Mistral environment not reset between deployments

[Julie Pichon]

When using "openstack overcloud deploy --templates" to deploy, the 'overcloud' plan gets recreated. However at the moment it would appear that only the templates in the Swift container are updated, while the Mistral environment stays on. This causes issues and confusing on redeployments because environment parameters that are no longer valid are still being used.

$ openstack overcloud deploy --templates --control-scale 1 --compute-scale 1
Removing the current plan files
Uploading new plan files
Started Mistral Workflow tripleo.plan_management.v1.update_deployment_plan. Execution ID: 27c896d8-638c-4350-92f0-b04f0f67d17c
Plan updated
Deploying templates in the directory /tmp/tripleoclient-f8n2Vw/tripleo-heat-templates
Started Mistral Workflow tripleo.deployment.v1.deploy_plan. Execution ID: 57e92bac-370c-4db7-b14b-413d91d489fa
[...]

The code at https://github.com/openstack/python-tripleoclient/blob/a62216/tripleoclient/workflows/plan_management.py#L124 suggests that we actually mean to be resetting the environment, though it doesn't seem to be working at the moment.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/394419

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/394419
Committed: https://git.openstack.org/cgit/openstack/python-tripleoclient/commit/?id=fb6f500ffed3696f2b8cffdb5338ff4aa2bdd12f
Submitter: Jenkins
Branch: master

commit fb6f500ffed3696f2b8cffdb5338ff4aa2bdd12f
Author: Dougal Matthews <email address hidden>
Date: Mon Nov 7 13:12:08 2016 +0000

    Reset the parameter_defaults between deployments via the CLI

    When the CLI makes subsequent deployments it manually updates the plan
    in Mistral. We need to do this for backwards compatibility as the deploy
    command previously didn't have any plans or history behind it. So if a
    user deploys with a specific parameter or template and then they remove
    it in the next deploy it shouldn't still exist. In short, the CLI
    expects everything to be passed to the deploy command every time.
    Therefore, the easiest way to handle this is by manually resetting the
    plan.

    This has mostly been the case, but we neglected to reset the
    parameter_defaults. Meaning, that a user may deploy with a parameter
    like "ComputeScale: 5" and then deploy again, without defining it and it
    would use the previous value when they would expect it to use the
    default.

    Change-Id: I09048a550599146ef9bd81a309d80b10497df7c8
    Closes-Bug: #1639787

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/newton
Review: https://review.openstack.org/396524

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/396524
Committed: https://git.openstack.org/cgit/openstack/python-tripleoclient/commit/?id=6c08ab7a4d35b3574291de04693543398ee6a571
Submitter: Jenkins
Branch: stable/newton

commit 6c08ab7a4d35b3574291de04693543398ee6a571
Author: Dougal Matthews <email address hidden>
Date: Mon Nov 7 13:12:08 2016 +0000

    Reset the parameter_defaults between deployments via the CLI

    When the CLI makes subsequent deployments it manually updates the plan
    in Mistral. We need to do this for backwards compatibility as the deploy
    command previously didn't have any plans or history behind it. So if a
    user deploys with a specific parameter or template and then they remove
    it in the next deploy it shouldn't still exist. In short, the CLI
    expects everything to be passed to the deploy command every time.
    Therefore, the easiest way to handle this is by manually resetting the
    plan.

    This has mostly been the case, but we neglected to reset the
    parameter_defaults. Meaning, that a user may deploy with a parameter
    like "ComputeScale: 5" and then deploy again, without defining it and it
    would use the previous value when they would expect it to use the
    default.

    Change-Id: I09048a550599146ef9bd81a309d80b10497df7c8
    Closes-Bug: #1639787
    (cherry picked from commit fb6f500ffed3696f2b8cffdb5338ff4aa2bdd12f)

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/python-tripleoclient 5.5.0 release.

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/python-tripleoclient 5.4.1 release.