Circuit Breaker

Intent

Prevent cascading failures by failing fast when a dependency is unhealthy instead of repeatedly sending requests that are likely to fail or time out.

Use when

• A downstream dependency becomes flaky/unavailable and causes retries/timeouts to pile up.
• You want to protect your own resources (threads, pools, event loop) under partial failure.

Avoid / watch-outs

• Breakers need a recovery/probing strategy (half-open) and a UX/fallback plan.
• Avoid hiding persistent errors; surface stable error codes and make behavior observable.

Skill mapping

• resilience: configure breaker thresholds, time budgets, and fallbacks; expose breaker metrics.
• observability: log breaker state transitions and correlate with dependency errors/latency.
• testing: simulate dependency failure and assert breaker open/close behavior where appropriate.