Circuit Breaker

A circuit breaker is a resilience pattern that wraps calls to a remote dependency and stops sending requests once too many of them fail. Instead of waiting on timeouts, callers get an immediate error or a fallback answer while the dependency recovers.

Why it matters

A slow or dead dependency is more dangerous than a fast failure. Threads pile up waiting, queues fill, and one broken payment API can take down your whole checkout. A breaker turns that slow failure into a fast, contained one and gives the struggling service room to recover instead of hammering it with retries.

Example

A breaker has three states:

1. Closed: requests flow normally and failures are counted.
2. Open: after, say, 5 failures in 30 seconds, calls fail instantly for a cooldown period such as 60 seconds.
3. Half-open: after the cooldown, a few trial requests are allowed. If they succeed the breaker closes; if they fail it opens again.

Your monitoring should alert when a breaker opens, because users are now seeing degraded behavior even though no server is technically down.

Common mistakes

• Setting the failure threshold so low that a handful of ordinary timeouts opens the breaker and causes an outage on its own.
• Opening the breaker with no fallback, so users just see errors faster.
• Not exposing breaker state as a metric, which hides the problem from dashboards until customers complain.
• Using one shared breaker for several unrelated endpoints, so one bad route blocks healthy ones.

Related

• Graceful degradation
• Retry with backoff
• Bulkhead
• Failover

यह पेज सामान्य जानकारी है, जिसे हमारी संपादकीय टीम ने लिखा है और दस्तावेज़ीकरण बदलने पर अपडेट किया जाता है। सर्वर और प्रदाता अलग-अलग होते हैं, इसलिए बदलाव पहले सुरक्षित तरीके से जाँचें। देखें: संपादकीय नीति और अस्वीकरण.