logging-best-practices

Structured logging emits single, context-rich wide events per request to improve observability. By capturing high cardinality data, business context, and environment details in JSON format, wide events replace scattered logs to simplify debugging.

Implement structured logging by using a single logger instance and middleware patterns to capture request ID, user info, and business data. This approach ensures all context is emitted in one queryable JSON event per request.

Common application logging anti-patterns include emitting scattered logs across multiple statements and using unstructured strings. These practices hinder debugging by fragmenting context and preventing efficient querying of high cardinality data.

Using a single logger instance ensures consistent structured logging and prevents scattered logs. It allows middleware to reliably attach business context, environment characteristics, and request IDs to one wide event for better observability.

A canonical log event should include high cardinality fields, business context, request IDs, user information, and environment characteristics. Capturing this contextual data in a single JSON event prevents scattered logs and accelerates debugging.