Local Kubernetes Deployment with Kind

This directory contains declarative Kubernetes manifests for deploying the Personal Lifecycle Tracker stack to a local Kubernetes cluster using Kind (Kubernetes in Docker), providing containerized multi-service deployment with persistent volumes and health probes.

1. Kubernetes Architecture & Components

Database: PostgreSQL 16 containerized deployment backed by a 1Gi PersistentVolumeClaim (postgres-pvc) on port 5432 with pg_isready health probes.
Backend Service: FastAPI backend (personal-lifecycle-backend:latest) configured with container probes targeting /health on port 8000.
Frontend / Ingress Gateway: Caddy 2 reverse proxy serving the static React 19 SPA and reverse proxying API traffic, exposed via NodePort 30088.
Kustomize Composition: Managed declaratively through k8s/kustomization.yaml.

2. Step-by-Step Deployment Instructions

Step 1: Create a Local Kind Cluster

# Using Makefile
make kind-up

# Or directly with Kind
kind create cluster --name lifecycle-cluster

Step 2: Build & Load Container Images

Build images and load them into the Kind control-plane node without pushing to a remote registry:

# Build images
docker build -t personal-lifecycle-backend:latest backend/
docker build -t personal-lifecycle-frontend:latest -f frontend/Dockerfile --target prod frontend/

# Load into Kind cluster
kind load docker-image personal-lifecycle-backend:latest --name lifecycle-cluster
kind load docker-image personal-lifecycle-frontend:latest --name lifecycle-cluster

Step 3: Apply Kubernetes Manifests

# Using Makefile
make kind-deploy

# Or directly with kubectl
kubectl apply -k k8s/

Step 4: Verify Pod Readiness & Rollout

# Check rollout status
kubectl rollout status deployment/lifecycle-postgres
kubectl rollout status deployment/lifecycle-backend
kubectl rollout status deployment/lifecycle-frontend

# Inspect running pods
kubectl get pods -o wide

Step 5: Port-Forward & Validate Health

# Port-forward frontend service
kubectl port-forward svc/lifecycle-frontend 8088:80 &

# Verify live healthcheck
curl -f http://localhost:8088/health

Step 6: Teardown Kind Cluster

make kind-down
# Or: kind delete cluster --name lifecycle-cluster

3. Zero-Cost Kubernetes to AWS Serverless Lambda Integration

To fulfill Kubernetes project criteria while maintaining $0.00 / month cost on AWS (avoiding the ~$73/mo mandatory Amazon EKS control-plane fee):

Kubernetes-to-Lambda Invocation Bridge (k8s/lambda-bridge-job.yaml):
- Deploys an automated Kubernetes Job inside the local cluster that invokes the Serverless AWS Lambda Function URL, performs health validations, and bridges Kubernetes workloads with serverless backend APIs.
AWS Controllers for Kubernetes (ACK) Declarative Manifest (k8s/ack-lambda-resource.yaml):
- Demonstrates declarative Kubernetes management of AWS Lambda functions via the lambda.services.k8s.aws/v1alpha1 CustomResource.
- Allows Kubernetes control plane manifests to manage, configure, and monitor AWS serverless infrastructure with 0 EC2 worker nodes and 0 cloud cluster charges.