http-echo • CI/CD + Kubernetes (Kind) Demo

This repository contains a tiny Go web server http-echo and a complete PR-based CI/CD flow using GitHub Actions, Kind (Kubernetes in Docker), ingress-nginx, and k6 for load testing.

What you get

• Go app that echoes text (/app/http-echo).
• Docker image (distroless) built from a statically linked binary.
• K8s manifests for two sample apps: foo and bar, each running http-echo.
• Ingress with hosts foo.localhost and bar.localhost.
• GitHub Actions workflow that:
  • Creates a Kind cluster
  • Builds the app and a Docker image
  • Loads the image into Kind
  • Deploys Deployments + Services + Ingress
  • Installs ingress-nginx so Ingress works on localhost
  • Runs k6 load tests and posts results to the PR

---

Prerequisites (for local runs)

• Docker (with buildx)
• Go 1.21+ (for building locally)
• kubectl
• Kind
• Helm
• k6
• jq

On GitHub Actions these are installed automatically by the workflow.

---

Project structure

devOps-basic/
├─ .github/workflows/load-test-ci.yml # Main PR CI/CD workflow
├─ app/http-echo/                     # Go app
├─ k8s/
│  ├─ deployments.yaml                # Two Deployments: foo, bar (image placeholder)
│  ├─ services.yaml                   # Services for foo and bar (added)
│  └─ ingress.yaml                    # Ingress for foo.localhost / bar.localhost
├─ loadtest/k6-script.js              # Basic k6 load test
├─ scripts/wait-for-ingress.sh        # Waits for ingress to be reachable
└─ k8s/kind-config.yaml               # Kind cluster config (port mappings added)

---

How the workflow works

The workflow runs on Pull Requests to main when changes touch app/**, k8s/**, scripts/**, loadtest/**, or .github/workflows/**.

High level steps:

1. Checkout repo.
2. Install tools: kind, kubectl, helm, k6, jq (and Go).
3. Create Kind cluster with extra port mappings (80/443) so ingress is reachable at localhost.
4. Install ingress-nginx using the official Kind manifest and wait for readiness.
5. Build http-echo binary to dist/linux/amd64/http-echo via make -C app/http-echo OS=linux ARCH=amd64 bin.
6. Build Docker image with build args and tag http-echo:pr-<PR_NUMBER>.
7. Load image into Kind.
8. Deploy:
   • Replace IMAGE_PLACEHOLDER in k8s/deployments.yaml with http-echo:pr-<PR_NUMBER> and apply
   • Apply k8s/services.yaml
   • Apply k8s/ingress.yaml
9. Wait for ingress (foo.localhost and bar.localhost) to become reachable.
10. Run k6 load test, compute metrics (avg, p90, p95, req/s), and post a PR comment.
11. (Optional) Install Prometheus stack.
12. Cleanup Kind cluster.

---

Run end-to-end locally (optional)

# 1) Create cluster
kind create cluster --config k8s/kind-config.yaml

# 2) Install ingress-nginx (Kind)
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/main/deploy/static/provider/kind/deploy.yaml
kubectl wait --namespace ingress-nginx   --for=condition=Ready pods   --selector=app.kubernetes.io/component=controller   --timeout=180s

# 3) Build app binary to dist/
make -C app/http-echo OS=linux ARCH=amd64 bin

# 4) Build Docker image
PRODUCT_VERSION=$(cat app/http-echo/version/VERSION)
docker build --platform linux/amd64   --build-arg BIN_NAME=http-echo   --build-arg PRODUCT_VERSION=${PRODUCT_VERSION}   -t http-echo:dev app/http-echo

# 5) Load into Kind and deploy
kind load docker-image http-echo:dev --name kind
sed "s|IMAGE_PLACEHOLDER|http-echo:dev|g" k8s/deployments.yaml | kubectl apply -f -
kubectl apply -f k8s/services.yaml
kubectl apply -f k8s/ingress.yaml

# 6) Test
bash scripts/wait-for-ingress.sh foo.localhost 120
bash scripts/wait-for-ingress.sh bar.localhost 120
curl -s http://foo.localhost/
curl -s http://bar.localhost/

---

Common pitfalls

• Make sure Docker Desktop (or daemon) is running; Kind needs it.
• If port 80/443 on your host are in use, change k8s/kind-config.yaml to map to different hostPorts and update the README commands accordingly.

---

Sample PR Comment Output

License

http-echo is © HashiCorp, MPL-2.0.