Dell Isilon (PowerScale) MCP Server
Enables AI assistants to interact with Dell PowerScale (Isilon) OneFS clusters through over 2,400 tools generated from the REST API specification. It supports managing storage protocols, snapshots, networking, and cluster configurations using natural language.
README
Isilon MCP Server
A professional-grade Model Context Protocol (MCP) server for Dell PowerScale (Isilon) OneFS API integration. This server enables AI assistants like Claude to interact with PowerScale storage clusters through a comprehensive set of tools generated from the OneFS 9.7 REST API specification.
Features
- 🚀 Complete API Coverage: 2,434 operations across 46 categories
- 🔐 Secure Authentication: Basic Auth with TLS support
- 📦 Dynamic Tool Generation: Auto-generates MCP tools from OpenAPI spec
- 🔄 Multiple Transports: stdio (Claude Desktop) and HTTP/SSE (n8n)
- 📝 Comprehensive Logging: Structured JSON and colored console output
- ⚡ Async/Await: Built on httpx for high-performance async operations
- 🛡️ Credential-Free Mode: Explore available tools without credentials
Supported API Categories
| Category | Operations | Description |
|---|---|---|
| Auth | 373 | Authentication, users, groups, roles, providers |
| Protocols | 351 | SMB, NFS, HDFS, S3, HTTP, FTP, Swift |
| Sync | 185 | SyncIQ replication policies, reports, jobs |
| Cluster | 167 | Cluster management, nodes, configuration |
| Network | 153 | Networking, subnets, pools, interfaces |
| Datamover | 151 | Data migration, accounts, policies |
| Cloud | 136 | CloudPools, tiering, accounts |
| Snapshot | 117 | Snapshot schedules, aliases, locks |
| Upgrade | 85 | Cluster upgrade management |
| Event | 83 | Event channels, alerts, settings |
| Quota | 78 | SmartQuotas management |
| Job | 72 | Job engine, policies, reports |
| ... | ... | And 34 more categories! |
Quick Start
Prerequisites
- Python 3.10 or higher
- Access to a Dell PowerScale cluster with REST API enabled
- API credentials with appropriate permissions
Installation
From PyPI (Recommended)
# Install the package
pip install isilon-mcp-server
# Or install with HTTP server support for n8n
pip install "isilon-mcp-server[http]"
From Source
# Clone the repository
git clone https://github.com/sachdev27/isilon-mcp-server.git
cd isilon-mcp-server
# Create virtual environment
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# Install the package
pip install -e .
# Or install with HTTP server support for n8n
pip install -e ".[http]"
Configuration
- Copy the example environment file:
cp .env.example .env
- Edit
.envwith your PowerScale credentials:
ISILON_HOST=your-powerscale-host
ISILON_USERNAME=admin
ISILON_PASSWORD=your-secure-password
ISILON_PORT=8080
ISILON_TLS_VERIFY=true
Running the Server
stdio Transport (for Claude Desktop)
# Using the CLI
isilon-mcp
# Or using Python module
python -m isilon_mcp.main
HTTP/SSE Transport (for n8n or web clients)
# Start HTTP server
python -m isilon_mcp.http_server
# With custom host/port
python -m isilon_mcp.http_server --host 0.0.0.0 --port 8000
Claude Desktop Integration
Add to your Claude Desktop configuration (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": {
"isilon": {
"command": "/path/to/venv/bin/python",
"args": ["-m", "isilon_mcp.main"],
"cwd": "/path/to/isilon-mcp-server",
"env": {
"ISILON_HOST": "your-powerscale-host",
"ISILON_USERNAME": "admin",
"ISILON_PASSWORD": "your-password"
}
}
}
}
n8n Integration
- Start the HTTP server:
python -m isilon_mcp.http_server --host 0.0.0.0 --port 8000
- In n8n, add an MCP Client node pointing to:
http://localhost:8000/sse
Example Usage
Once connected, you can ask Claude to:
- "List all SMB shares on the cluster"
- "Show me the cluster health status"
- "Create a new NFS export for /ifs/data/project1"
- "Get all snapshots older than 30 days"
- "Show active SyncIQ policies"
- "List users in the local provider"
Project Structure
isilon-mcp-server/
├── isilon_mcp/
│ ├── __init__.py # Package exports
│ ├── api_client.py # Async HTTP client with Basic Auth
│ ├── config.py # Pydantic configuration models
│ ├── exceptions.py # Exception hierarchy
│ ├── http_server.py # HTTP/SSE transport for n8n
│ ├── logging_config.py # Structured logging
│ ├── main.py # CLI entry point (stdio)
│ ├── server.py # Core MCP server
│ └── tool_generator.py # OpenAPI to MCP tool generator
├── powerscale_9.7_comprehensive_openapi.json # Full API spec
├── pyproject.toml # Package configuration
├── requirements-mcp.txt # Dependencies
├── .env.example # Configuration template
└── README.md # This file
Development
Running Tests
# Install dev dependencies
pip install -e ".[dev]"
# Run tests
pytest
# With coverage
pytest --cov=isilon_mcp --cov-report=html
Code Quality
# Format code
black isilon_mcp/
isort isilon_mcp/
# Lint
ruff check isilon_mcp/
# Type checking
mypy isilon_mcp/
API Reference
Tool Naming Convention
Tools are named using the pattern:
{category}_{operationId}
Examples:
auth_list_providers- List authentication providersprotocols_list_smb_shares- List SMB sharessnapshot_create_snapshot- Create a new snapshotcluster_get_cluster_config- Get cluster configuration
Authentication
The server uses HTTP Basic Authentication with your PowerScale credentials. All requests are made over HTTPS by default.
Error Handling
The server provides detailed error messages for:
- Authentication failures
- Connection errors
- Rate limiting
- API validation errors
Security Considerations
- Never commit
.envfiles - They contain sensitive credentials - Use TLS verification - Set
ISILON_TLS_VERIFY=truein production - Principle of least privilege - Use API users with minimal required permissions
- Read-only mode - Set
READ_ONLY_MODE=truefor monitoring-only access
Troubleshooting
Connection Issues
# Test connection manually
curl -k -u username:password https://your-host:8080/platform/18/cluster/config
Certificate Errors
If using self-signed certificates:
ISILON_TLS_VERIFY=false # Only for testing!
Debug Mode
Enable detailed logging:
LOG_LEVEL=DEBUG
DEBUG_HTTP=true
License
MIT License - see LICENSE for details.
Contributing
Contributions are welcome! Please read our contributing guidelines and submit pull requests.
Acknowledgments
- Dell PowerScale for the comprehensive REST API
- Model Context Protocol for the MCP specification
- Anthropic Claude for AI assistant integration
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。