Using b with Docker

b provides official Docker images that make it easy to use binary management in containerized environments without installing b locally.

Docker Images

The official Docker images are available at ghcr.io/fentas/b with support for both AMD64 and ARM64 architectures.

Quick Usage

# List available binaries
docker run --rm -v ./.bin:/.bin ghcr.io/fentas/b list

# Install binaries using Docker
docker run --rm -v ./.bin:/.bin ghcr.io/fentas/b install jq kubectl

# Install b itself and other tools
docker run --rm -v ./.bin:/.bin ghcr.io/fentas/b install b jq kubectl

Volume Mounting

The key to using b with Docker is properly mounting your binary directory:

• Host path: ./.bin (your local binary directory)
• Container path: /.bin (where b expects to find/install binaries)

# Mount current directory's .bin folder
docker run --rm -v ./.bin:/.bin ghcr.io/fentas/b install terraform

# Use absolute paths if needed
docker run --rm -v /path/to/project/.bin:/.bin ghcr.io/fentas/b list

Environment Variables

You can also use environment variables to control where b installs binaries:

# Set PATH_BIN to specify installation directory
docker run --rm -v ./.bin:/usr/local/bin -e PATH_BIN=/usr/local/bin ghcr.io/fentas/b install helm

# Set PATH_BASE for project-relative installation
docker run --rm -v ./:/app -e PATH_BASE=/app ghcr.io/fentas/b install kubectl

Using in Your Own Dockerfiles

Multi-stage Copy

Copy the b binary from the official image into your own containers:

FROM alpine:latest

# Copy the b binary from the official image
COPY --from=ghcr.io/fentas/b:latest /b /usr/local/bin/b

# Set up environment for b to work
ENV PATH_BIN=/usr/local/bin

# Install binaries during build
RUN b install jq kubectl helm

# Your application code
COPY . /app
WORKDIR /app

CI/CD Integration

Use b in your CI/CD pipelines with Docker:

# .github/workflows/ci.yml
name: CI
on: [push, pull_request]

jobs:
  test:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      
      - name: Install project tools with b
        run: |
          docker run --rm -v ./.bin:/.bin ghcr.io/fentas/b install jq kubectl
          
      - name: Add tools to PATH
        run: echo "${{ github.workspace }}/.bin" >> $GITHUB_PATH
        
      - name: Use installed tools
        run: |
          kubectl version --client
          jq --version

Docker Compose

Integrate b into your Docker Compose workflows:

# docker-compose.yml
version: '3.8'
services:
  tools:
    image: ghcr.io/fentas/b:latest
    volumes:
      - ./.bin:/.bin
    command: install kubectl helm terraform
    
  app:
    build: .
    depends_on:
      - tools
    volumes:
      - ./.bin:/usr/local/bin
    environment:
      - PATH=/usr/local/bin:$PATH

Available Tags

• ghcr.io/fentas/b:latest - Latest stable version (multi-arch)
• ghcr.io/fentas/b:v4.3.0 - Specific version (multi-arch)
• ghcr.io/fentas/b:latest-amd64 - AMD64 specific
• ghcr.io/fentas/b:latest-arm64 - ARM64 specific

Best Practices

1. Use Volume Mounts

Always mount your binary directory to persist installed tools:

# ✅ Good - tools persist after container exits
docker run --rm -v ./.bin:/.bin ghcr.io/fentas/b install jq

# ❌ Bad - tools are lost when container exits
docker run --rm ghcr.io/fentas/b install jq

2. Pin Versions in Production

Use specific image tags in production environments:

# ✅ Good - predictable builds
COPY --from=ghcr.io/fentas/b:v4.3.0 /b /usr/local/bin/b

# ❌ Avoid in production - may break with updates
COPY --from=ghcr.io/fentas/b:latest /b /usr/local/bin/b

3. Set Environment Variables

Explicitly set PATH_BIN when using custom installation directories:

ENV PATH_BIN=/usr/local/bin
RUN b install kubectl

Troubleshooting

Permission Issues

If you encounter permission issues with volume mounts:

# Run with current user ID
docker run --rm --user $(id -u):$(id -g) -v ./.bin:/.bin ghcr.io/fentas/b install jq

Binary Not Found

Ensure your PATH includes the binary directory:

# Add .bin to PATH
export PATH="$PWD/.bin:$PATH"

# Or use absolute path
/path/to/project/.bin/kubectl version

Environment Configuration

If b fails to install, ensure environment variables are set:

# Check if PATH_BIN or PATH_BASE is needed
docker run --rm -e PATH_BIN=/.bin -v ./.bin:/.bin ghcr.io/fentas/b install jq