spark-edge-db — Database & Persistence Layer

This package manages the local database connection, schema definitions, and migration logic for Spark Edge.

It uses Drizzle ORM and SQLite (via @libsql/client) to store application state, configurations, command queues, and monitoring metrics locally.

🏗️ Architecture

• Database Engine: SQLite running in WAL (Write-Ahead Logging) mode for robust local persistence.
• ORM: Drizzle ORM for type-safe queries and schema definition.
• Migrations: Automatically executed or resolved during application startup or migration setup, with customizable path resolution using the SPARK_EDGE_MIGRATIONS_DIR environment variable.

📡 Schemas & Entities

The package defines the following entities in src/db/schemas.ts:

• edge_settings: Stores global configurations (e.g. edge name, cloud broker credentials).
• monitored_servers: Configured servers that this Spark Edge instance is monitoring.
• metrics: Historical resource utilization logs (CPU, memory, disk).
• commands: Pending and executed remote commands sent by the Spark Cloud.
• failed_mqtt_messages: Local message fallback queue for offline message buffering.

🚀 Usage

This package is consumed by the main spark-edge package to handle local state.

Initializing and Querying

import { db, schemas } from 'spark-edge-db';
import { eq } from 'drizzle-orm';

// Query edge settings
const settings = await db.select().from(schemas.edgeSettings).limit(1);

Running Migrations Programmatically

import { runMigrations } from 'spark-edge-db';

await runMigrations();