pipeline-example

example for jamf docker desktop run in kind kind create cluster --name local-test

kubectl get pods

🧩 Basic API Pipeline Example

A lightweight Go web service with automated Kubernetes deployment using Kind, Helm, and GitHub Actions.

---

📘 Overview

This repository demonstrates a full CI/CD workflow for a simple Go-based API deployed to a local Kind (Kubernetes in Docker) cluster.
The GitHub Actions pipeline handles building, pushing, deploying, and verifying your application end-to-end — making it easy to reproduce locally or debug via tmate.

---

What's it do?

A minimal Go API running on port 8080:

func main() {
    http.HandleFunc("/", func(w http.ResponseWriter, r *http.Request) {
        fmt.Fprint(w, "Hello World")
    })
    http.ListenAndServe(":8080", nil)
}

When running, http://localhost:8080

⚙️ Prerequisites

Install these tools locally to mirror the CI environment: Tool - Version - Install Go ≥ 1.22 - golang.org/dl Docker ≥ 24 - docs.docker.com (I use Docker Desktop) Kubectl ≥ 1.30 - kubernetes.io Helm ≥ 3.14 - helm.sh Kind ≥ 0.23 - kind.sigs.k8s.io

Running the App Locally

Run locally

Pull the repo down and cd into it then run it locally to verify it works

go run main.go

The included workflow automates the full build → deploy → verify cycle.

Make sure Kubernetes is enabled in Docker

📄 .github/workflows/ci.yml Pipeline steps: Build & Test Go App Build & Push Docker Image — Builds and pushes to Docker Hub. Create Kind Cluster — Spins up a temporary Kubernetes cluster inside GitHub Actions. Export kubeconfig — Ensures Helm and kubectl point to the correct cluster. Deploy Helm Chart Verify API Response — Uses kubectl port-forward + curl to confirm Hello World. (Optional) action-tmate — Opens a live debugging session for interactive testing. link

🔐 Required GitHub Infp Name Description DOCKERHUB_USERNAME - Your Docker Hub username (new variable) DOCKERHUB_TOKEN - Docker Hub access token for docker push (new secret) Public SSH Key - Optional, for tMate validation

Trigger Workflow

1. Make a change to a file and push it up to Github. Create a new PR and the workflow will kick off.

   • When the workflow gets to the 'deploy-and-validate' stage, click into the workflow details.

   • Inside, wait for 'Start tMate Session' to start

     • Click in, and wait for a series of ssh code to show up, should look similar to: 🔹 How to Connect When the workflow reaches the tmate step, it prints something like: SSH: ssh abcdefgh@nyc1.tmate.io

       SSH URL: Connect from your local terminal: ssh abcdefgh@nyc1.tmate.io

   • 🔹 Once Inside tmate Check your pods: kubectl get pods

   Run: kubectl port-forward deployment/pipeline-example 9091:8080 & Wait a couple of seconds, then run: curl http://localhost:9091

   You should see a 'Hello World' returned to you. To close the instance, run exit. This will also finish your pipeline. You the pipeline will continue to run until you cancel it you dont use the tMate option.

🧹 Cleanup To reset your environment: kind delete cluster --name local-test docker image rm USERNAME/pipeline_example:latest

🔍 Troubleshooting Issue - Cause - Fix port already in use - Existing forward active pkill -f "kubectl port-forward"

Structure

├── main.go
├── Dockerfile
├── pipeline-example/
│   └─
│       ├── Chart.yaml
│       ├── templates/
│       │   ├── deployment.yaml
│       │   └── service.yaml
│       └── values.yaml
└── .github/
    └── workflows/
        └── ci.yml

🏁 Summary

This project is a simple, reproducible template for: Building and deploying a Go API to Kubernetes. Managing deployments with Helm. Automating CI/CD with GitHub Actions. Debugging live clusters with action-tmate.