Making a deployment resource FAILED should also fail the deployment itself

[Steve Baker]

Currently if a deployment resource goes to a FAILED state (for example due to stack timeout) the corresponding deployment will remain in an IN_PROGRESS state indefinitely.

The deployment resource implementation should also make an RPC call to change the deployment state to FAILED too.

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/330060
Committed: https://git.openstack.org/cgit/openstack/heat/commit/?id=40036afff2f9ae95d133cad85d3197f0b12c5947
Submitter: Jenkins
Branch: master

commit 40036afff2f9ae95d133cad85d3197f0b12c5947
Author: Zane Bitter <email address hidden>
Date: Wed Jul 13 17:03:56 2016 -0400

    Add a Resource plugin API for cancelling actions

    Allow a resource type plugin to provide a handle_*_cancel() method to
    handle cancelling an in-progress action.

    Change-Id: I20d243328daf1ffa1a40d55c5fcf889c70a8a59c
    Related-Bug: #1585815
    Related-Bug: #1591337

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/484741

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/484741
Committed: https://git.openstack.org/cgit/openstack/heat/commit/?id=199e4ac322f815c774895b0b20479264dec8853f
Submitter: Jenkins
Branch: master

commit 199e4ac322f815c774895b0b20479264dec8853f
Author: rabi <email address hidden>
Date: Mon Jul 10 16:40:12 2017 +0530

    Implement handle_<action>_cancel for SoftwareDeployment

    This implements handle_<action>_cancel for SoftwareDeployment
    to update the deployment to failed, when the resource fails due
    to scheduler timeout or some other external reason.

    Change-Id: Ibaa551d35e96f34cf950811a5b5a05e1cda6c364
    Closes-Bug: #1585815

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/heat 9.0.0.0rc1 release candidate.