From owner-freebsd-questions Sun Oct 29 20:47: 3 2000 Delivered-To: freebsd-questions@freebsd.org Received: from mailhost01.reflexnet.net (mailhost01.reflexnet.net [64.6.192.82]) by hub.freebsd.org (Postfix) with ESMTP id 428D837B479 for ; Sun, 29 Oct 2000 20:47:00 -0800 (PST) Received: from 149.211.6.64.reflexcom.com ([64.6.211.149]) by mailhost01.reflexnet.net with Microsoft SMTPSVC(5.5.1877.197.19); Sun, 29 Oct 2000 20:45:38 -0800 Received: (from cjc@localhost) by 149.211.6.64.reflexcom.com (8.11.0/8.11.0) id e9U4kcx78002; Sun, 29 Oct 2000 20:46:38 -0800 (PST) (envelope-from cjc) Date: Sun, 29 Oct 2000 20:46:38 -0800 From: "Crist J . Clark" To: Arun Sharma Cc: freebsd-questions@FreeBSD.ORG Subject: Re: softupdates and (avoiding) fsck Message-ID: <20001029204638.A75251@149.211.6.64.reflexcom.com> Reply-To: cjclark@alum.mit.edu References: <20001029183443.A27388@sharmas.dhs.org> <20001029190021.Z75251@149.211.6.64.reflexcom.com> <20001029191544.A27575@sharmas.dhs.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0i In-Reply-To: <20001029191544.A27575@sharmas.dhs.org>; from adsharma@sharmas.dhs.org on Sun, Oct 29, 2000 at 07:15:44PM -0800 Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Sun, Oct 29, 2000 at 07:15:44PM -0800, Arun Sharma wrote: > On Sun, Oct 29, 2000 at 07:00:21PM -0800, Crist J . Clark wrote: > > There is no way around fsck'ing when the system is brought down > > hard. (Not absolutely true, but you dowanna mount an unclean FS.) > > Soft updates does _NOT_ guarantee that you will come out of such a > > nasty thing without damage. It does make it much less likely, > > however. Carefully read, > > > > /usr/src/sys/ufs/ffs/README > > > > For more information on what soft updates really does. > > That file doesn't exist in -current anymore. But I found some other > file, which pointed me to: > > http://www.mckusick.com/softdep/ > > which says: > > > By ensuring that the only inconsistencies are unclaimed blocks or > > inodes, soft updates can eliminate the need to run a filesystem check > > program after every system crash. Instead, the system is brought up > > immediately. When it is convenient, a snapshot is taken and a background > > task can be run on on that snapshot to reclaim any lost blocks and > > inodes. The use of a snapshot allows normal filesystem activity to > > continue concurrently. Good you found the paper. Section 6 deals with the issue. Also, have a look at the fsck(8) source to see how it handles checking softupdate filesystems. -- Crist J. Clark cjclark@alum.mit.edu To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message