Dockerignore

Creating a .dockerignore file

To create a .dockerignore file:

1. Create a new file named .dockerignore in the root directory of your project (the same directory that contains your Dockerfile)
2. Add patterns for files and directories you want to exclude, one per line

Here’s a simple example of a .dockerignore file:

# Exclude node modules
node_modules

# Exclude any git files
.git
.gitignore

# Exclude log files
*.log

# Exclude the README
README.md

.dockerignore syntax

The .dockerignore file uses pattern matching similar to .gitignore. Here’s what you need to know:

Basic patterns

# Comment line (ignored)
file.txt           # Ignores file.txt in any directory
path/to/file.txt   # Ignores specific file
*.log              # Ignores all files with .log extension
**/*.log           # Ignores .log files in any subdirectory
temp?              # ? matches a single character (temp1, tempA, etc.)

Directory patterns

node_modules/      # Ignores the node_modules directory and all its contents
build/             # Ignores the build directory and all its contents

Negation patterns

You can use ! to negate a pattern, meaning Docker will include a file that would otherwise be excluded:

# Ignore all .md files
*.md

# But include important-notes.md
!important-notes.md

Path patterns

# Matches directories and files named temp at any level
**/temp

# Matches files or directories in the root directory only
/temp

# Matches any file or directory named temp or starting with temp
temp*

Practical examples

Let’s look at some real-world examples for different types of projects:

Node.js project

# Development dependencies
node_modules
npm-debug.log
yarn-debug.log
yarn-error.log

# Environment and configuration
.env
.env.local
.env.development
.env.test
.env.production

# Testing
coverage
.nyc_output

# Build outputs
dist
build

# Version control
.git
.gitignore

# IDE specific files
.idea
.vscode
*.swp
*.swo

# OS specific files
.DS_Store
Thumbs.db

---

Python project

# Python bytecode
__pycache__/
*.py[cod]
*$py.class

# Virtual environments
venv/
env/
ENV/
.env

# Testing
.pytest_cache/
.coverage
htmlcov/

# Distribution / packaging
dist/
build/
*.egg-info/

# Documentation
docs/_build/

# Version control
.git
.gitignore

# IDE specific files
.idea/
.vscode/

---

Java project

# Compiled class files
*.class
target/
build/

# Package files
*.jar
*.war
*.ear

# Maven
.mvn/
mvnw
mvnw.cmd

# Gradle
.gradle/
gradlew
gradlew.bat

# Testing
test-output/

# Logs
*.log

# Version control
.git/
.gitignore

# IDE specific files
.idea/
.vscode/
*.iml
.classpath
.project
.settings/

Example: Complete workflow

Let’s walk through a complete example of setting up a .dockerignore file for a Node.js web application:

1. Project structure:

my-node-app/
├── .git/
├── node_modules/
├── src/
│   ├── index.js
│   └── components/
├── public/
│   ├── images/
│   └── styles/
├── tests/
├── .env
├── .env.example
├── package.json
├── package-lock.json
└── Dockerfile

2. Create a .dockerignore file:

# Version control
.git
.gitignore

# Dependencies
node_modules

# Environment variables (except examples)
.env
.env.*
!.env.example

# Testing
tests
coverage

# Build artifacts
build
dist

# Logs
logs
*.log

# OS files
.DS_Store
Thumbs.db

# Editor directories
.idea
.vscode
*.swp

3. Dockerfile:

FROM node:14-alpine

WORKDIR /app

# Copy package files first (for better caching)
COPY package*.json ./

RUN npm install --production

# Copy the rest of the application
COPY . .

EXPOSE 3000

CMD ["node", "src/index.js"]

4. Build the Docker image:

docker build -t my-node-app .