update attach status when instance is not a valid UUID

[John Griffith]

Issue reported on ask.openstack (https://ask.openstack.org/question/1828/attach-volume-get-exception-when-instance_uuid-is-invalidbut-the-status-of-volume-is-still-attaching/).

We should do an update on the volume status after we catch the InvalidUUID exception on attach.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/33560

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/33560
Committed: http://github.com/openstack/cinder/commit/4aaa3b64a3ca3050ebed3d748d46facafc09413b
Submitter: Jenkins
Branch: master

commit 4aaa3b64a3ca3050ebed3d748d46facafc09413b
Author: John Griffith <email address hidden>
Date: Tue Jun 18 16:49:13 2013 -0600

    Update attach status when instance id invalid.

    In cinder.manager do_attach we have a check to make sure the
    supplied instance id is a valid UUID which is great, but if it's
    not we just raise and dump out, leaving the volume in an attaching
    state.

    We should update the status to error_attaching in this case and then
    raise/return.

    Fixes bug: 1192370

    Change-Id: Ia759ea7a387544913953ca84e9ed927480c4edfd