one-deploy-validation

Checkout

1. Resursively clone the engineering-deploy repository

   cd ~/ && git clone --recursive [email protected]:OpenNebula/one-deploy-validation.git

Requirements

Note

If Makefile is used then it will create python virtual environments using hatch (on demand).

1. Install hatch

   pip install hatch

   or

   pipx install hatch

   or use any other method you see fit

2. Install the opennebula.deploy collection with dependencies

   make requirements

   if you'd like to pick specific branch (instead of master), tag or a custom fork

   make requirements ONE_DEPLOY_URL:=git+https://github.com/OpenNebula/one-deploy.git,release-1.2.1

   the one-deploy repository checkout should be available in ~/.ansible/collections/ansible_collections/opennebula/deploy/

3. (OPTIONAL) Update the vendor/ceph-ansible submodule (if you want to deploy Ceph clusters)

   cd ~/ && git submodule update --init --recursive -- ./

Playbooks/Roles

1. You can create new playbooks and roles as you desire and still reuse the code from one-deploy

   ---
   - ansible.builtin.import_playbook: opennebula.deploy.site
   
   - hosts:
       - "{{ frontend_group | d('frontend') }}"
       - "{{ node_group | d('node') }}"
     roles:
       - role: { opennebula.deploy.helper.facts, _force: true }
       - role: example

Inventory/Execution

Note

It's exactly the same as with one-deploy.

1. Inventories are kept in the ./inventory/ directory, please take a look at example.yml

2. To execute ansible-playbook you can run

   make I=inventory/example.yml main

   all the normal targets are available by default

   • ceph
   • infra
   • main
   • pre
   • site