rdeploy

A CLI tool that makes it easy to build and deploy applications to Kubernetes on GCP. It wraps common tools (helm, kubectl, gcloud) into simple commands configured via a project-level YAML file.

Installation

pip install rdeploy

Quick Start

1. Create an rdeploy.yaml in your project root (see Configuration)
2. Run commands:

# Set cluster context
rdeploy set-context staging

# Build and push docker image
rdeploy cloudbuild staging v1.2.3

# Deploy to kubernetes
rdeploy upgrade staging v1.2.3

# Create a new release tag
rdeploy git-release patch

Commands

Command	Description
set-project	Set the active GCP project
set-cluster	Set the active Kubernetes cluster
set-context	Switch cluster and namespace context
activate	Fetch and set project, cluster, and namespace
install	Install a Helm chart deployment
upgrade	Upgrade a Helm deployment to a new image tag (atomic by default: waits for the rollout and rolls back on failure; use --no-atomic to skip, --timeout to adjust the wait, default 15m)
helm	Run arbitrary Helm commands
helm-setup	Download and configure a local Helm binary
git-release	Bump version and push a git tag
next-version	Show what the next version number would be
latest-version	Show the current latest version from git tags
build	Build and push a Docker image
cloudbuild	Build a Docker image via Google Cloud Build
live-image	Display the currently deployed image and tag
shell	Exec into a management container
manage	Run a Django management command in-cluster
create-namespace	Create a Kubernetes namespace
upload-secrets	Upload secrets from an env file to Kubernetes
decode-secret	Print decoded values of a Kubernetes secret
upload-static	Upload static files to a GCS bucket
create-bucket	Create a private GCS bucket
create-public-bucket	Create a public GCS bucket
create-volume	Create a GCE persistent disk
compose	Wrapper for docker-compose

Configuration

rdeploy is configured via an rdeploy.yaml file in your project root. The file supports versioned schemas (v1, v2, v3).

Version 3 (recommended)

version: '3'
configs:
  staging:
    project_name: my-service
    namespace: staging
    kube_context: gke_my-project_us-central1_cluster-1
    docker_image: gcr.io/my-project/my-service:latest
    cloud_provider:
      name: gcp
      project: my-project
      kube_cluster: cluster-1
      region: us-central1
      helm_registry: us-central1-docker.pkg.dev
    helm_values_path: ./etc/helm/staging/values.yaml
    helm_chart: rehive-service
    helm_chart_version: 0.2.0
    helm_version: 3.14.0
    use_system_helm: true

  production:
    project_name: my-service
    namespace: production
    kube_context: gke_my-project_us-central1_cluster-1
    docker_image: gcr.io/my-project/my-service:latest
    cloud_provider:
      name: gcp
      project: my-project
      kube_cluster: cluster-1
      region: us-central1
      helm_registry: us-central1-docker.pkg.dev
    helm_values_path: ./etc/helm/production/values.yaml
    helm_chart: rehive-service
    helm_chart_version: 0.2.0
    helm_version: 3.14.0
    use_system_helm: true

Version 2

version: '2'
configs:
  staging:
    project_name: my-service
    namespace: staging
    docker_image: gcr.io/my-project/my-service:latest
    cloud_provider:
      name: gcp
      project: my-project
      kube_cluster: cluster-1
      region: us-central1
      helm_registry: us-central1-docker.pkg.dev
    helm_values_path: ./etc/helm/staging/values.yaml
    helm_chart: rehive/rehive-service
    helm_chart_version: 0.1.38
    helm_version: 3.14.0
    use_system_helm: true

Configuration Reference

Field	Description
version	Config schema version (1, 2, or 3)
project_name	Name of the Helm release and Kubernetes deployment
namespace	Kubernetes namespace
kube_context	(v3) Direct kubectl context name
docker_image	Docker image path
cloud_provider.name	Cloud provider (currently gcp)
cloud_provider.project	GCP project ID
cloud_provider.kube_cluster	Kubernetes cluster name
cloud_provider.region	GCP region
cloud_provider.zone	GCP zone (alternative to region)
cloud_provider.helm_registry	OCI Helm registry (GCP Artifact Registry)
helm_values_path	Path to Helm values file
helm_chart	Helm chart name or OCI reference
helm_chart_version	Helm chart version
helm_version	Helm binary version (for local installs)
use_system_helm	Use system Helm binary (default: true)

Versioning

rdeploy uses semantic versioning for releases. The git-release command automates tagging:

# Patch release: 1.2.3 → 1.2.4
rdeploy git-release patch

# Minor release: 1.2.3 → 1.3.0
rdeploy git-release minor

# Major release: 1.2.3 → 2.0.0
rdeploy git-release major

# Pre-release: 1.2.3 → 1.2.4-rc.1
rdeploy git-release pre-patch

# Skip confirmation prompt
rdeploy git-release patch --force

Development

# Clone and install in development mode
git clone https://github.com/rehive/rdeploy.git
cd rdeploy
python -m venv .venv
source .venv/bin/activate
pip install -e ".[dev]"

# Run tests
pytest

# Run with coverage
coverage run -m pytest
coverage report

Requirements

• Python 3.9+
• kubectl configured with cluster access
• gcloud CLI configured and authenticated
• Helm 3.x

Publishing to PyPI

rm -rf dist
python -m build
twine upload dist/*