We have several requests on attaching specific subscriptions to specific roles, defining multiple pools, etc. Instead of re-implementing subscription manager, keep our default way and provide a way on how to achieve advanced use cases (subscription per role, multiple pools, etc). This should be already doable via pre-/post-deployment hooks. If not, please comment and we will need to re-evaluate the scope. Related BZs will be linked as dependent on this specific RFE. Procedure will need to be defined by Eng, tested by QE and documented by Docs team.
*** Bug 1396351 has been marked as a duplicate of this bug. ***
*** Bug 1368037 has been marked as a duplicate of this bug. ***
*** Bug 1408571 has been marked as a duplicate of this bug. ***
Keeping in OSP12 since this is best practices / documentation only (of course supported by QE) but no code is involved - no risk on disrupting product delivery.
*** Bug 1462084 has been marked as a duplicate of this bug. ***
See https://blueprints.launchpad.net/tripleo/+spec/rhsm-integ-ansible where I think this blueprint would help to achieve this RFE.
There is an ansible module already existing for these needs: http://docs.ansible.com/ansible/latest/redhat_subscription_module.html we can just document the proceedure and publish some examples for now and build something around it in future releases
I think I just got this working on OSP10 and without needing to hack into rhel-registration. rhel-registration is just a pre-config hook. It means it is usually called from your env file in this manner: resource_registry: [...] OS::TripleO::NodeExtraConfig: all-pre-config.yaml ..which, in turn, points to something like this: resources: RhelRegistrationPreConfig: type: all-pre-config/rhel-registration.yaml properties: server: {get_param: server} So, in order to override -some- parameters for some node Roles, I split the pre-config hooks this way: resource_registry: [...] OS::TripleO::ControllerExtraConfigPre: controller-pre-config.yaml OS::TripleO::ComputeExtraConfigPre: compute-pre-config.yaml OS::TripleO::CephStorageExtraConfigPre: ceph-pre-config.yaml OS::TripleO::NodeExtraConfig: all-pre-config.yaml Each of these sub-files looks like this (here, the ceph pre-config file, ceph-pre-config.yaml ): $ cat ceph-pre-config.yaml heat_template_version: 2014-10-16 description: 'Extra Pre-Deployment Config, Ceph' parameters: server: type: string rhel_reg_repos: type: string resources: RhelRegistrationPreConfig: type: all-pre-config/rhel-registration.yaml properties: server: {get_param: server} rhel_reg_repos: "rhel-7-server-rpms,rhel-7-server-supplementary-rpms,rhel-7-server-optional-rpms,rhel-7-server-extras-rpms,rhel-7-server-openstack-10-rpms,rhel-7-server-openstack-10-tools-rpms,rhel-7-server-openstack-10-optools-rpms,rhel-7-server-openstack-10-devtools-rpms,rhel-7-server-rhceph-2-osd-rpms,rhel-7-server-rhceph-2-mon-rpms,rhel-7-server-rhceph-2-tools-rpms,rhel-7-server-satellite-tools-6.2-rpms" outputs: deploy_stdout: value: list_join: - '' - - {get_attr: [RhelRegistrationPreConfig, deploy_stdout]} Here, the 'rhel_reg_repos' parameter is role-dependent and the deployment will re-use the rest of the global parameters found in all-pre-config/rhel-registration.yaml before calling the actual script (rhel-registration). Note, that I used 'rhel_reg_repos' in this example but really any parameter found in rhel-registration.yaml [1] can be overridden in this fashion. Works for me (tm), tested on OSP10z6:
*** Bug 1456957 has been marked as a duplicate of this bug. ***
Feature test consistently passing downstream CI since puddle 2018-10-17.2
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-2019:0045