Kong in Kubernetes

This project seeks to provide an initial structure for developers to work with Kong API Gateway in a Kubernetes cluster. This structure can be used to design, develop, and deploy microservices architectures in a local development environment. Nonetheless, it is a fresh and minimal environment perfect for learning about Kong API Gateway.

Goals

• Local Kubernetes cluster: use Kind to deploy a single node cluster.
• Kong API Gateway with declarative configuration: deploy Kong OSS 2.6 with a database using Helm. The Admin API will be exposed through a ClusterIP, and a Pod containing decK will sync the local configuration file to Kong.
• Learn about Kong API Gateway: I believe that an API Gateway enables different kinds of architectures and Kong seems to offer a myriad of tools to both Free and Enterprise editions. Also, Kong allows developers to create custom plugins using Go, and I want to explore it.

Out of Scope

• Teach about Kubernetes: this document assumes readers have previous knowledge of the orchestrator.
• Production configuration: this project is meant to be used for local development and personal projects. Although I have plans to elaborate a production-ready Kong and Kuberenetes environment.
• Load Balancers: I will rely on NodePorts because I do not want to add complexity to my local environment.

Setup

Create the Kubernetes cluster using kind

kind create cluster --config kind.yaml

It will expose the ports 8000 and 8443 on the running machine that are going to forward the HTTP requests to Kong's proxy.

Launch the Postgres database in kind network

docker-compose up -d

Create Service for external Postgres database

kubectl apply -f postgres/

Install Kong in Kubernetes

helm repo add kong https://charts.konghq.com

helm repo update

helm install -n kong --create-namespace api-gateway kong/kong --set ingressController.installCRDs=false --values kong/values.yaml

Sync local configuration to Kong

./deck-sync

Finally, check http://localhost:8000/mock in your browser