In designing solutions for observability, which of the following practices is essential?

Question

Options

• ALimiting log collection to error logs only to reduce storage requirements
• BImplementing comprehensive logging across services and applications
• CRelying exclusively on manual monitoring and alerting
• DUsing custom metrics and logs only for high-severity incidents

• A
  4% (1)
• B
  77% (20)
• C
  15% (4)
• D
  4% (1)

Explanation

Comprehensive logging across all services and applications (B) is the foundation of observability because it gives you full visibility into system behavior, enabling faster diagnosis of issues, performance bottlenecks, and unexpected interactions between components. Limiting logs to errors only (A) is wrong because you lose the contextual trail - many bugs only become visible when you can trace the sequence of events leading up to an error. Manual monitoring (C) doesn't scale and introduces human latency into detection, defeating the purpose of observability. Restricting custom metrics to high-severity incidents only (D) means you lack baseline data, making it impossible to detect anomalies or distinguish normal from abnormal behavior.

Memory tip: Think of observability as a "flight recorder" - you need continuous, comprehensive data capture before something goes wrong, not just when alarms are already firing. The word comprehensive is your signal: observability means all services, all the time.

No community discussion yet for this question.