MSA에서 분산 추적은 장애 원인을 어떻게 찾을까요?

In a microservices environment, a single user request passes through multiple services. When the request flow is fragmented—such as through a gateway, order service, payment service, and inventory service—it is difficult to pinpoint exactly where the slowdown occurred or where a failure happened solely from logs. This is where distributed tracing comes in as a technology that tracks the entire journey of a request. To use a simple analogy, it is similar to a shipping tracking number. Just as a single tracking number allows you to identify which hubs a package has passed through and where it has been held for a long time, distributed tracing displays the services a request has passed through and the processing time as a single flow based on a Trace ID. In this video, we summarize the core concepts of distributed tracing and observability. We explain the roles of logs, metrics, and traces, as well as Trace and Span, context propagation, the W3C traceparent standard, OpenTelemetry, Spring Boot automatic instrumentation, custom spans, asynchronous context propagation, sampling strategies, the correlation between logs and traces, and even the cardinality pitfalls, all from a practical perspective. 📌 Recommended for: ✅ Backend developers who have struggled to trace the cause of failures in microservice environments ✅ Those curious about the working principles of OpenTelemetry and distributed tracing ✅ Those who want to understand observability by viewing logs, metrics, and traces together ✅ Those who want to apply Trace ID, custom spans, and sampling strategies in Spring Boot #DistributedTrace #Observability #OpenTelemetry #Microservices #BackendDevelopment