🔒 Paid
Service Discovery Failure
Service A depends on Service C, which it can only reach through endpoints the service registry currently reports as healthy. The registry has begun handing out a mix of stale and genuinely unhealthy endpoints for Service C, and connection failures against it are climbing. Service A's calls to Service C must stay reliable without Service A simply hammering the same failing instances harder. Diagnose why discovery keeps producing bad endpoints and restore a durable, trustworthy path to Service C.
- Difficulty
- Senior
- Expected Time
- ~15 min
- Category
- Microservices & Dependencies
Skills
- Diagnosing stale service-discovery data
- Recognizing retry amplification against a fixed bad endpoint set
- Balancing discovery refresh speed against fallback routing
- Reasoning about health-check eviction policy tradeoffs
System Architecture
- API Gateway
- Service A
- Service B
- Service C
- Service Registry
This is a Pro scenario. Upgrade to Pro to unlock it.
Upgrade to ProBack to Incidents