From owner-freebsd-geom@FreeBSD.ORG Mon Jul 31 10:22:43 2006 Return-Path: X-Original-To: freebsd-geom@freebsd.org Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id BBB1516A4DD for ; Mon, 31 Jul 2006 10:22:43 +0000 (UTC) (envelope-from arne_woerner@yahoo.com) Received: from web30305.mail.mud.yahoo.com (web30305.mail.mud.yahoo.com [68.142.200.98]) by mx1.FreeBSD.org (Postfix) with SMTP id 4074843D45 for ; Mon, 31 Jul 2006 10:22:43 +0000 (GMT) (envelope-from arne_woerner@yahoo.com) Received: (qmail 41082 invoked by uid 60001); 31 Jul 2006 10:22:42 -0000 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=Message-ID:Received:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=30gcNfcCcWkhhIQhWMHEzhG7NLXNtzyF15ZLLElLXhdF8oqA7d3cMAcMwhY+AZtrmx75o+Lv04lEzLQK4ojd+AOED6YgcMn3fvVspKMUcqmYCzwYsLxcLXzEm1BxQhGw04+mRTCzuyoZdlwSdYx5duyXhu9Grh2h3SFL5pSpJAE= ; Message-ID: <20060731102242.41080.qmail@web30305.mail.mud.yahoo.com> Received: from [213.54.65.162] by web30305.mail.mud.yahoo.com via HTTP; Mon, 31 Jul 2006 03:22:42 PDT Date: Mon, 31 Jul 2006 03:22:42 -0700 (PDT) From: "R. B. Riddick" To: Pawel Jakub Dawidek In-Reply-To: <20060731101033.GA842@garage.freebsd.pl> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Cc: Brent Hostetler , freebsd-geom@freebsd.org Subject: Re: gmirror/gconcat: mkdir causes system reboot X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 31 Jul 2006 10:22:43 -0000 --- Pawel Jakub Dawidek wrote: > Gmirror protects against such situations - on power loss or system crash > it will resynchronize mirrors. > Hmm... The evil is a squirrel... E. g. the write cache could delay the commit regarding the last sector of the disc until the data has been written... Then the sudden power loss (or so) could happen, while one disc updates the data but before the other disc and the meta data are written. Or not? Or is gmirror like gjournal? In graid5 I mark the geom_device as clean as soon as the shutdown event occurs. Is that a good idea? That makes versioning of the various disks obsolete, doesn't it? -Arne __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com