API doesn't fail on Down Service for manage/get_manageable volumes/snapshots

[Gorka Eguileor]

Existing code will fail at API level if a service is disabled and we try to manage/get_manageable volumes/snapshots on that service, but it won't fail at API level if the service is down, and the message will wait in the message queue until the service is up again, which is misleading.

We should fail at API level when the service is down.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/345245

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/345245
Committed: https://git.openstack.org/cgit/openstack/cinder/commit/?id=889947c622245e143b369a908c5dcbd7c3371b1f
Submitter: Jenkins
Branch: master

commit 889947c622245e143b369a908c5dcbd7c3371b1f
Author: Gorka Eguileor <email address hidden>
Date: Thu Jul 21 10:11:30 2016 +0200

    Fail manage operations if service is down

    Existing code will fail at API level if a service is disabled and we try
    to manage/get_manageable volumes/snapshots on that service, but it won't
    fail at API level if the service is down, and the message will wait in
    the message queue until the service is up again, which is misleading.

    This patch changes this behavior and makes manage_existing,
    manage_existing_snapshot, get_manageable_volumes, and
    get_manageable_snapshots, fail at the API if the service is down.

    Change-Id: Ifd56a46fce9ac2fe72a30820c05a5cf9adee229f
    Closes-Bug: #1605109

[Thierry Carrez (ttx)]

This issue was fixed in the openstack/cinder 9.0.0.0b3 development milestone.