From owner-freebsd-stable@FreeBSD.ORG Wed Apr 2 16:27:38 2008 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id B43511065670 for ; Wed, 2 Apr 2008 16:27:38 +0000 (UTC) (envelope-from jdc@parodius.com) Received: from mx01.sc1.parodius.com (mx01.sc1.parodius.com [72.20.106.3]) by mx1.freebsd.org (Postfix) with ESMTP id A2BD58FC36 for ; Wed, 2 Apr 2008 16:27:38 +0000 (UTC) (envelope-from jdc@parodius.com) Received: by mx01.sc1.parodius.com (Postfix, from userid 1000) id 7BDE21CC060; Wed, 2 Apr 2008 09:27:38 -0700 (PDT) Date: Wed, 2 Apr 2008 09:27:38 -0700 From: Jeremy Chadwick To: Harald Schmalzbauer Message-ID: <20080402162738.GA54106@eos.sc1.parodius.com> References: <47F39F8C.1000004@omnisec.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <47F39F8C.1000004@omnisec.de> User-Agent: Mutt/1.5.17 (2007-11-01) Cc: freebsd-stable@freebsd.org Subject: Re: ataraid and glable interference X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 02 Apr 2008 16:27:38 -0000 On Wed, Apr 02, 2008 at 05:00:28PM +0200, Harald Schmalzbauer wrote: > usually I use gmirror but today I wanted to use ataraid (Intel Matrix) for > a new system. > My filesystems are always mounted by label reference, > "/dev/ufs/testbox1ROOT /" for example. > > The problem is that the system also sees the disks themself which are > components of the ataraid (ar0 = ad4+a6 and all 3 devices are present). > If I 'tunefs -L testbox1ROOT /dev/mirror/t1ROOT' then at startup geom sees > the "testbox1ROOT" label on ad4p2 and not on mirror/t1ROOT. > > Is there a way to hide the real disks if they are members of an ataraid? This sounds like something you should file a PR for. I'm a little baffled as to how/why the glabel is showing up at all on /dev/ad4 partitions (or slices; not sure which GEOM sets metadata on). What you're doing sounds completely legitimate, and it sounds like an oversight which should be addressed in some way. The only way I can think of is to get ata(4) to support some loader(8) tunables, something like hint.ad.4.disabled="1" to disable ad4, so that you can use those to "ignore" disks. Or is this a GEOM labelling bug of some sort? -- | Jeremy Chadwick jdc at parodius.com | | Parodius Networking http://www.parodius.com/ | | UNIX Systems Administrator Mountain View, CA, USA | | Making life hard for others since 1977. PGP: 4BD6C0CB |