From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:45698) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TBzsq-0000Un-P3 for qemu-devel@nongnu.org; Wed, 12 Sep 2012 23:14:29 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1TBzsp-0005hH-JQ for qemu-devel@nongnu.org; Wed, 12 Sep 2012 23:14:28 -0400 Received: from mx1.redhat.com ([209.132.183.28]:44628) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1TBzsp-0005h8-BA for qemu-devel@nongnu.org; Wed, 12 Sep 2012 23:14:27 -0400 Received: from int-mx02.intmail.prod.int.phx2.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id q8D3EQfq015296 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Wed, 12 Sep 2012 23:14:26 -0400 Message-ID: <50515004.5010801@redhat.com> Date: Thu, 13 Sep 2012 11:16:20 +0800 From: Guannan Ren MIME-Version: 1.0 References: <1347463350-28760-1-git-send-email-gren@redhat.com> <5050CAB8.6040702@redhat.com> In-Reply-To: <5050CAB8.6040702@redhat.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [libvirt] [PATCH] snapshot: fix rollback failure in transaction mode List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Eric Blake Cc: libvir-list@redhat.com, Jeff Cody , QEMU Developers On 09/13/2012 01:47 AM, Eric Blake wrote: > On 09/12/2012 09:22 AM, Guannan Ren wrote: >> After failure of qemu transaction command, the snapshot file still >> be there with non-zero in size. In order to unlink the file, the >> patch removes the file size checking. > Can you give some exact steps to reproduce this, so that I know who is > making the file have non-zero size? I'm worried that unlinking a > non-empty file is discarding data, so the commit message needs a lot > more details about how we are proving that the only way the file can be > non-zero size is because qemu happened to put data into a previously > empty file prior to the failed 'transaction' attempt. qemu left non-empty file. Steps: 1, Create a qemu instance with two drive images of qcow2 type (root user) /usr/libexec/qemu-kvm -m 1024 -smp 1 -name "rhel6u1" \ -drive file=/var/lib/libvirt/images/firstqcow2,if=none,id=drive-virtio-disk0,format=qcow2,cache=none -device virtio-blk-pci,scsi=off,bus=pci.0,addr=0x5,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1 \ -drive file=/var/lib/libvirt/images/secondqcow2,if=none,id=drive-virtio-disk1,format=qcow2,cache=none \ -device virtio-blk-pci,scsi=off,bus=pci.0,addr=0x7,drive=drive-virtio-disk1,id=virtio-disk1 -qmp stdio 2, Initialize qemu qmp {"execute":"qmp_capabilities"} 3, Remove the second drive image file rm -f /var/lib/libvirt/images/secondqcow2 4, Run 'transaction' command with snapshot qemu commands in. {"execute":"transaction","arguments": {"actions": [{"type":"blockdev-snapshot-sync","data":{"device":"drive-virtio-disk0","snapshot-file":"/var/lib/libvirt/images/firstqcow2-snapshot.img","format":"qcow2"}}, {"type":"blockdev-snapshot-sync","data":{"device":"drive-virtio-disk1","snapshot-file":"/var/lib/libvirt/images/secondqcow2-snapshot.img","format":"qcow2"}}] }, "id":"libvirt-6"} 5, Got the error as follows: {"id": "libvirt-6", "error": {"class": "OpenFileFailed", "desc": "Could not open '/var/lib/libvirt/images/secondqcow2-snapshot.img'", "data": {"filename": "/var/lib/libvirt/images/secondqcow2-snapshot.img"}}} 6, List first newly-created snapshot file: -rw-r--r--. 1 root root 262144 Sep 13 11:43 firstqcow2-snapshot.img > > That is, after re-reading context code just now, I'm fairly confident > that this code can only be reached when qemu supports the 'transaction' > monitor command, and libvirt's --reuse-ext flag was not specified, and > therefore libvirt must have just created the file. But I want that in > the commit message rather than having to re-read the code every time I > visit this commit in future reads of the git log. That's ok, I will add this information into commit log in v2. > It may also be that > qemu has a bug in that the 'transaction' command is modifying files even > when it fails, so even while this works around the bug, I'm cc'ing Jeff > to see if qemu also needs a bug fix.