Logging Strategy - Monitoring & Observability

Overview

Logging is not just about writing messages to a file. An effective logging strategy ensures you can debug production issues efficiently without overwhelming your storage.

Key Principles

Structured Logging: Use JSON format for machine-readability.
Context Enrichment: Include trace_id, user_id, and service_name in every log.
Log Levels: Use them appropriately (DEBUG, INFO, WARN, ERROR, FATAL).
Log Retention: Implement policies based on business needs and compliance (e.g., GDPR).

Example: Structured Log (JSON)

{
  "timestamp": "2026-04-10T10:00:00Z",
  "level": "ERROR",
  "service": "order-service",
  "trace_id": "a1b2c3d4e5f6",
  "message": "Database connection failed",
  "error_code": "DB_CONN_TIMEOUT"
}

Result: Tools like Loki or Elasticsearch can index these fields, allowing queries like: service="order-service" | json | error_code="DB_CONN_TIMEOUT"