From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:40778) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1a5vmn-0004Mc-OP for qemu-devel@nongnu.org; Mon, 07 Dec 2015 08:25:02 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1a5vml-00038U-2L for qemu-devel@nongnu.org; Mon, 07 Dec 2015 08:25:01 -0500 Received: from mx1.redhat.com ([209.132.183.28]:37150) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1a5vmk-00038I-Ss for qemu-devel@nongnu.org; Mon, 07 Dec 2015 08:24:58 -0500 Date: Mon, 7 Dec 2015 14:24:55 +0100 From: Kevin Wolf Message-ID: <20151207132455.GC4191@noname.str.redhat.com> References: <56654C56.7000600@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <56654C56.7000600@gmail.com> Subject: Re: [Qemu-devel] qemu-img: Could not open '....qcow2': Could not read snapshots: File too large List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Kis Gellert Cc: qemu-devel@nongnu.org Am 07.12.2015 um 10:07 hat Kis Gellert geschrieben: > Today one of my virtual machines failed starting. > Previously , I might made a snapshot while VM running. Never access disk images from more than one process at the same time (except if all of them are read-only), it causes corruption. The proper way of taking a snapshot of a running VM is the qemu monitor (i.e. in your case through libvirt). Your best bet for recovering at least some data is using either a hex editor or tests/qemu-iotests/qcow2.py from the qemu source tree to update the number of snapshots to 0 and then use qemu-img convert to copy the data out. Kevin