Periodic reclaiming of soft deleted instances stops on first error

[Hans Lindgren]

Same as for periodic cleanup (bug 1192712), if an instance reclaim fails with an exception, no more instances will be reclaimed.
If the failure is permanent, this will block other instances from being reclaimed for as long as it remains.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/34620

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/34620
Committed: http://github.com/openstack/nova/commit/b5053888b4db36272102f027b7e1d292c430d8f5
Submitter: Jenkins
Branch: master

commit b5053888b4db36272102f027b7e1d292c430d8f5
Author: Hans Lindgren <email address hidden>
Date: Wed Jun 26 23:28:50 2013 +0200

    Make sure periodic instance reclaims continues on error

    If an instance reclaim operation fails with an exception, no more
    instances will be reclaimed. If the failure is permanent, this will
    block other instances from being reclaimed for as long as it remains.

    Resolves bug 1195029.

    Change-Id: I0379d16a1f74344ef44c843ce19a505e31e79bfa