From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from list by lists.gnu.org with archive (Exim 4.71) id 1WrUEz-0000XK-1X for mharc-grub-devel@gnu.org; Mon, 02 Jun 2014 11:33:37 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:54158) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WrUEq-0000MO-W1 for grub-devel@gnu.org; Mon, 02 Jun 2014 11:33:34 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1WrUEl-0004il-CU for grub-devel@gnu.org; Mon, 02 Jun 2014 11:33:28 -0400 Received: from mail-lb0-x236.google.com ([2a00:1450:4010:c04::236]:41629) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WrUEl-0004iN-3w for grub-devel@gnu.org; Mon, 02 Jun 2014 11:33:23 -0400 Received: by mail-lb0-f182.google.com with SMTP id z11so2689478lbi.27 for ; Mon, 02 Jun 2014 08:33:22 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=date:from:to:cc:subject:message-id:in-reply-to:references :mime-version:content-type:content-transfer-encoding; bh=SqZxChEXHshPrRLDntaKdKNh8RKo9YEKterNUnu14TI=; b=yi0TeOhGaDBsLyhp8W5Vwa+Uf4jNIcGVoU/CTxqA4epmKxaubljWaCweBepDjb66Jy Wwl95HpL88OYWi4f4NWEQKRKB8hZ8HOWi9+YyTv4t+zppn01NQMZU44heMX4N8A2koME 6bX+ISOlzk+J9cRLc4oyR0qWTdJQS0x0WS7WqU4ZQH14QDhhAjkmpUgXz/IvBeKnyLrK hMNXYuhRCjpEN81teVyL2N4hmvtgFYMjG4I5fLO7jEAXMlN/ooJtBAkUPzLAMLXfulm1 3wUYCk8oFlMqEY5xgC6W/xv9PMxMpWD3M+cAW2n6F7YrhKgkVjXV658Hpc4++cVtb49N nt1w== X-Received: by 10.112.185.105 with SMTP id fb9mr2492223lbc.72.1401723201979; Mon, 02 Jun 2014 08:33:21 -0700 (PDT) Received: from opensuse.site (ppp37-190-15-130.pppoe.spdop.ru. [37.190.15.130]) by mx.google.com with ESMTPSA id zn1sm13194034lbb.17.2014.06.02.08.33.20 for (version=SSLv3 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Mon, 02 Jun 2014 08:33:21 -0700 (PDT) Date: Mon, 2 Jun 2014 19:33:20 +0400 From: Andrey Borzenkov To: The development of GNU GRUB Subject: Re: Why ignore UUID when linux root device is an LV? Message-ID: <20140602193320.43da908a@opensuse.site> In-Reply-To: <20140502142330.GC7028@riva.ucam.org> References: <53639575.6020109@totakura.in> <20140502142330.GC7028@riva.ucam.org> X-Mailer: Claws Mail 3.9.2 (GTK+ 2.24.22; x86_64-suse-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=KOI8-R Content-Transfer-Encoding: 8bit X-detected-operating-system: by eggs.gnu.org: Error: Malformed IPv6 address (bad octet value). X-Received-From: 2a00:1450:4010:c04::236 Cc: Sree Harsha Totakura , cjwatson@ubuntu.com X-BeenThere: grub-devel@gnu.org X-Mailman-Version: 2.1.14 Precedence: list Reply-To: The development of GNU GRUB List-Id: The development of GNU GRUB List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 02 Jun 2014 15:33:34 -0000 В Fri, 2 May 2014 15:23:30 +0100 Colin Watson пишет: > On Fri, May 02, 2014 at 02:54:13PM +0200, Sree Harsha Totakura wrote: > > I would like to know why UUID is not used for root device in Linux > > kernel's `root=' parameter when the root device uses LVM. > > Filesystem UUIDs on LVM aren't necessarily unique in the presence of > snapshots; and not all initramfses cope with root=UUID=* in any event, > for instance: > > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=612402 > > Every time this breaks in GRUB and we end up using UUIDs on LVM again, I > get a slew of Debian bug reports until I figure it out and fix it. > Note Debian Bug#747653 which results exactly from this behavior - btrfs which spans multiple LVM volumes; grub hits LVM and does not use UUID which results in passing multiple devices as root= argument. > The LVM device names themselves are sufficiently stable that there > should be no need to introduce additional layers of complexity by using > filesystem UUIDs. In general we only need them in cases where the > device names are not stable. > > > >From the commit log: > > > commit 507736c87c49712ac618169d17a659bd6c25eecc > > > Author: Colin Watson > > > Date: Fri Jul 2 12:32:05 2010 +0100 > > > > > > * util/grub-mkconfig_lib.in (uses_abstraction): New function. > > > * util/grub.d/10_linux.in: Use it to check for LVM, so that > > > LVM-on-RAID is handled correctly. > > > > I see that this condition is added to address LVM/RAID configurations. > > But, I guess it is similar to LVM/LUKS; is it not? > > I'm not sure, but perhaps GRUB_ENABLE_CRYPTODISK=y would help > grub-mkconfig notice this? > > Regards, >