From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:42726) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ZKVRF-0006FJ-Pi for qemu-devel@nongnu.org; Wed, 29 Jul 2015 13:46:49 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1ZKVRA-0002Vh-K0 for qemu-devel@nongnu.org; Wed, 29 Jul 2015 13:46:45 -0400 Received: from mx1.redhat.com ([209.132.183.28]:33235) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ZKVRA-0002VM-DU for qemu-devel@nongnu.org; Wed, 29 Jul 2015 13:46:40 -0400 Message-ID: <55B9117E.2030006@redhat.com> Date: Wed, 29 Jul 2015 13:46:38 -0400 From: John Snow MIME-Version: 1.0 References: <20150729084653.GB10617@stefanha-thinkpad.redhat.com> In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] Using the one disk image file on 2 virtual machines at the same time List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Manjong Han , Stefan Hajnoczi Cc: qemu-devel@nongnu.org On 07/29/2015 01:29 PM, Manjong Han wrote: > Thanks, Stefan. > > 2015-07-29 17:46 GMT+09:00 Stefan Hajnoczi : >> >> You should probably use qcow2 backing files instead: >> >> 10G.qcow2 <-- vm001.qcow2 >> ^-- vm002.qcow2 >> >> The command to create these files is: >> >> qemu-img create -f qcow2 -o backing_file=10G.qcow2 vm001.qcow2. >> >> Both VMs share the data in 10G.qcow2. All writes go to vm001.qcow2 or >> vm002.qcow2, respectively, so they don't corrupt each other. >> > > I tried to create a backing files, using the commands which you told. > > $ qemu-img create -f qcow2 -o backing_file=10G.qcow2 vm001.qcow2 > $ qemu-img create -f qcow2 -o backing_file=10G.qcow2 vm002.qcow2 > > And, I used these backing files on each virtual machines. > But, new files weren't written on original disk image(10G.qcow2).. > The backing files were working each other. > >> Standard file systems (ext4, xfs) and volume managers (LVM) are not >> cluster-aware by default. They must only be accessed from one machine >> at a time. Otherwise you risk data corruption. >> > > I think that I must probably use a shared file system like NFS.. > Yes, any files written using the backing files like outlined above will put new files in the overlays (e.g. vm001.qcow2 or vm002.qcow2) and NOT into the backing file (10G.qcow2) this is a safe way to share a base image for an OS, but it's not a method of accomplishing a concurrent fileshare. You'll want to configure an NFS or CIFS share (etc) in the base image and then allow the multiple VMs to utilize that share. --js