🚀 Automated Docker + NGINX Deployment Script (deploy.sh)

Author: John Durodola

Language: Bash

---

🧭 Overview

This project provides a robust, production-grade Bash script (deploy.sh) that automates the setup, deployment, and configuration of a Dockerized web application on a remote Linux server, complete with NGINX reverse proxy, error handling, validation, and logging.

It is designed for repeatable, idempotent deployments, suitable for real-world DevOps workflows or infrastructure automation tasks.

---

🎯 Objectives

By the end of this project, you should be able to:

• Automate the deployment of a Dockerized app onto a remote Linux server.
• Handle SSH-based communication and file transfer securely.
• Configure NGINX as a reverse proxy to expose the app publicly.
• Implement logging, validation, and error handling best practices in Bash.
• Understand the foundations of server provisioning, container orchestration, and infrastructure automation.

---

🧩 Features

✅ Collects user input interactively and validates all parameters. ✅ Supports both Dockerfile and docker-compose.yml-based projects. ✅ Automatically installs Docker, Docker Compose, and NGINX if missing. ✅ Configures NGINX reverse proxy dynamically for your app. ✅ Idempotent: Safe to re-run without breaking existing setups. ✅ Logs every action with timestamps and clear success/failure indicators. ✅ Supports --cleanup flag to remove all deployed resources gracefully. ✅ Fully self-contained — no dependencies beyond bash, git, rsync, and ssh.

---

⚙️ Requirements

Component	Purpose	Example
Local machine	Runs the script	macOS / Linux
Remote server	Target deployment host	Ubuntu 20.04+
SSH access	Required for remote commands	Key-based (recommended)
Docker	Application runtime	Installed automatically if missing
NGINX	Reverse proxy	Installed automatically
Git	To clone project repo	Must be installed locally

---

🧰 Installation

1️⃣ Clone this script repository

git clone https://github.com/<your-username>/deploy-script.git
cd deploy-script

2️⃣ Make the script executable

chmod +x deploy.sh

---

🚀 Usage

✅ To deploy your app:

Run the script interactively:

./deploy.sh

You’ll be prompted for:

Prompt	Description
Git repo URL	URL of your app’s repository
Branch	Optional (defaults to main)
Remote username	SSH username on target server
Server IP	Public IP or domain of remote host
SSH key path	Path to private SSH key (default: ~/.ssh/id_rsa)
App port	Internal container port (default: 80)

Once complete, your app will be accessible at:

http://<your-server-ip>

---

🧹 Cleanup Mode

To remove deployed containers, files, and NGINX configs:

./deploy.sh --cleanup

This will:

• Stop and remove Docker containers.
• Delete project files from the remote server.
• Remove NGINX site configurations.
• Reload NGINX cleanly.

---

📦 What the Script Does (Step-by-Step)

Step	Description
1. Parameter Collection	Prompts user for repo, branch, server, key, and port with validation.
2. Repository Setup	Clones or updates the repository using git.
3. Docker Validation	Ensures presence of Dockerfile or docker-compose.yml.
4. SSH Validation	Tests remote connection using ssh dry-run.
5. Environment Setup	Installs docker, docker-compose, and nginx remotely.
6. File Sync	Transfers project files securely with rsync.
7. Docker Deployment	Builds and runs containers using docker or docker-compose.
8. NGINX Proxy Config	Creates reverse proxy config and reloads NGINX.
9. Verification	Checks Docker containers and logs final status.
10. Logging	Saves all actions to deploy_YYYYMMDD_HHMMSS.log.
11. Cleanup (optional)	Removes containers, files, and configs when --cleanup is used.

---

🧾 Logging

• Every run generates a timestamped log file:

  deploy_20251022_143501.log
  

• All actions are logged with [INFO], [WARN], and [ERROR] tags.

• You can review logs to debug or audit your deployments.

---

⚠️ Error Handling

The script uses:

• set -Eeuo pipefail → catches all unhandled errors.

• trap → logs the failing line and exits gracefully.

• Validation checks for:

  • Missing or invalid SSH keys
  • Missing required binaries (git, ssh, rsync)
  • Unreachable hosts
  • Missing Docker configuration files

Each stage exits with a clear error message and a distinct exit code.

---

🔁 Idempotency

The script is safe to re-run. It will:

• Pull latest repo updates if the folder already exists.
• Stop and remove old containers before rebuilding.
• Avoid duplicating NGINX configs.
• Only re-deploy changed files.

This ensures consistent, repeatable deployments every time.

---

🌐 Example Walkthrough

$ ./deploy.sh
=== Automated Docker + NGINX Deployment ===
Git repo URL: https://github.com/example/myapp.git
Branch (default: main): main
Remote username: ubuntu
Remote server IP: 18.223.45.67
SSH key path (default: ~/.ssh/id_rsa):
Application port (default: 80): 5000
[INFO]  Validating inputs...
[INFO]  SSH connection successful.
[INFO]  Preparing remote environment...
[INFO]  Deploying application...
[INFO]  Configuring NGINX...
[INFO]  Verifying Docker container status...
[INFO]  Deployment completed successfully!
App accessible at: http://18.223.45.67
Full log saved in: deploy_20251022_143501.log

---

🧩 Directory Structure

deploy-script/
│
├── deploy.sh             # Main Bash script
├── README.md             # Documentation
└── deploy_YYYYMMDD.log   # Auto-generated logs

---

🔒 Security Notes

• SSH keys are never copied or exposed — only used locally.
• PATs (Personal Access Tokens) are not stored; if added, use them cautiously.
• Always prefer key-based SSH authentication over passwords.
• For production, integrate Certbot or real SSL instead of HTTP-only NGINX.

---