Abandon stack lacks locking

[Steven Hardy]

delete_stack in service.py contains locking logic, so we can avoid races where other actions are performed on a stack while it's being deleted.

abandon_stack does essentially the same thing, but lacks any locking, so there will be problems if one engine does an abandon while another is servicing some other lifecycle request, like a stack update for example.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/86663

[Openstack Gerrit (openstack-gerrit)]

Reviewed: https://review.openstack.org/86663
Committed: https://git.openstack.org/cgit/openstack/heat/commit/?id=f9c0ae9ca3b7187ee831d14b5143356deeb6785f
Submitter: Jenkins
Branch: master

commit f9c0ae9ca3b7187ee831d14b5143356deeb6785f
Author: Vijendar Komalla <email address hidden>
Date: Thu Apr 10 12:46:04 2014 -0500

    Implement locking in abandon stack

    This change implements locking in abandon stack. If lock
    acquire fails throws ActionInPrgoress exception.

    Change-Id: I930123efa95b2863c4e2de01f4646b6e3d68c5cc
    Closes-bug: #1301311