Trove resource deletion fails to clean up resource

[Franciraldo Cavalcante]

We delete a Heat stack containing OS::Trove::Cluster, and it fails:

+---------------+--------------------------------------+--------------------+-----------------+---------------------+
| resource_name | physical_resource_id                 | resource_type      | resource_status | updated_time        |
+---------------+--------------------------------------+--------------------+-----------------+---------------------+
| mysql_cluster | dd82ebac-9e78-487b-9eb7-41d115fcd38b | OS::Trove::Cluster | DELETE_FAILED   | 2016-06-28T18:13:27 |
+---------------+--------------------------------------+--------------------+-----------------+---------------------+

Here's the reason:

stack_status_reason   | Resource DELETE failed: UnprocessableEntity:                                                                                             |
|                       | resources.mysql_cluster: This action cannot be                                                                                           |
|                       | performed on the cluster while the current cluster task                                                                                  |
|                       | is 'BUILDING'. (HTTP 422)

We then delete the trove cluster from cmd line, with no failure, and retry the heat stack deletion. The stack deletion still fails with the same message above. Apparently Heat is not properly reading the status from Trove service.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/390828

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/390828
Committed: https://git.openstack.org/cgit/openstack/heat/commit/?id=0fb991c332583ad57574274afb146c2f78bcad48
Submitter: Jenkins
Branch: master

commit 0fb991c332583ad57574274afb146c2f78bcad48
Author: Peter Razumovsky <email address hidden>
Date: Wed Oct 26 15:02:12 2016 +0300

    Move call trove cluster delete to check complete

    There could be situation, when cluster cannot be available for
    deleting, for example, it has BUILDING status. In such case delete
    call will raise UnprocessableEntity exc. So, need to move cluster
    delete to check_delete_complete and try to delete cluster.

    Also, test successful cluster delete was fixed.

    Change-Id: Ie05f33d6a3f653406d80a362066cac3ffec1e1ef
    Closes-bug: #1597374

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/heat 8.0.0.0b2 development milestone.