errors-to: xfs-bounce@oss.sgi.com List-Id: xfs To: xfs@oss.sgi.com --nextPart1755204.IqRS27UC8K Content-Type: text/plain; charset="iso-8859-6" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline The solution to Dave's problem seems obvious to me. If you care about your= =20 data and your hardware, buy a UPS with power conditioning, configure Linux = to=20 show down when the battery gets low and enjoy the peace of mind knowing tha= t=20 even if your away from your machine and the power goes off, the system will= =20 take care of itself. Joe On Monday 08 January 2007 08:49, you wrote: > On Mon, 2007-01-08 at 05:13 -0800, Dave N wrote: > > Hi, > > > > Can someone enlighten me what the issue is with XFS? I've been hearing a > > lot of good things on the Net about XFS. How it's lightening fast, how = it > > has features other file systems do not have (like GRIO, real time > > volumes, allocate on flush, etc), how it scales very well, etc... but > > what I didn't hear about is how fast XFS screws things up if something > > wrong happens. Because of the good things I heard about XFS, I too > > decided to try it out (been using Ext3 or ReiserFS here for most of the > > time). Now I'm very disappointed in XFS. I live in an area where power > > outages are common and I do not have an UPS here. I have a few computers > > all running on XFS and thought that XFS will give me similar > > data-integrity like Ext3 or ReiserFS. Now, for the past few weeks I've > > been experiencing "strange behavior" from XFS. One time, I was reading = an > > article on the Net and had only my Firefox browser open. Then we had a > > power outage for a short period of time, and when I logged in again into > > KDE, I was surprised to find out that all my desktop icons were messed = up > > all over the place. The other time, again power outage, only this time I > > was working on a small text file. Booted up again only to find out that > > the file I was working on contained garbage and I had to start all over > > again. > > > > I also heard that XFS depends heavily on the application side for its > > data-integrity. XFS "thinks" that the application will use the proper > > calls when writing to disk. What???? How is it the task of the > > application to ensure the safety of your files??? IMO, programs are the= re > > to provide the tools to be productive, NOT to ensure the data safety of > > your files, that's the task of the file system. Even MySQL provides me > > with better data-integrity here. If I'm doing some database transaction > > and the power fails, I can be pretty sure that *most* of the time, MySQL > > will be just fine next time I boot up. > > > > Why oh why such a beautiful file system like XFS is so terrible at > > data-integrity? Look what Sun Microsystems did with their new ZFS file > > system... full atomicity, CRC checksumming and other features to ensure > > data-integrity... why can't XFS have such things? > > > > Thanks for listening to my preaching here guys > > > > Cheers! > > Hi, > > It is nothing wrong with XFS - your expectations are wrong. > > You expect data to be journaled, but XFS does journal metadata only, not > data. So, the thing that you get is filesystem integrity not data > integrity. > If you want data integrity you need properly written applications and > __it is__ application's job to care about it's data. It is nothing > unusual here. > > If you need data journaling then you need another filesystem - eg. ext3. > > I suppose that you find all of it in FAQ. > > Regards, > > Olaf --=20 A Cringester who requested anonymity says when a friend ran=20 Microsoft BS (Baseline Security) Analyzer on a XP Pro SP2 machine, the cumulative size of the patches that were required exceeded=20 the size of the original OS. I'm not surprised. The volume of Microsoft BS I've analyzed could fill Bill Gates' house. Source: Robert X. Cringely, InfoWorld, Sept. 4, 2006, Issue 36 Penguin: Linux version 2.6.16, 8010.09 BogoMips --nextPart1755204.IqRS27UC8K Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (GNU/Linux) iD8DBQBFomTILtK/s7Pra28RAr03AKCELLtTC2Nt5ZM5KDQDZsW7cQQBFwCeLn7Q 6hKiX3Z67OVHCmIcM5vAD8w= =Is2P -----END PGP SIGNATURE----- --nextPart1755204.IqRS27UC8K--