Docker Networking: How Containers Talk to Each Other

In a modern application, a React Frontend needs to talk to a Node.js API, which needs to talk to a MongoDB Database. If these were three separate physical servers, you would need cables. In Docker, we use Virtual Networks.

1. The "Gated Community" Analogy

Think of Docker Networking like a Gated Community:

• The Containers: These are the houses.
• The Bridge Network: This is the private internal road inside the community. Houses can talk to each other easily using their "Names" (DNS).
• Port Mapping: This is the "Main Gate." If someone from the outside world (the Internet) wants to visit, they must come through a specific gate number.

2. The Three Main Network Drivers

Docker provides different "Drivers" depending on how much isolation you need:

A. Bridge Network (The Default)

The most common choice for CodeHarborHub projects. It creates a private space on your computer.

• Key Feature: Containers can talk to each other using their Container Name.
• Usage: docker run --network my-bridge-net ...

B. Host Network

The container shares the exact same IP and Ports as your actual laptop/server. There is no isolation.

• Key Feature: Best for high-performance apps, but dangerous because ports can clash.

C. None

The container has no network access at all. Total "Air-gap" security.

3. Port Mapping: Opening the Gates

By default, an app running inside a container on port 3000 is invisible to your browser. You must "Map" a port from your Host to the Container.

The Formula: -p [Host_Port]:[Container_Port]

docker run -p 8080:3000 my-web-app

• Host Port (8080): What you type in your browser (localhost:8080).
• Container Port (3000): Where the app is actually listening inside the "box."

4. Automatic Service Discovery

This is the "Magic" of Docker. If you put two containers on the same user-defined network, you don't need to know their IP addresses!

Docker acts as a mini DNS Server. It knows that api-server belongs to the Backend container's IP address. This means you can write your code using friendly names instead of hard-coding IPs, which can change every time you restart the containers.

Essential Networking Commands

Command	Action
docker network ls	See all virtual networks on your machine.
docker network create hub-net	Create a new private road for your apps.
docker network inspect hub-net	See which containers are currently connected.
docker network connect hub-net app1	Add an existing container to a network.

The Math of Conflict

If you try to run two containers mapping to the same Host Port, Docker will throw an error:

$Error = \text{Host\_Port\_8080} \in \text{Already\_In\_Use}$

Solution: Map them to different host ports:

• Container A: -p 8080:80
• Container B: -p 8081:80

Summary Checklist

• I understand that Bridge is the default network for isolated apps.
• I can explain Port Mapping (-p host:container).
• I know that containers on the same network can talk using Names instead of IPs.
• I understand that localhost inside a container refers to the container itself, not my laptop.

Note

Never use the default "bridge" network for production. Always create your own named network (docker network create ...). Only named networks allow containers to look each other up by name!