QuestionQ92

VerifiedApplying site reliability engineering practices

You manage a microservice that exposes a public-facing API (Service A). Service A is time-critical and has a response SLO of 500 ms. Service A makes synchronous calls to an internal API (Service B), which is known to become unreliable under heavy load, causing connection timeout errors or 500 errors. Service B collects request information for the transactions handled by Service A.

You need to reduce the impact of Service B problems on Service A users. What should you do?

Explanation

A circuit breaker prevents an unreliable downstream service from causing cascading failures or consuming Service A’s response-time budget. When Service B is unhealthy, a fallback that stores the request data in a database preserves the information for later processing without making Service A users wait on failing synchronous calls. Google Cloud recommends circuit breakers and graceful degradation to reduce traffic to failing or overloaded services and prevent cascading failures.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!