mcvsphere/README_DOCKER.md
Ryan Malloy eb59cd5e9a rename project: esxi-mcp-server → mcvsphere
mcvsphere = Model Control for vSphere

Updates:
- Package renamed from esxi_mcp_server to mcvsphere
- CLI entry point: mcvsphere (was esxi-mcp-server)
- All imports and references updated
- Docker configs updated
- Test suites updated
2025-12-26 20:58:48 -07:00

286 lines
5.7 KiB
Markdown

# mcvsphere - Docker Guide
This guide provides instructions for running the mcvsphere using Docker and Docker Compose.
## Quick Start
### Prerequisites
- Docker 20.10+
- Docker Compose 2.0+
- Access to a VMware vCenter Server or ESXi host
### 1. Setup
```bash
# Clone the repository
git clone <repository-url>
cd mcvsphere
# Create necessary directories and configuration
make setup
# Create environment variables file (optional)
make env-example
cp .env.example .env
```
### 2. Configuration
You have two options for configuration:
#### Option A: Configuration File (Recommended)
Edit `config/config.yaml`:
```yaml
vcenter_host: "your-vcenter-ip"
vcenter_user: "administrator@vsphere.local"
vcenter_password: "your-password"
datacenter: "your-datacenter"
cluster: "your-cluster"
datastore: "your-datastore"
network: "VM Network"
insecure: true
api_key: "your-api-key"
log_level: "INFO"
```
#### Option B: Environment Variables
Edit `.env` file:
```bash
VCENTER_HOST=your-vcenter-ip
VCENTER_USER=administrator@vsphere.local
VCENTER_PASSWORD=your-password
VCENTER_DATACENTER=your-datacenter
VCENTER_CLUSTER=your-cluster
VCENTER_DATASTORE=your-datastore
VCENTER_NETWORK=VM Network
VCENTER_INSECURE=true
MCP_API_KEY=your-api-key
MCP_LOG_LEVEL=INFO
```
### 3. Run the Server
```bash
# Build and run
make dev
# Or run in background
make run
# Check status
make status
# View logs
make logs
```
## Available Commands
Use `make help` to see all available commands:
```bash
make help
```
### Build Commands
- `make build` - Build Docker image
- `make build-no-cache` - Build without cache
### Run Commands
- `make run` - Run in background
- `make run-logs` - Run with logs
- `make stop` - Stop containers
- `make restart` - Restart containers
### Development Commands
- `make dev` - Development mode (build + run with logs)
- `make logs` - Show logs
- `make shell` - Open bash shell in container
- `make status` - Show container status
- `make health` - Check container health
### Maintenance Commands
- `make clean` - Remove containers and volumes
- `make clean-all` - Remove everything
- `make update` - Rebuild and restart
## Docker Architecture
### Multi-stage Build
The Dockerfile uses a multi-stage build process:
1. **Builder Stage**: Installs build dependencies and Python packages
2. **Production Stage**: Creates a minimal runtime image
### Security Features
- Runs as non-root user (`mcpuser`)
- Minimal base image (python:3.11-slim)
- Only necessary runtime dependencies
- Configurable resource limits
### Directory Structure
```
/app/
├── server.py # Main application
├── config.yaml.sample # Configuration template
├── docker-entrypoint.sh # Startup script
├── config/ # Configuration directory (mounted)
│ └── config.yaml # Runtime configuration
└── logs/ # Log directory (mounted)
└── vmware_mcp.log # Application logs
```
## Configuration Options
### Volume Mounts
- `./config.yaml:/app/config/config.yaml:ro` - Configuration file (read-only)
- `./logs:/app/logs` - Log directory
### Environment Variables
All configuration options can be set via environment variables:
| Variable | Description | Default |
|----------|-------------|---------|
| `VCENTER_HOST` | vCenter/ESXi hostname | Required |
| `VCENTER_USER` | Username | Required |
| `VCENTER_PASSWORD` | Password | Required |
| `VCENTER_DATACENTER` | Datacenter name | Auto-detect |
| `VCENTER_CLUSTER` | Cluster name | Auto-detect |
| `VCENTER_DATASTORE` | Datastore name | Auto-detect |
| `VCENTER_NETWORK` | Network name | VM Network |
| `VCENTER_INSECURE` | Skip SSL verification | true |
| `MCP_API_KEY` | API authentication key | None |
| `MCP_LOG_LEVEL` | Log level | INFO |
### Resource Limits
Default resource limits in docker-compose.yml:
- **Memory**: 512MB limit, 256MB reserved
- **CPU**: 0.5 cores limit, 0.25 cores reserved
## Health Checks
The container includes automatic health checks:
- **Interval**: 30 seconds
- **Timeout**: 10 seconds
- **Retries**: 3
- **Start Period**: 40 seconds
Check health manually:
```bash
make health
```
## Networking
The server exposes:
- **Port 8080**: HTTP API endpoint
- **Path `/sse`**: Server-Sent Events endpoint
- **Path `/sse/messages`**: JSON-RPC messages endpoint
## Troubleshooting
### Check Logs
```bash
make logs
```
### Check Container Status
```bash
make status
```
### Open Shell in Container
```bash
make shell
```
### Common Issues
1. **Configuration not found**: Ensure `config/config.yaml` exists or environment variables are set
2. **Permission denied**: Check that the `logs` directory is writable
3. **Connection failed**: Verify vCenter/ESXi connectivity and credentials
4. **Health check failed**: Check if the server is responding on port 8080
### Debug Mode
Run with debug logging:
```bash
# Set in .env file
MCP_LOG_LEVEL=DEBUG
# Or in config.yaml
log_level: "DEBUG"
```
## Production Deployment
### Security Recommendations
1. Use a dedicated user account for vCenter access
2. Enable API key authentication
3. Use valid SSL certificates (set `insecure: false`)
4. Limit container resources
5. Use Docker secrets for sensitive data
### High Availability
For production deployments, consider:
- Running multiple container instances
- Using a load balancer
- Implementing persistent storage for logs
- Setting up monitoring and alerting
## Examples
### Basic Usage
```bash
# Start the server
make run
# Check if it's working
curl http://localhost:8080/sse
```
### API Authentication
```bash
# With API key
curl -H "Authorization: Bearer your-api-key" http://localhost:8080/sse
```
### Development
```bash
# Development workflow
make build
make dev
# Make changes to code
# Rebuild and restart
make update
```