Skip to content

README.md

Latest commit

 

History

History
551 lines (314 loc) · 11 KB

README.md

File metadata and controls

551 lines (314 loc) · 11 KB

Dev8.dev Docker Development Environment# Dev8.dev Docker Workspace

Complete containerized development environment with VS Code Server, AI tools, and language support.Cloud development container with VS Code Server, multiple languages, and AI coding tools.

Quick Start## 🏗️ Architecture

bash

1. Configure environment00-base → 10-languages → 20-vscode → 30-ai-tools (FINAL)

cp .env.example .env```

Edit .env and set your GITHUB_TOKEN

Layers:

2. Build and start (one command)1. 00-base - Ubuntu 22.04, SSH, system packages, SDKMAN, Homebrew

make up2. 10-languages - Node.js, Python, Go, Rust

  1. 20-vscode - VS Code Server (code-server)

3. Access VS Code4. 30-ai-tools - GitHub Copilot CLI, AI tool setup scripts

Open browser: http://localhost:8080

No password required by default## 🚀 Quick Start


### Build All Layers

**That's it!** You now have a complete development environment.

```bash

---cd docker



## Contents# Build all layers in correct order (REQUIRED)

make build-all

- [Architecture](#architecture)```

- [Configuration](#configuration)

- [Usage](#usage)**Note:** You must use `make build-all` because the layers must be built sequentially. Plain `docker compose build` won't work due to layer dependencies.

- [Production Deployment](#production-deployment)

- [Advanced Topics](#advanced-topics)### Run Locally



---```bash

# Start the workspace

## Architecturemake up



### Layer Structure# Or use docker compose directly

docker compose up -d workspace

00-base → Ubuntu + essential tools + dev8 user# Access:

10-languages → Go, Python, Node.js, Rust# - VS Code: http://localhost:8080 (password: dev8dev)

20-vscode → code-server (VS Code in browser)# - SSH: ssh -p 2222 dev8@localhost

30-ai-tools → GitHub Copilot, AI CLI tools


make logs

Each layer builds on the previous, allowing efficient caching.

# Stop

### Servicesmake down

  • VS Code Server (port 8080): Browser-based VS Code

  • SSH (port 2222): Optional shell access### Environment Variables

  • Supervisor API (port 9000): Workspace monitoring

Create a .env file in the docker/ directory:

Volumes

  • dev8-data: Single persistent volume containing:
    • User home directory (/home/dev8) - configs, packages, extensions
    • Workspace subdirectory (/home/dev8/workspace) - your code and projects
# Required



---# Optional

CODE_SERVER_PASSWORD=your_password

## ConfigurationANTHROPIC_API_KEY=sk-ant-xxx...

OPENAI_API_KEY=sk-xxx...

### Environment VariablesGEMINI_API_KEY=xxx...

Required:


GITHUB_TOKEN=ghp_xxxxx  # GitHub token for Copilot

```Everything in `/home/dev8` persists across container restarts:



**Optional:**```

```bash/home/dev8/

# Authentication (default: none)├── .sdkman/           # Java, Kotlin, Scala (sdk install ...)

CODE_SERVER_AUTH=none           # or 'password'├── .linuxbrew/        # Ruby, PostgreSQL (brew install ...)

CODE_SERVER_PASSWORD=           # set if auth=password├── .npm/              # Node packages (npm install -g ...)

├── .local/            # Python packages (pip install --user ...)

# SSH (optional)├── .cargo/            # Rust packages (cargo install ...)

SSH_PASSWORD=                   # simple password auth└── .vscode-server/    # VS Code extensions

SSH_PUBLIC_KEY=                 # or public key auth```



# Git identity**Examples:**

GIT_USER_NAME=Your Name```bash

GIT_USER_EMAIL=you@example.com# Install Java

sdk install java 17.0.8-amzn

# AI API keys (optional)

ANTHROPIC_API_KEY=# Install Ruby

OPENAI_API_KEY=brew install ruby

GEMINI_API_KEY=

```# Install Node package

npm install -g typescript

### Security Notes

# Install Python package

- **Default**: No authentication (suitable for local development)pip install --user pytest

- **Production**: Set `CODE_SERVER_AUTH=password` and strong password

- **SSH**: Not required for VS Code access; only enable if needed# All persist in /home/dev8 volume!

🤖 AI Coding Tools

Usage

GitHub Copilot CLI

Development (Local)```bash

gh copilot suggest "create a REST API in Node.js"


# Start workspace```

make up

### AI APIs (if configured)

# View logs- Anthropic Claude API

make logs- OpenAI API

- Google Gemini API

# Stop workspace

make downSetup scripts in `/usr/local/share/ai-tools/`



# Rebuild after changes## � Services

make rebuild

```| Service | Port | Purpose |

|---------|------|---------|

### Accessing the Workspace| VS Code Server | 8080 | Web-based VS Code |

| SSH Server | 2222 | Terminal access |

**VS Code (Primary):**| Workspace Supervisor | 9000 | Monitoring |

- URL: http://localhost:8080

- No password required (default)## � Directory Structure



**SSH (Optional):**```

```bashdocker/

# If SSH_PASSWORD is set├── images/

ssh -p 2222 dev8@localhost│   ├── 00-base/              # Base system + SDKMAN + Homebrew

│   ├── 10-languages/         # Node, Python, Go, Rust

# If SSH_PUBLIC_KEY is set│   ├── 20-vscode/            # VS Code Server

ssh -p 2222 -i ~/.ssh/your_key dev8@localhost│   └── 30-ai-tools/          # AI CLI tools

```├── shared/

│   ├── config/               # Shared configs

### Common Tasks│   └── scripts/              # Helper scripts

├── docker-compose.yml        # Local development

**View running containers:**├── Makefile                  # Build shortcuts

```bash└── README.md                 # This file

docker compose ps```

🛠️ Development

Execute command in container:


docker exec -it dev8-workspace bash

``````bash

make build-base

**Check container health:**make build-languages

```bashmake build-vscode

docker compose logs workspace | tail -50make build-ai-tools

---### Test

Production Deployment```bash

Start container

Azure Container Instances (ACI)docker compose up -d workspace

Quick Deploy:# Check services


# 1. Configure production environment

cp .env.prod.example .env.prod# View logs

# Edit .env.prod with your Azure settingsdocker compose logs -f workspace

2. Build and push image

make prod-build### Clean Up

make prod-push

# 3. Deploy to ACIdocker compose down -v  # Remove containers and volumes

make prod-deploymake clean              # Remove images

4. Check status

make prod-status## � Documentation


- **ARCHITECTURE.md** - Architecture decisions and design

### Production Configuration- **CONTAINER_CAPABILITIES.md** - What the container can do

- **CHANGELOG.md** - Version history

**Key differences from development:**

- Use pre-built images (no build context)## 📄 License

- Azure Files for persistence

- Resource limits enforcedPart of Dev8.dev - See [LICENSE](../LICENSE)

- Authentication recommended

**Environment setup:**
```bash
# Required
ENVIRONMENT_ID=dev8-prod-001
GITHUB_TOKEN=ghp_xxxxx
CONTAINER_REGISTRY=yourregistry.azurecr.io
AZURE_STORAGE_ACCOUNT=youraccount
AZURE_STORAGE_KEY=xxxxx

# Recommended
CODE_SERVER_AUTH=password
CODE_SERVER_PASSWORD=$(openssl rand -base64 32)

Deploy script:

./deploy-to-aci.sh

See Production Deployment Details below.

Advanced Topics

Manual Layer Building

# Build individual layers (in order)
docker compose build base
docker compose build languages
docker compose build vscode
docker compose build workspace

Custom Configuration

Disable authentication:

CODE_SERVER_AUTH=none

Enable password protection:

CODE_SERVER_AUTH=password
CODE_SERVER_PASSWORD=your_secure_password

SSH with password:

SSH_PASSWORD=your_ssh_password

SSH with public key:

SSH_PUBLIC_KEY="ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAA..."

Troubleshooting

Container won't start:

# Check logs
docker compose logs workspace

# Check environment
docker exec dev8-workspace env

# Restart
make rebuild

Port conflicts:

# Change ports in docker-compose.yml
ports:
  - "8081:8080"  # VS Code on 8081 instead

Volume issues:

# List volumes
docker volume ls

# Inspect volume
docker volume inspect dev8-workspace

# Clean rebuild
make clean
make up

Production Deployment Details

Prerequisites

  1. Azure Container Registry (ACR)

    az acr create --name yourregistry --resource-group yourgroup --sku Basic

  2. Azure Storage Account

    az storage account create --name yourstorage --resource-group yourgroup

  3. Azure File Share

    # Create single file share for all persistent data (home + workspace)
az storage share create --name dev8-data --account-name yourstorage --quota 200

Deployment Process

1. Build production image:

docker build -t yourregistry.azurecr.io/dev8-workspace:latest -f images/30-ai-tools/Dockerfile .

2. Push to registry:

az acr login --name yourregistry
docker push yourregistry.azurecr.io/dev8-workspace:latest

3. Deploy using script:

./deploy-to-aci.sh

Or manually:

az container create \
  --resource-group yourgroup \
  --name dev8-workspace \
  --image yourregistry.azurecr.io/dev8-workspace:latest \
  --cpu 2 --memory 4 \
  --ports 8080 2222 9000 \
  --environment-variables \
    GITHUB_TOKEN=ghp_xxxxx \
    CODE_SERVER_AUTH=password \
    CODE_SERVER_PASSWORD=xxxxx

Production Checklist

  • Set strong passwords (use openssl rand -base64 32)
  • Configure Azure Files for persistence
  • Set up backup strategy
  • Configure monitoring/logging
  • Test failover scenarios
  • Document access procedures
  • Set resource limits
  • Enable HTTPS (use Azure Application Gateway)

Monitoring

# Check container status
az container show --resource-group yourgroup --name dev8-workspace

# View logs
az container logs --resource-group yourgroup --name dev8-workspace

# Restart container
az container restart --resource-group yourgroup --name dev8-workspace

📚 Documentation

File Reference

  • docker-compose.yml: Local development configuration
  • docker-compose.prod.yml: Production (ACI) configuration
  • .env.example: Development environment template
  • .env.prod.example: Production environment template
  • deploy-to-aci.sh: Automated ACI deployment script
  • Makefile: Common commands and shortcuts

License

See LICENSE for details.