The Platform: How CNDI Templates Automate Project Setup

CNDI Projects are GitOps and Infrastructure-as-Code repositories.

They have 2 user-facing files you need to know about:

• cndi_config.yaml - This file describes the project's infrastructure, applications, and configuration.
• .env - This file contains secrets and other environment variables that are used by the cluster and CI.

If you understand those files, you understand CNDI and your cluster.

But where do they come from? How do you create them?

CNDI Templates 🏗️

CNDI Templates are a way to declare prompts and how the responses should flow into those config files. Let's look at an example:

# each prompt is used to collect information from a user
prompts:
  - name: "project_name"
    message: "What is the name of your project?"
    type: "Input"
    default: "my-project" # default value, used if `cndi` is run non-interactively

  - name: "deployment_target_provider"
    message: "What cloud provider are you using?"
    type: "Select"
    default: "aws"
    options:
      - "aws"
      - "gcp"
      - "azure"
      - "dev"

- name: "git_token"
  message: "What is your GitHub Personal Access Token?"
  type: "Secret"
  default: ""

outputs:
  # data to be written to the cndi_config.yaml file
  cndi_config:
    # basic metadata
    cndi_version: v2
    project_name: "{{ $cndi.get_prompt_response(project_name) }}"
    provider: "{{ $cndi.get_prompt_response(deployment_target_provider) }}"

    # high-level infra like nodes and platform services
    infrastructure: { ... }

    # generally Helm charts for the applications you're deploying
    applications: { ... }

    # kubernetes manifests to apply to the cluster
    cluster_manifests: { ... } 

  # a set of environment variables to be written to the .env file
  env:
    GIT_TOKEN: "{{ $cndi.get_prompt_response(git_token) }}"

  # a markdown file to be written to the README.md
  readme:
    title: "# {{ $cndi.get_prompt_response(project_name) }}"
    example_markdown_section: |
        ## Example
        ```bash
        echo "Hello, World!"
        ```

The best way to get started with CNDI is to call cndi create and point to a Template. Templates are just YAML files like the one above, and they can be hosted anywhere.

Officially recognized Templates can be called by name:

cndi create --template kafka

You can also point to a local file or a remote URL:

cndi create -t https://raw.githubusercontent.com/polyseam/cndi/refs/heads/main/templates/basic.yaml

When you run cndi create -t airflow for example, you'll be prompted for some elements that are relevant for any CNDI cluster, and some values which are specific to your deployment target, and some for Airflow itself.

We have all of the official Templates hosted in the ./templates directory of the cndi GitHub Repo, and they can be helpful when learning to write your own Templates.

Configurator 🤖

We've just launched another brand new tool to make getting started with CNDI Templates even easier called the CNDI Configurator! This tool enables running through the interactive Template prompts in the browser rather than on the command line.

The Configurator can help you get a feel for the tool and which questions you'll be asked, and when you are happy with your responses, you can download a my_responses.yaml file to use with the CLI:

cndi create -r ./my_responses.yaml

Conclusion 🎯

Now that you have a clear understanding of how CNDI Templates work, you can feel confident in using them and even creating your own!

After you've tried the configurator you can find the cndi project on GitHub! ⭐️