Description of problem: After deploying an overcloud, NeutronControlPlaneID param is set to '' in tuskar plan. This causes that both "openstack overcloud update stack" and "openstack overcloud node delete" commands fail. The reason why "openstack overcloud deploy" command works is that it sets NeutronControlPlaneID parameter before any stack create/update API call. This doesn't make much sense - I think one of following should be done: a) don't set empty NeutronControlPlaneID param in tuskar at all b) update NeutronControlPlaneID param in tuskar on overcloud create/update Steps to Reproduce: 1. deploy overcloud following doc ("openstack overcloud deploy") 2. fetch tuskar plan templates: tuskar plan-templates -O tpls $PLAN_UUID 3. check NeutronControlPlaneID is set to '' in tpls/environment.yaml This is a regression - in pre-unifiedCLI age there was not problem with this.
Jan you are right and in fact it is the same for every other variable that is set for a deploy. The difference between [1] (old way) and [2] is that before we were actually patching the tuskar plan with the user-supplied params before requesting the deploy templates. Now, the templates are requested, the user params are also passed like [3] so you get what you want but it isn't persisted in Tuskar. We came across this in scrum on the 17th and it was my impression that someone was already working on changing that behaviour in the osc plugin. I think possibly dmathews or regebro? [1] https://github.com/rdo-management/instack-undercloud/blob/master/scripts/instack-deploy-overcloud#L289 [2] https://github.com/rdo-management/python-rdomanager-oscplugin/blob/master/rdomanager_oscplugin/v1/overcloud_deploy.py#L395 [3] https://github.com/rdo-management/python-rdomanager-oscplugin/blob/master/rdomanager_oscplugin/v1/overcloud_deploy.py#L340
I believe this bug is fixed by this patch: https://review.gerrithub.io/#/c/237713/
(In reply to Lennart Regebro from comment #5) > I believe this bug is fixed by this patch: > > https://review.gerrithub.io/#/c/237713/ This is in place of the previous patch
stack@puma33 ~]$ openstack management plan list +--------------------------------------+-----------+-------------+------------------------------------------------------------------+ | uuid | name | description | roles | +--------------------------------------+-----------+-------------+------------------------------------------------------------------+ | 27f1ac33-6c9d-4d02-a3c9-e0ca463f3b11 | overcloud | None | Compute, Ceph-Storage, Cinder-Storage, Controller, Swift-Storage | +--------------------------------------+-----------+-------------+------------------------------------------------------------------+ [stack@puma33 ~]$ grep NeutronControlPlaneID tpls/environment.yaml NeutronControlPlaneID: 548752ae-51fd-491c-b281-b48a6cbb6d9a python-rdomanager-oscplugin-0.0.8-18.el7ost.noarch
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHEA-2015:1549