glusterfs: Rebased snapshot volume is not deleted

[Thang Pham]

When deleting a volume that was snapshot, the volume is not properly deleted.

Steps to recreate bug:
1. Create a volume

2. Attach volume to a running instance.

3. Take an online snapshot of the volume.
Note that the active volume used by the instance is now switched to volume-<uuid>.<snapshot-uuid>.

4. Delete the snapshot.
The snapshot volume will be rebased, and the "active" volume as seen in volume-<uuid>.info is now set to
volume-<uuid>.<snapshot-uuid>.

Under the glusterfs mount point, there are now only 2 files related to the volume:
  - volume-<uuid>.info
  - volume-<uuid>.<snapshot-uuid>
The original volume base volume-<uuid> is deleted after the rebase.

5. Detach the volume from the running instance.

6. Delete the volume.
Under the glusterfs mount point, the volume-<uuid>.<snapshot-uuid> file is not deleted as expected.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/86158

[Eric Harney (eharney)]

I'd suggest that this is probably "medium" priority since it causes leaking of backend space.

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/86158
Committed: https://git.openstack.org/cgit/openstack/cinder/commit/?id=5f00cad02da1093d71f636add0810a538cbd444f
Submitter: Jenkins
Branch: master

commit 5f00cad02da1093d71f636add0810a538cbd444f
Author: Thang Pham <email address hidden>
Date: Tue Apr 8 16:45:34 2014 -0400

    GlusterFS: Delete active snapshot file on volume delete.

    If a snapshot is taken of a volume that is attached to an active
    instance, the volume file used by the instance will be switched to
    the new snapshot file that is created. When you delete the
    snapshot, the base volume file will be merged with the snapshot
    file and the base volume is deleted. Upon a deleting the active
    volume, the active snapshot file is not deleted because it does not
    have the expected name that cinder is looking for, i.e.
    volume-<uuid>. Instead, the snapshot file has the name
    volume-<uuid>.<snapshot-uuid>. This patch looks at the volume info
    file to find any active snapshot file and properly delete it when
    the volume is deleted.

    Change-Id: Ib0af4401d839ec3bd1eb3a81e1671811e0d4a288
    Closes-Bug: #1300303

[Openstack Gerrit (openstack-gerrit)]

Fix proposed to branch: stable/icehouse
Review: https://review.openstack.org/89560

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/89560
Committed: https://git.openstack.org/cgit/openstack/cinder/commit/?id=560740f7348da143d8739980a06496d6e24d1c4e
Submitter: Jenkins
Branch: stable/icehouse

commit 560740f7348da143d8739980a06496d6e24d1c4e
Author: Thang Pham <email address hidden>
Date: Tue Apr 8 16:45:34 2014 -0400

    GlusterFS: Delete active snapshot file on volume delete.

    If a snapshot is taken of a volume that is attached to an active
    instance, the volume file used by the instance will be switched to
    the new snapshot file that is created. When you delete the
    snapshot, the base volume file will be merged with the snapshot
    file and the base volume is deleted. Upon a deleting the active
    volume, the active snapshot file is not deleted because it does not
    have the expected name that cinder is looking for, i.e.
    volume-<uuid>. Instead, the snapshot file has the name
    volume-<uuid>.<snapshot-uuid>. This patch looks at the volume info
    file to find any active snapshot file and properly delete it when
    the volume is deleted.

    Change-Id: Ib0af4401d839ec3bd1eb3a81e1671811e0d4a288
    Closes-Bug: #1300303
    (cherry picked from commit 5f00cad02da1093d71f636add0810a538cbd444f)