From owner-freebsd-fs@FreeBSD.ORG Sun Jan 29 17:51:20 2012 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id B9488106566B for ; Sun, 29 Jan 2012 17:51:20 +0000 (UTC) (envelope-from jhellenthal@gmail.com) Received: from mail-gy0-f182.google.com (mail-gy0-f182.google.com [209.85.160.182]) by mx1.freebsd.org (Postfix) with ESMTP id 6E7EF8FC0C for ; Sun, 29 Jan 2012 17:51:20 +0000 (UTC) Received: by ghbg15 with SMTP id g15so1596179ghb.13 for ; Sun, 29 Jan 2012 09:51:19 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=sender:date:from:to:cc:subject:message-id:references:mime-version :content-type:content-disposition:in-reply-to; bh=zywtK9h9++WYyKSThtaP6fHsTnEv+tsXkVWvpCYYZ4U=; b=jGKi2argFoGEkMyq5vP5080a2RM8MOe/AuDmg92xwN7CWFOfmePEq2DVoZ5EMTfTt4 lkN7f/dYGzzFaE+V2ll8ImcWhY3/hmYwV/7OsZiTpwkqwf7Ys510VlwE9vAqcsDY1hfu M33RCBcUXmuqCeXcn78sWCmlphvqyhSEXGquE= Received: by 10.236.131.38 with SMTP id l26mr22054068yhi.70.1327859479646; Sun, 29 Jan 2012 09:51:19 -0800 (PST) Received: from DataIX.net (adsl-99-19-42-1.dsl.klmzmi.sbcglobal.net. [99.19.42.1]) by mx.google.com with ESMTPS id j41sm11825824yhm.16.2012.01.29.09.51.17 (version=TLSv1/SSLv3 cipher=OTHER); Sun, 29 Jan 2012 09:51:18 -0800 (PST) Sender: Jason Hellenthal Received: from DataIX.net (localhost [127.0.0.1]) by DataIX.net (8.14.5/8.14.5) with ESMTP id q0THpDKr043614 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Sun, 29 Jan 2012 12:51:13 -0500 (EST) (envelope-from jhell@DataIX.net) Received: (from jhell@localhost) by DataIX.net (8.14.5/8.14.5/Submit) id q0THp6tQ043494; Sun, 29 Jan 2012 12:51:06 -0500 (EST) (envelope-from jhell@DataIX.net) Date: Sun, 29 Jan 2012 12:51:06 -0500 From: Jason Hellenthal To: Randy Bush Message-ID: <20120129175106.GA86040@DataIX.net> References: <20120129085045.GA26210@DataIX.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Cc: FreeBSD FS Subject: Re: trying to whack a glabel for a zfs mirror X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 29 Jan 2012 17:51:20 -0000 On Sun, Jan 29, 2012 at 06:00:25PM +0900, Randy Bush wrote: > > Once again the use of glabel(8) that causes and can cause loss of data > > within ZFS disks... DO NOT USE GLABEL! it is not a solution that you > > are looking for and in the long run you will shoot yourself in the > > foot for using it. > > > > What you are seeing is glabel blatently refusing to write meta-data to > > parts of the disk where something may already exist. This is a good > > thing. > > > > In turn use something like gpart(8) to adjust the gpt label and/or set > > your disks up properly. This is not the same thing as glabel(8) which > > in turn is a hack and not a solution and severely needs to be shot > > into outerspace from world. > > i gather you do not like glabel :) > The idea is good but the end result usually turns out to be what you are seeing. > as you might have guessed from the ad0s3, the disks are gparted. the > reason that they are also glabeled is that this is on a disk controller > from hell, an hpt 16-port, which seems to occasionally renumber the > drives, for example when one is removed. so i wanted constant labels. > Jeremy's solution is quite sensible, but seeing you have these disks gpart(1)'d not neccesarily meaning they are using GPT but if they were then you would be able to use /dev/gpt/ that would be statically available until you used gpart(8) to remove it. gpart create -s GPT ad0 gpart add -t freebsd-zfs -l zfs-disk1 ad0 and then add /dev/gpt/zfs-disk1 to your pool. This is different than creating a MBR partitioning scheme that leaves you to using glabel(8) to store meta-data within possibly used space and the "It is not compatible with other systems shoot yourself in the foot.". glabel(8) should be for thumbdrives only! I see you are using the original ATA driver still. You might want to also check into ATA_CAM option for the kernel. It might break away part of your headaches with all these suggestions. Jeremy's suggesting is explicitly hinting at using that as well. "hint.ada." -- ;s =;