@log-ingestor/mssql

📘 Article: Logging Node.js Applications Directly into MSSQL — Without ELK or Kafka
https://medium.com/@aritvyas/logging-node-js-applications-directly-into-mssql-without-elk-or-kafka-57d668b54077

Disk-first Node.js logger with automatic ingestion into Microsoft SQL Server — no ELK, no log management headaches.

⚠️ MSSQL database name is mandatory in the connection string. This package is the MSSQL adapter for the log-ingestor ecosystem.
It writes structured logs to disk and reliably ingests them into Microsoft SQL Server using a high‑performance background ingestor.

Why this exists

Most Node.js applications:

• Print logs to console
• Rotate files manually
• Lose logs on crashes
• Overpay for ELK

This solves that with a simple model:

App → disk → background ingestor → MSSQL

No Kafka. No ELK. No ops pain.

Features

• Structured JSON logging
• Disk-first (crash-safe)
• Automatic background ingestion
• Batch inserts with retry
• Auto table creation
• One ingestor per process
• Production-ready defaults

Installation

npm install @log-ingestor/mssql

Basic Usage

const { createLogger } = require("@log-ingestor/mssql");

const logger = createLogger({
  service: {
    name: "order-service",
    version: "1.0.0"
  },
  logging: {
    logDir: "./logs"
  },
  db: {
    type: "mssql",
    connection:
      "Server=localhost,1433;" +
      "Database=logsdb;" +
      "User Id=sa;" +
      "Password=***;" +
      "Encrypt=true;" +
      "TrustServerCertificate=true;"
  },
  batch: {
    size: 200
  }
});

logger.info("order created", { orderId: 123 });
logger.error("payment failed", { reason: "timeout" });

That’s it.
The ingestor starts automatically in the background.

How it works

1. Logs are written as JSON lines to disk
2. A background ingestor binary starts automatically
3. Logs are batched and inserted into MSSQL
4. Retries happen automatically on failure

Your app never blocks on DB writes.

Configuration

Logger options

{
  service: {
    name: string,
    version?: string
  },
  logging: {
    logDir?: string // default: ./logs
  },
  db: {
    type: "mssql",
    connection: string,
    table?: string // default: unified_logs
  },
  batch?: {
    size?: number        // default: 200
    maxRetries?: number  // default: 3
  }
}

Output format (example)

{
  "logId": "uuid",
  "timestamp": "2026-01-05T10:20:30.000Z",
  "level": "INFO",
  "service": {
    "name": "order-service",
    "version": "1.0.0"
  },
  "message": "order created",
  "data": {
    "orderId": 123
  }
}

Internal components

• @log-ingestor/core – shared logger logic
• @log-ingestor/mssql – MSSQL ingestion adapter
• Ingestor binary – high-performance background worker

What this is NOT

• Not a console logger
• Not a replacement for application metrics
• Not an ELK stack

This is for durable, queryable application logs.

Roadmap

• PostgreSQL adapter
• MySQL adapter
• MongoDB adapter
• DynamoDB adapter

License

MIT