Author: Abbas Aziz

Date: February 23, 2026

Tech Stack: Kubernetes (Kind), Docker, Python (FastAPI), PostgreSQL, Prometheus, Grafana, Helm, WSL2.

Overview:

This project demonstrates the deployment of a full-stack containerized application into a local Kubernetes cluster, focusing on overcoming resource constraints and resolving complex service discovery issues in an observability pipeline.

The Mission

Deploy a resilient, monitored API that tracks real-time traffic and database health using industry-standard SRE (Site Reliability Engineering) practices.

Architecture Overview

• Infrastructure: 3-Node Kubernetes Cluster (1 Control Plane, 2 Workers) running via Kind (Kubernetes in Docker).
• Application: FastAPI (Python 3.10) with custom Prometheus middleware.
• Database: PostgreSQL 15 deployed as a StatefulSet for data persistence.
• Monitoring: Prometheus Operator (Kube-Prometheus-Stack) managing the metrics lifecycle via ServiceMonitors.

Infrastructure Setup

1. Cluster Configuration (Kind)

To simulate a multi-node production environment locally, I configured a 3-node Kind cluster.

YAML

`# kind-config.yaml kind: Cluster apiVersion: kind.x-k8s.io/v1alpha4 name: ai-cluster nodes:

• role: control-plane
• role: worker
• role: worker`

Command: