Deploying Docker to a Virtual Private Server
Taking your application live is one of the most exciting parts of software development. While managed cloud platforms (like Heroku or Vercel) are easy to use, they can become incredibly expensive as your traffic grows. Learning how to deploy your Docker containers directly to a Virtual Private Server (VPS)—such as DigitalOcean, Linode, or AWS EC2—gives you complete architectural control and keeps your hosting costs completely predictable.
- How to securely connect to your Linux server over the internet using SSH.
- The three-stage lifecycle of a Docker deployment (Build, Push, Pull).
- How to fetch your compiled application image from a Container Registry.
- How to run your container in the cloud and map its ports so the public can access it.
1. Understanding the VPS Deployment Workflow
Before typing any commands, you need to understand how your code travels from your laptop to the public internet. You never write code directly on the production server. Instead, you use a three-step pipeline.
| Environment | Primary Role | Common Command |
|---|---|---|
| Local Machine | Write code, build the Docker image, and upload it. | docker push |
| Container Registry | Safely store your built images (e.g., Docker Hub, AWS ECR). | (Storage only) |
| Cloud VPS | Download the image and run the application 24/7. | docker pull & docker run |
2. Connecting to Your Server via SSH
To control your VPS, you need to open a secure remote terminal session. We do this using a protocol called Secure Shell (SSH). When you create a VPS on a provider like DigitalOcean, they will give you a public IP address and a default username (usually root or ubuntu).
Opening the Remote Connection
Example 1: SSH into your Linux server
# Open your local terminal and type:
ssh root@203.0.113.50
Expected output
The authenticity of host '203.0.113.50' can't be established. Are you sure you want to continue connecting (yes/no)? yes Welcome to Ubuntu 22.04.3 LTS (GNU/Linux 5.15.0-82-generic x86_64) root@webapp-production:~#
3. The Architecture & Execution Flow
Once you are logged into your server, the server acts as a blank slate. It does not have your code yet. The server must reach out to your Container Registry, grab the image, and spin it up into a running container that listens for user traffic.
The Docker Deployment Lifecycle
Your local machine pushes the built image to a registry. The production VPS pulls that exact image and runs it, exposing it to the internet.
4. Step-by-Step Execution Breakdown
Assuming you have already installed Docker on your new Linux server, you are ready to pull your application and expose it to the world. We will use a standard web application as an example.
Run
docker pull yourusername/yourapp:latest. This tells your VPS to reach out to Docker Hub, find your repository, and download the exact compiled container image to the server's hard drive.Run
docker run -d --name my-live-app -p 80:8080 yourusername/yourapp:latest. The -d flag is crucial. It means "detached." It tells Docker to run the container in the background. If you omit this, your app will immediately crash and turn off the moment you close your SSH terminal window.The
-p 80:8080 flag bridges the server's network to the container's network. The number on the left (80) is the public port on the VPS (Port 80 is the standard for HTTP web traffic). The number on the right (8080) is the internal port your code is actually listening on inside the container.5. Hands-On Practice
Testing these commands gives you the muscle memory needed for smooth production deployments.
You deployed your app, but you want to verify it is actually running and didn't crash on startup.
Hint: Use docker ps to see a list of all active, background containers.
Your web app is running in detached mode, but users are reporting an error. You need to see the application's terminal output to find the stack trace.
Expected result: You should use docker logs my-live-app to print the standard output of the background container to your SSH session.
Your app needs a database connection string to work in production, but you didn't hardcode it into the image for security reasons.
Hint: Use the -e flag during the docker run command to inject environment variables, like this: docker run -d -p 80:8080 -e DB_PASS="supersecret" yourapp:latest.
6. Interview Questions
If you are applying for DevOps, Backend, or Full-Stack roles, interviewers will heavily test your knowledge of how containers behave on real infrastructure.
Answer: docker run creates a brand new container from an image and starts it. docker start simply wakes up and resumes an existing container that was previously stopped. You usually use "run" for deployments, and "start/stop" for maintenance.
Answer: In enterprise systems, you rarely expose a Docker container directly to the public internet. Instead, you run the container on an internal port (like 8080), and place a Reverse Proxy (like NGINX, Traefik, or an AWS Application Load Balancer) in front of it on Port 80/443. The proxy handles SSL/TLS certificates, load balancing, and security filtering, while Docker just focuses on running the application logic.
Conclusion
Deploying Docker to a VPS bridges the gap between writing code on your laptop and serving real users globally. By mastering SSH, understanding the detached run mode, and properly mapping your network ports, you have taken full control of your infrastructure. This approach not only saves significant money compared to managed cloud providers, but it also builds the foundational system administration skills required for senior engineering roles.
Official Documentation
- Docker Documentation – Docker Run Reference
- DigitalOcean Reference – How to Connect to Droplets with SSH
Upgradix by Shrash Studio – Learn Technology. Build Your Future.