cinder-backup hangs when restoring to vol larger than restore

[Edward Hope-Morley]

Performing a cinder-backup restore to a volume that is bigger than the original restore hangs since we overflow the restore. It should be zeroing the rest of the volume since it may contain previous data. This only occurs with the ceph backup driver and when specifying a previoulsy created volume.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/46683

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/46683
Committed: http://github.com/openstack/cinder/commit/e54f34245880c105225675d9fb91d2ad1997f7ba
Submitter: Jenkins
Branch: master

commit e54f34245880c105225675d9fb91d2ad1997f7ba
Author: Edward Hope-Morley <email address hidden>
Date: Fri Sep 13 15:20:03 2013 -0400

    Fixes ceph volume restore to larger image than source

    If a backup is restored to a volume that is bigger than the
    backup itself, which could be the case if --volume-id is
    supplied, we need the option to set extraneous extents to zero
    i.e. set extents after those from the backup to zero.

    Change-Id: I3c226b9e7429371f6cd9b9bd81071ff765712975
    Fixes: bug 1225126