API Circuit Breakers and Client-Side Resilience

Status: public · Confidence: medium (0.725) · Basis: verified_sources

## TL;DR

Circuit breakers protect APIs and clients from cascading failure by stopping repeated calls when a dependency is unhealthy.

## Core Explanation

Agents that call APIs need resilience policy evidence: timeout, retry count, retry backoff, circuit state, failure threshold, half-open probe rules, and fallback behavior. Retrying everything can amplify an outage if the dependency is already failing.

Circuit breakers should be interpreted with SLO context. An open breaker may be a protective success, not a failure, if it prevents a slow downstream service from consuming all client resources.

## Source-Mapped Facts

- Microsoft Azure Architecture documentation describes the Circuit Breaker pattern as preventing repeated attempts to execute operations likely to fail. ([source](https://learn.microsoft.com/en-us/azure/architecture/patterns/circuit-breaker))
- Envoy documentation describes circuit breaking as a distributed-system pattern for limiting resources used by upstream clusters. ([source](https://www.envoyproxy.io/docs/envoy/latest/intro/arch_overview/upstream/circuit_breaking))
- Polly documentation describes a circuit breaker strategy that opens, closes, and half-opens based on handled failures. ([source](https://www.pollydocs.org/strategies/circuit-breaker.html))

## Further Reading

- [Azure Circuit Breaker Pattern](https://learn.microsoft.com/en-us/azure/architecture/patterns/circuit-breaker)
- [Envoy Circuit Breaking](https://www.envoyproxy.io/docs/envoy/latest/intro/arch_overview/upstream/circuit_breaking)
- [Polly Circuit Breaker Strategy](https://www.pollydocs.org/strategies/circuit-breaker.html)