Automation and GitOps Overview

Automating control plane deployments with GitOps enables declarative, version-controlled infrastructure management. This section covers integrating GitOps workflows with Upbound control planes using Argo CD and related tools.

What is GitOps?

GitOps is an approach for managing infrastructure by:

• Declaratively describing desired system state in Git
• Using controllers to continuously reconcile actual state with desired state
• Treating Git as the source of truth for all configuration and deployments

Upbound control planes are fully compatible with GitOps patterns and we strongly recommend integrating GitOps in the platforms you build on Upbound.

Key Concepts

Argo CD

Argo CD is a popular Kubernetes-native GitOps controller. It continuously monitors Git repositories and automatically applies changes to your infrastructure when commits are detected.

Deployment Models

The way you configure GitOps depends on your deployment model:

Aspect	Cloud Spaces	Self-Hosted Spaces
Access Method	Upbound API with tokens	Kubernetes native (secrets/kubeconfig)
Configuration	Kubeconfig via up CLI	Control plane connection secrets
Setup Complexity	More involved (API integration)	Simpler (native Kubernetes)
Typical Use Case	Managing Upbound resources	Managing workloads on control planes

Getting Started

Choose your path based on your deployment model:

Cloud Spaces

If you're using Upbound Cloud Spaces (Dedicated or Managed):

1. Start with GitOps with Upbound Control Planes
2. Learn how to integrate Argo CD with Cloud Spaces
3. Manage both control plane infrastructure and Upbound resources declaratively

Self-Hosted Spaces

If you're running self-hosted Spaces:

1. Start with GitOps with ArgoCD in Self-Hosted Spaces
2. Learn how to configure control plane connection secrets
3. Manage workloads deployed to your control planes

Common Workflows

Workflow 1: Managing Control Planes with GitOps

Create and manage control planes themselves declaratively using provider-kubernetes:

apiVersion: kubernetes.crossplane.io/v1alpha2
kind: Object
metadata:
  name: my-controlplane
spec:
  forProvider:
    manifest:
      apiVersion: spaces.upbound.io/v1beta1
      kind: ControlPlane
      # ... control plane configuration

Workflow 2: Managing Workloads on Control Planes

Deploy applications and resources to control planes using standard Kubernetes GitOps patterns:

apiVersion: v1
kind: Namespace
metadata:
  name: my-app
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: my-app
  namespace: my-app
# ... deployment configuration

Workflow 3: Managing Upbound Resources

Use provider-upbound to manage Upbound IAM and repository resources:

• Teams
• Robots and their team memberships
• Repositories and permissions

Advanced Topics

Argo CD Plugin for Upbound

Learn more in the ArgoCD Plugin guide for enhanced integration with self-hosted Spaces.

Declarative Control Plane Creation

See Declaratively create control planes for advanced automation patterns.

Consuming Control Plane APIs

Understand how to consume control plane APIs in your app cluster with Argo CD.

Prerequisites

Before implementing GitOps with control planes, ensure you have:

For Cloud Spaces:

• Access to Upbound Cloud Spaces
• up CLI installed and configured
• API token with appropriate permissions
• Argo CD or similar GitOps controller running
• Familiarity with Kubernetes RBAC

For Self-Hosted Spaces:

• Self-hosted Spaces deployed and running
• Argo CD deployed in your infrastructure
• Kubectl access to the cluster hosting Spaces
• Understanding of control plane architecture

Next Steps

1. Choose your deployment model above
2. Review the relevant getting started guide
3. Set up your GitOps controller (Argo CD)
4. Deploy your first automated control plane
5. Explore advanced topics as needed

tip

Start with simple deployments to test your GitOps workflow before moving to production. Use simulations to preview changes before applying them.