From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:35582) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WEKO3-0005IK-VN for qemu-devel@nongnu.org; Fri, 14 Feb 2014 10:09:11 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1WEKO3-00004z-70 for qemu-devel@nongnu.org; Fri, 14 Feb 2014 10:09:07 -0500 Date: Fri, 14 Feb 2014 16:09:02 +0100 From: Stefan Hajnoczi Message-ID: <20140214150902.GN17391@stefanha-thinkpad.redhat.com> References: <201402121635195912042@163.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <201402121635195912042@163.com> Subject: Re: [Qemu-devel] qemu-1.7.0 vm migration with nbd usage List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: xuanmao_001 Cc: qemu-devel , qemu-discuss On Wed, Feb 12, 2014 at 04:35:19PM +0800, xuanmao_001 wrote: > Hi, Is there a document describe vm migration with nbd server or any example about it? NBD supports migration because it is a network storage protocol. Imagine you have a "source" host and a "destination" host. The source host is currently running the guest. The destination host is the live migration target. There is also an NBD server host that is independent of the source and destination hosts. Launch QEMU on the destination host with the same command-line as the source host, but append "-incoming tcp::1234". Start live migration using the "migrate" command in the QEMU monitor on the source host. Live migration will run until it completes. Finally, you should see your VM has migrated to the destination host. Disk I/O will still use the NBD server host. So there's nothing really special here. NBD is just considered shared storage. Stefan