Skip to content

settings.md

Latest commit

 

History

History
196 lines (144 loc) · 5.6 KB

settings.md

File metadata and controls

196 lines (144 loc) · 5.6 KB

Dremio AI Settings Configuration

Default Configuration

The default configuration file is located at:

  • $HOME/.config/dremioai/config.yaml (Unix/macOS)
  • %XDG_CONFIG_HOME%/dremioai/config.yaml (if XDG_CONFIG_HOME is set)

You can view the default configuration (if it exists) and its location using the dremio-mcp-server config list command:

$ uv run dremio-mcp-server config list
Default config file: /home/user/.config/dremioai/config.yaml (exists = True)
dremio:
  uri: https://api.dremio.cloud
  pat: your-pat-here
  project_id: project123
  experimental_enabled: false
  allow_dml: false
tools:
  server_mode: FOR_SELF

To only show the configuration file path without displaying its contents:

$ dremio-mcp-server config list --show-filename
Default config file: /home/user/.config/dremioai/config.yaml (exists = True)

Overview

The dremioai.config.settings module provides a comprehensive configuration system for managing various aspects of the Dremio AI tools and servers. It uses Pydantic for robust validation and type checking of configuration values.

Configuration Structure

Base Settings

Section Description
dremio Dremio connection settings
tools Tool-specific configurations
prometheus Prometheus integration settings
langchain LangChain framework settings
beeai BeeAI framework settings

Configuration Sections

Dremio Settings

dremio:
  uri: <string|DremioCloudUri> # Dremio instance URI
  pat: <string> # Personal Access Token
  project_id: <string> # Optional: Project ID for Dremio Cloud
  enable_search: <bool> # Optional: Enable semantic search
  allow_dml: <bool> # Optional: Allow MCP Server to create views in Dremio
  wlm:  # Optional: WLM settings, for running MCP server in stdio mode
    engine_name: <string> # Optional: Direct all SQL to this engine if set

URI can be specified as:

PAT can be provided:

  • Directly as a string
  • As a file path prefixed with '@' (e.g., "@~/tokens/dremio.token")

Tools Settings

tools:
  server_mode: <string|ToolType|int> # Tool types to enable

Server modes:

  • FOR_SELF: Dremio cluster introspection
  • FOR_PROMETHEUS: Prometheus integration
  • FOR_DATA_PATTERNS: Data pattern analysis
  • EXPERIMENTAL: Experimental features

NOTE: if you intend to use experimental features you must also set experimental_enabled: true as shown above.

Multiple modes can be combined using comma separation: FOR_SELF,FOR_PROMETHEUS

Prometheus Settings

prometheus:
  uri: <string|HttpUrl> # Prometheus server URI
  token: <string> # Authentication token

LangChain Settings (Experimental)

The Tools defined in the tools section can be used with LangChain if required as well. The LangChain integration depends on openai or ollama for LLM. The [src/dremoiai/servers/frameworks/langchain/server.py] supplies a command line interface showing the integration.

langchain:
  llm: <Model> # LLM type (ollama/openai)
  openai:
    api_key: <string> # OpenAI API key
    model: <string> # Model name (default: gpt-4)
    org: <string> # Optional: Organization ID
  ollama:
    model: <string> # Model name (default: llama3.1)

BeeAI Settings (Experimental)

BeeAI framework supports MCP as well as multiple LLM providers. The [src/dremoiai/servers/frameworks/beeai/server.py] supplies a command line interface showing this integration.

beeai:
  mcp_server:
    command: "uv"
    args:
    - "run"
    - "--directory"
    - "<toplevel git directory>"
    - "dremio-mcp-server"
    - "run"
  sliding_memory_size: <int> # Memory window size
  anthropic:
    api_key: <string> # Anthropic API key
    chat_model: <string> # Chat model name
  openai: <OpenAI> # OpenAI settings (same as LangChain)
  ollama: <Ollama> # Ollama settings (same as LangChain)

Configuration Methods

File-based Configuration

Default configuration location: ~/.config/dremioai/config.yaml

Example:

dremio:
  uri: "https://api.dremio.cloud"
  pat: "@~/tokens/dremio.pat"
  project_id: "project123" # required only for DC
  enable_search: <bool> # Optional: Enable semantic search features
  allow_dml: <bool> # Optional: Allow MCP Server to create views in Dremio
tools:
  server_mode: "FOR_SELF,FOR_DATA_PATTERNS"

Environment Variables

Settings can be configured using environment variables with nested delimiter '_':

DREMIOAI_DREMIO__URI="https://api.dremio.cloud"
DREMIOAI_DREMIO__PAT="your-pat-here"
DREMIOAI_TOOLS__SERVER_MODE="FOR_SELF"

Programmatic Configuration

from dremioai.config import settings

# Load from file
settings.configure("path/to/config.yaml")

# Get current settings
current = settings.instance()

# Override settings
current.with_overrides({
    "dremio.uri": "https://new-uri.com",
    "tools.server_mode": "FOR_SELF"
})

Validation

  • All settings are validated using Pydantic models
  • Type checking and conversion is performed automatically
  • Custom validators are implemented for special types (URIs, file paths, etc.)
  • Invalid configurations will raise appropriate validation errors

Security Notes

  • Token files should have appropriate permissions
  • Environment files (.env) should be properly secured
  • Avoid committing sensitive configuration to version control