Ask Your Question

Revision history [back]

click to hide/show revision 1
initial version

This post does not tend to be an answer to this question, but rather add a comment with a legible formatting.

I never used the pre-provisioned method to deploy overcloud, but wanted to share my thoughts on some of your questions.

  • Jinja2 templates (.j2) will be automatically rendered into yaml files during deployment. You just need to pretend the file is already there when you pass it to the deploy command. So you are doing it right.
  • EC2MetadataIp and ControlPlaneDefaultRoute need to be provided in an environment file. You may add them into your test-net.yaml file, immediately under the 'parameter_defaults' section.
  • When creating and using environment files outside of the default heat template directory, try to use the absolute path in the 'resource_registry' to avoid issues. In the test-net.yaml, set OS::TripleO::DeployedServer::ControlPlanePort: /usr/share/openstack-tripleo-heat-templates/deployed-neutron-port.yaml
  • Some command line parameters you used have been deprecated . Instead of using --control-flavor --control-scale and the like, it is prefered to set them in an environment file under the 'parameter_defaults' section such as:
    • OvercloudControllerFlavor: control
    • ControllerCount: 1

It would be nice to know more details on your network and NIC setup in this case. How many NICs do you have on each node? Do you plan to use the provisioning network (ctlplane) for all overcloud services?

This post does not tend to be an answer to this question, but rather add adds a comment with a legible formatting.

I never used the pre-provisioned method to deploy overcloud, but wanted to share my thoughts on some of your questions.

  • Jinja2 templates (.j2) will be automatically rendered into yaml files during deployment. You just need to pretend the file is already there when you pass it to the deploy command. So you are doing it right.
  • EC2MetadataIp and ControlPlaneDefaultRoute need to be provided in an environment file. You may add them into your test-net.yaml file, immediately under the 'parameter_defaults' section.
  • When creating and using environment files outside of the default heat template directory, try to use the absolute path in the 'resource_registry' to avoid issues. In the test-net.yaml, set OS::TripleO::DeployedServer::ControlPlanePort: /usr/share/openstack-tripleo-heat-templates/deployed-neutron-port.yaml
  • Some command line parameters you used have been deprecated . Instead of using --control-flavor --control-scale and the like, it is prefered to set them in an environment file under the 'parameter_defaults' section such as:
    • OvercloudControllerFlavor: control
    • ControllerCount: 1

It would be nice to know more details on your network and NIC setup in this case. How many NICs do you have on each node? Do you plan to use the provisioning network (ctlplane) for all overcloud services?

This post does not tend to be an answer to this question, but rather adds a comment with a legible formatting.

I never used the pre-provisioned method to deploy overcloud, but wanted to share my thoughts on some of your questions.

  • Jinja2 templates (.j2) will be automatically rendered into yaml files during deployment. You just need to pretend the file is already there when you pass it to the deploy command. So you are doing it right.
  • EC2MetadataIp and ControlPlaneDefaultRoute need to be provided in an environment file. You may add them into your test-net.yaml file, immediately under the 'parameter_defaults' section.
  • When creating and using environment files outside of the default heat template directory, try to use the absolute path in the 'resource_registry' to avoid issues. In the test-net.yaml, set OS::TripleO::DeployedServer::ControlPlanePort: /usr/share/openstack-tripleo-heat-templates/deployed-neutron-port.yaml/usr/share/openstack-tripleo-heat-templates/deployed-server/deployed-neutron-port.yaml
  • Some command line parameters you used have been deprecated . Instead of using --control-flavor --control-scale and the like, it is prefered to set them in an environment file under the 'parameter_defaults' section such as:
    • OvercloudControllerFlavor: control
    • ControllerCount: 1

It would be nice to know more details on your network and NIC setup in this case. How many NICs do you have on each node? Do you plan to use the provisioning network (ctlplane) for all overcloud services?