From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from [140.186.70.92] (port=45971 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1OvuRg-0001zp-Ve for qemu-devel@nongnu.org; Wed, 15 Sep 2010 12:02:58 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69) (envelope-from ) id 1OvuRb-0003or-VD for qemu-devel@nongnu.org; Wed, 15 Sep 2010 12:02:52 -0400 Received: from mx1.redhat.com ([209.132.183.28]:30925) by eggs.gnu.org with esmtp (Exim 4.69) (envelope-from ) id 1OvuRb-0003ol-P1 for qemu-devel@nongnu.org; Wed, 15 Sep 2010 12:02:47 -0400 From: Juan Quintela In-Reply-To: <4C8D1339.2000803@redhat.com> (Avi Kivity's message of "Sun, 12 Sep 2010 19:51:53 +0200") References: <1284213896-12705-1-git-send-email-aliguori@us.ibm.com> <1284213896-12705-2-git-send-email-aliguori@us.ibm.com> <4C8CAD7C.5020102@redhat.com> <4C8CD06A.7060507@codemonkey.ws> <4C8CD57D.60508@redhat.com> <4C8CF122.6080000@codemonkey.ws> <4C8CFA6C.6080202@redhat.com> <4C8D0999.9040405@codemonkey.ws> <4C8D1339.2000803@redhat.com> Date: Wed, 15 Sep 2010 18:00:07 +0200 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Subject: [Qemu-devel] Re: [PATCH 1/3] block: allow migration to work with image files List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Avi Kivity Cc: Kevin Wolf , qemu-devel@nongnu.org, Stefan Hajnoczi Avi Kivity wrote: > On 09/12/2010 07:10 PM, Anthony Liguori wrote: >> On 09/12/2010 11:06 AM, Avi Kivity wrote: >>> On 09/12/2010 05:26 PM, Anthony Liguori wrote: >>>> On 09/12/2010 08:28 AM, Avi Kivity wrote: >>>>> On 09/12/2010 03:06 PM, Anthony Liguori wrote: >>>>>> >>>>>> Backing files and logical size shouldn't change during live >>>>>> migration. >>>>> >>>>> Why not? >>>> >>>> To make our lives easier. >>> >>> It means management needs to block volume resize while a live >>> migration takes place. Since live migration is typically done by >>> the system automatically, while volume resize happens in response >>> to user request, this isn't a good idea. Both in terms of user >>> experience, and in terms of pushing more complexity to management. >> >> We don't do volume resize today so it's a moot point. >> > > Let's be prepared for the future. This is better done when we would be able to create the machine during migration. As we have everything organized today, this is basically imposible :( The problem is not related with live migration, is related with the fact that we need to give the same arguments in both source & target machines. Later, Juan.