gRPC Server

Laurus includes a built-in gRPC server that keeps the search engine resident in memory, eliminating the per-command startup overhead of the CLI. This is the recommended way to run Laurus in production or when integrating with other services.

Features

• Persistent engine — The index stays open across requests; no WAL replay on every call
• Full gRPC API — Index management, document CRUD, commit, and search (unary + streaming)
• Health checking — Standard health check endpoint for load balancers and orchestrators
• Graceful shutdown — Pending changes are committed automatically on Ctrl+C / SIGINT
• TOML configuration — Optional config file with CLI override support

Quick Start

# Start the server with default settings
laurus serve

# Start with a custom data directory and port
laurus --data-dir ./my_index serve --port 8080

# Start with a configuration file
laurus serve --config config.toml

Sections

• Getting Started — Installation, startup options, and configuration
• gRPC API Reference — Full API documentation for all services and RPCs