degradation-circuit-breaking

What problem does it solve?

It prevents outages and cascading failures by designing bounded, observable degradation behavior when dependencies time out, error, or slow down, so core user journeys stay protected.

Core Features & Use Cases

• Timeouts and bounded retries: Enforces explicit connection and read timeouts and retries that are capped, jittered, and restricted to safe/idempotent operations.
• Circuit breakers and bulkheads: Configures failure thresholds to trip and recover safely while isolating downstream dependencies to avoid exhausting shared capacity.
• Fallback and fail-open vs fail-closed: Defines product-approved fallback behaviors (including stale-data limits) and explicit security/correctness decisions for each dependency.
• Chaos-tested, observable degradation: Ensures fallback paths are verified with fault injection and that circuit/fallback/timeout events are emitted as metrics and structured logs.

Quick Start

Use the degradation-circuit-breaking capability to produce a resilience plan for your service’s external dependencies, including timeouts, bounded retry rules, circuit breaker thresholds, bulkhead isolation, approved fallback behavior, and required observability signals.