abdm64 / nest-graph-ql

📝 Table of Contents

• About

• Getting Started

• Folder Structure

• Development

  • App Architecture
  • API Architecture Explained
  • Running the API

• Deployment

  • Deployment Architecture
  • Deployment Architecture Explained
  • Docker
  • Kubernetes GKE

• Built Using

• Author

🧐 About

• This graphql api is connected to Postgres database to retrieve data .

🏁 Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes. See deployment for notes on how to deploy the project on a live Kubernetes system.

Prerequisites

You need to install the fellowing software in order to get the application up and running :

• Node.js and npm.
• NestJS.
• Postgres database.
• Docker.
• Kubernetes on cloud.

Development

APP Architecture

• 

APP Architecture Explained

The application architecture contain 2 part

• Simple GraphQl API that connect to Postgres database with define tables using TypeORM as interface to retrieve information

Installing

• Download and Install node.js and NPM from https://nodejs.org/en/download/

Running the Nest.JS application

• Download or clone the project code from https://github.com/abdm64/nest-graph-ql

• install NestJs Framework from the npm cli for more information please visit NestJS Web site

• Install all required npm packages by running npm install from the command line in the api directory (where the package.json is located).

cd api

npm install

• Before start the application you should change the keys in the api/src/config directory (host, user, password and db name ) in order to connect to the postgres database.

• Start the application by running npm start from the command line in the api directory , you should see the message: "Nest application successfully started".

npm start

• Now you have graphQl api that connected to the database and you can make query ( please see the attached email ).

🚀 Deployment

Prerequisites

• In order to deploy this application in production we need :
  • Docker to build image for this application and push it to the docker hub or private registry .
  • Kubernetes cluster to run the application in production mode from the image that was created .

Docker

Installing

• Download and install docker on your machine Please Visit Docker

Build Docker image

• Build your own docker image and push it to your repo by running "docker build -t my-app-name:v1 . " from the command line in api directory

cd api

docker build -t my-app-name:v1 .

you need to push the image to Docker hub or your own private registry .

Run Docker image

• Run your Docker image for the application by the command line

docker run -e [inject your env variable here] my-app-name:v1

PS: you need to run the docker image and attached it to a running Postgres image

Docker compose

• Also you can run the application connected to a database as service by running with pg admin to manage the Postgres database at port 5050 you can change it in docker-compose file

docker-compose up

and the docker-compose.yaml handle all the task for you running the application with a database attached

• to drop the running container use the command

docker-compose  down

Kubernetes

Deployment Architecture

• 

Deployment Architecture Explained

• Please fellow the instruction in the app-dep.yaml file in the k8s folder to update all information needed ( env var ) in order to the application work on Kubernetes cluster

• create secret in the same namespace to store your pg password by running the Cammand

kubectl create secret generic <secret_name>  --from-literal  PGPASSWORD=<your password>

• To run the application on Kubernetes cluster just run the fellowing command

kubectl apply -f k8s

this will create the fellowing Kubernetes objects:

• Namespace a virtual cluster for all your resource related to this application ( pods services secret)
• Deployment for the application with one pod ( running container) insuring high availability for that service.
• Deployment for the Postgres database in one pod ( not the perfect solution for stateful application )
• PVC to store the data from the database
• Cluster ip service that connected to the pod .
• ingress service that connect the cluster ip service with ingress-nginx load balancer to expose it outside (Public) .

PS : you must install NGINX Ingress Controller on your k8s cluster before applying the final deployment please check this link https://kubernetes.github.io/ingress-nginx/deploy/ for more information.

• to drop the application just run the command :

kubectl delete  -f k8s

this will delete all kubernetes objects

⛏️ Built Using

• NestJS - A progressive Node.js framework.
• Postgres - Open source relationel database.
• TypeORM - promise-based Node.js ORM for Postgres, MySQL, MariaDB, SQLite and Microsoft SQL Server.
• GraphQL query language for APIs.
• Docker - Software platform for building and packaging applications.
• Kubernetes - Container Orchestration.

✍️ Author

• @abdm64 Backend | DevOps Engineer @ Djezzy

Made with ❤️ by Abdellah