argocd-preview-environments

This repository, argocd-preview-environments, is designed to demonstrate and test ArgoCD's ability to automatically create preview environments for pull requests. Using ArgoCD's ApplicationSet with the pull request generator, each new pull request triggers a temporary environment that mirrors the changes in the PR.

Purpose

The main goal of this repository is to facilitate testing and exploration of ArgoCD’s pull request generator functionality, where every pull request can be deployed as a unique environment for testing.

How It Works

1. Automatic Deployment on Pull Requests: Whenever a new pull request is created, ArgoCD detects the PR and creates a dedicated preview environment based on the branch and pull request number.
2. ApplicationSet Configuration: The repository uses an ApplicationSet with a pull request generator that watches for PRs labeled with "preview".
3. Environment Cleanup: After a PR is closed or merged, ArgoCD can automatically prune the environment, ensuring only active PRs have live deployments.

Repository Structure

├── ansible.cfg
├── ansible-navigator.log
├── bootstrap.yaml
├── collections
│   └── requirements.yaml
├── gitops
│   ├── applications
│   │   └── bgd
│   │       ├── base
│   │       │   ├── bgd-deployment.yaml
│   │       │   ├── bgd-route.yaml
│   │       │   ├── bgd-svc.yaml
│   │       │   └── kustomization.yaml
│   │       └── overlays
│   │           └── default
│   │               ├── bgd-deployment-patch.yaml
│   │               └── kustomization.yaml
│   ├── bootstrap
│   │   └── bgd-boostrap.yaml
│   └── installation
│       ├── argocd-app-bootstrap.yaml
│       ├── argocd-server.yaml
│       └── gitops-operator.yaml
├── group_vars
│   ├── all
│   │   └── main.yml
│   └── cluster1
│       └── vault.yaml
├── inventory
├── README.md

Setup Instructions

1. Install the required packages

$  python3 -m pip install ansible-navigator --user
$  python3 -m pip install ansible-builder --user

2. Fork and Clone the Repository

   • Fork this repository to your GitHub account and clone it locally to test it with your own instance.

   git clone https://github.com/your-username/argocd-preview-environments.git
   cd argocd-preview-environments

   • Update the Application configuration in gitops/installation/argocd-app-bootstrap.yaml with your GitHub repository details if needed.

   source:
     repoURL: 'https://github.com/owner/repository_name.git'

   • Update the ApplicationSet configuration in gitops/bootstrap/bgd-boostrap.yaml with your GitHub repository details if needed.

   github:
     owner: user or organization
     repo: repository name
     labels: (Optional)
       - label that must match (e.g., "preview") for creating an Application for each pull request.
   
   source:
     repoURL: 'https://github.com/owner/repository_name.git'

   - Commit and push all changes
   
   ```bash
   $ git add --all; git commit -am "Update my repository"; git push
   

3. Populate and encrypt your credentials and custom data

$ vim group_vars/cluster1/vault.yaml
$ vim group_vars/all/main.yml
$ ansible-vault encrypt group_vars/*/vault.yaml
$ vim .vault_password

4. Create OCP object needed to deploy Openshift GitOps and bootstrap the application

$ ansible-navigator run bootstrap.yaml -i inventory -l cluster1 -m stdout --eei quay.io/automationiberia/ee-ocp-aap-iac-casc --vault-password-file .vault_password

5. Verify application is running

$ oc get appset -n openshift-gitops

6. Success

This configuration will create an Application for each pull request labeled as "preview."

$ oc get app -n openshift-gitops

Variables

Variable Name	Default Value	Required	Description	Example
ocp_api_key	n/a	yes	Token used to authenticate with the Openshift API	'sha256~Po6ydC7CVs12drESQeNiUW9poUT84aFrj7zL3VQfvrS'
ocp_api_host	n/a	yes	Provide a URL for accessing the Openshift API	'=https://api.cluster-ocp.lab.example.com:6443'
ocp_api_validate_certs	false	no	Whether or not to verify the API server’s SSL certificates	'true'
aws_access_key	n/a	yes	Amazon Access Key to access AWS API	'R767AKIFYSF5INA6QKB6'
aws_secret_key	n/a	yes	Amazon Secret Key to access AWS API	'qKCYpd/jQX6gRhucQwIT1d2lzrapZ/O4lpEKGGqR'
aws_region	n/a	yes	Amazon region where the S3 will be deploy it	'us-central-3'