8 June 2018, in openstack, devstuff
My tripleo-lab project just saw some major improvements, as well as my Ansible skills.
Its main purpose was to allow to deploy a full tripleo-based openstack, with its undercloud and a bunch of VM playing the role of baremetal nodes to provision. It worked well, and I was pretty happy with it.
But, after running it again and again in order to ensure I actually HAVE my running tripleo, I saw some improvements to make. Like better tagging, better grouping with blocks, and some new features.
In the current process, without any custom environment, you will still get the “full set”, with the preparation of about 9 VMs. Of course, not all computers can run that kind of workload, not even my current one.
In addition, you will find the tripleo-ci tools in the stack home directory, as well as some utilities such as tmux and vim. In this condition, you’re ready to test your changes, run an overcloud deploy and so on.
But, as said, you might not be able to run the complete stuff. In order to allow more people to run this playbook, I’ve added a smaller environment, starting “only” three controllers and a compute (and, of course, the undercloud). With only 5 VMs, this setup should match almost all of the needs.
I also documented in the README how to actually create your own environment. Of course, compared to Quickstart, it’s really light. But that was the intended goal of this tripleo-lab: be quick, light, easy. And help me learning Ansible. And jinja2.
From now on, if you want to run a small env, you just have to issue this command:
And if you want to upgrade to the full set, just re-run the same command
If you don’t want the tripleo-ci thingy, you can just disable it. There are two ways for that: either exclude a tag:
Or pass a custom environment file with the following content:
And in the shell: