From owner-freebsd-hackers Sat Feb 14 15:36:12 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id PAA11336 for freebsd-hackers-outgoing; Sat, 14 Feb 1998 15:36:12 -0800 (PST) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from allegro.lemis.com (allegro.lemis.com [192.109.197.134]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id PAA10909 for ; Sat, 14 Feb 1998 15:33:51 -0800 (PST) (envelope-from grog@lemis.com) Received: from freebie.lemis.com (freebie.lemis.com [192.109.197.137]) by allegro.lemis.com (8.8.7/8.8.5) with ESMTP id KAA24296; Sun, 15 Feb 1998 10:03:31 +1030 (CST) Received: (from grog@localhost) by freebie.lemis.com (8.8.8/8.8.7) id KAA08162; Sun, 15 Feb 1998 10:03:31 +1030 (CST) (envelope-from grog) Message-ID: <19980215100330.12416@freebie.lemis.com> Date: Sun, 15 Feb 1998 10:03:30 +1030 From: Greg Lehey To: Capriotti , hackers@FreeBSD.ORG Subject: Disk mirroring (was: SFT) References: <3.0.32.19691231210000.00a226f0@pop.mpc.com.br> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89i In-Reply-To: <3.0.32.19691231210000.00a226f0@pop.mpc.com.br>; from Capriotti on Sat, Feb 14, 1998 at 07:52:33PM -0300 WWW-Home-Page: http://www.lemis.com/~grog Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia Phone: +61-8-8388-8286 Fax: +61-8-8388-8725 Mobile: +61-41-739-7062 Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Sat, 14 February 1998 at 19:52:33 -0300, Capriotti wrote: > At 11:33 AM 1/6/98 -0800, you wrote: >> >> On Tue, 6 Jan 1998, Capriotti wrote: >> >>> Does anyone have any news about System Fault Tolerance under Free ? >>> >>> Like what Novell has, from mirrowed disks to mirrowed servers ? >> >> Mirrored disks can be done with ccd. >> >> Mirrored servers are basically what Unix types call a cluster. On >> Novell this is easy, because Novell boxes are basically just file servers, >> but a Unix box could be doing many different things. Migrating tasks >> from the failed system to the working system, and assumption >> of the IP traffic, is difficult. > > Hey, boys, just one question that crossed my mind: > > Why can't the kernel do something like: > > (consider a system w/ mirrowed disks using ccd) > > if (main disk) fails then start using (mirror disk) > > What - technically - prevents ccd/kernel/something else from being able to > do it ? Nothing. You've presumably missed some discussion on the matter in the -fs list. Unfortunately, CCD does not recover at all gracefully from this kind of situation. I'm currently writing a replacement for ccd which will do this (sort of), amongst other things. Why sort of? Because you wouldn't want to use one disk exclusively if you have two. You'll share read accesses as evenly as possible across the available spindles, and of course you must write to all of them. Greg To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message