/learning-docker

This repository is designed for learning and experimenting with Nginx as a reverse proxy using Docker and Docker Compose. It supports running a frontend, a Node.js backend, and MongoDB with a single command. Additionally, it includes instructions for setting up a local subdomain and configuring SSL.

Primary LanguageJavaScript

Learning Docker With Nginx

Dockerized Express Application with NGINX and MongoDB

This repository is designed for learning and experimenting with Nginx as a reverse proxy using Docker and Docker Compose. It supports running a frontend, a Node.js backend, and MongoDB with a single command. Additionally, it includes instructions for setting up a local subdomain and configuring SSL.

This repository contains a Dockerized setup for an Express.js application, an NGINX server, and a MongoDB database. It uses Docker Compose to manage multi-container applications efficiently.

Table of Contents

Prerequisites

Generate Self-Signed Certificates

To enable HTTPS, you need to generate a self-signed SSL certificate. You can do this using OpenSSL with the following command:

openssl req -nodes -new -x509 -keyout cert/server.key -out cert/server.cert -days 365

Update Hosts File

To access the application using the subdomain api.localhost, add the following entry to your hosts file:

For Windows

  1. Open Notepad as Administrator.
  2. Open the file located at C:\Windows\System32\drivers\etc\hosts.
  3. Add the following line: 127.0.0.1 api.localhost

For macOS/Linux

  1. Open a terminal.
  2. Run the following command: sudo nano /etc/hosts
  3. Add the following line: 127.0.0.1 api.localhost

Getting Started

  1. Clone the repository:

    git clone https://github.com/singh-govind/learning-docker.git

cd learning-docker

  2. Create a .env file in the root directory to set environment variables. Example:

    TZ=Asia/Kolkata

MONGO_URL=mongodb://mongodb/your-database // the second mongodb is name of the container which is running on same network bridge

HOST='0.0.0.0'

  3. Build and run the containers:

    • To build and run the containers: docker-compose up --build
    • To start multiple instances of the Express server (e.g., scaling to 3 instances): docker-compose up --build --scale express-app=3
    • To run the containers in detached mode (in the background): docker-compose up --build -d or docker-compose up --build --scale express-app=3 -d

  4. Access the application:

Directory Structure

learning-docker/
│
├── cert/                  # SSL certificates
│   ├── server.cert        # Self-signed SSL certificate
│   └── server.key         # Private key for the SSL certificate
|
├── index.html             # Static HTML file served by NGINX
├── index.js               # Main entry point for the Express application
├── rabbit-mq.js           # RabbitMQ setup for the Express server, acts as a message sender, runs in a Docker container
├── worker.js              # Message receiver, runs on a different machine or environment to process tasks received from the queue
├── package.json           # Node.js dependencies and scripts
├── package-lock.json      # Locked versions of dependencies
|
├── nginx/
│   ├── nginx.conf         # NGINX configuration file
│   └── logs/              # Logs directory for NGINX
|
├── Dockerfile              # Dockerfile for the Express app
├── docker-compose.yml      # Docker Compose configuration file
└── .env                    # Environment variables

Environment Variables

  • TZ: Set the timezone for your application, Asia/Kolkata.
  • MONGO_URL: Connection string for MongoDB, because in docker-compose its container name is mongodb thats why mongodb://mongodb/database-name.
  • HOST: keep it '0.0.0.0'.

Docker Commands

  • Build the images and start the containers: 
    docker-compose up --build
docker-compose up --build -d
docker-compose up --build --scale express-app=3
docker-compose up --build --scale express-app=3 -d
  • Stop the containers: docker-compose down
  • View logs: docker-compose logs
  • Remove dangling images and volumes: 
    docker image prune -f
docker volume prune -f

Note:

Start worker.js on host machine with below command for listening to the messages of queue sent from server.

node ./worker.js