feat: 분산 트레이싱 컨텍스트 전파 구현

build-logic/src/main/kotlin/transentia/KafkaConfigConventionPlugin.kt

@@ -2,7 +2,9 @@ package transentia
 
 import org.gradle.api.Plugin
 import org.gradle.api.Project
+import org.gradle.kotlin.dsl.configure
 import org.gradle.kotlin.dsl.dependencies
+import org.jetbrains.kotlin.gradle.dsl.KotlinJvmProjectExtension
 
 class KafkaConfigConventionPlugin : Plugin<Project> {
     override fun apply(target: Project) = with(target) {
@@ -12,6 +14,12 @@ class KafkaConfigConventionPlugin : Plugin<Project> {
         pluginManager.apply("org.jetbrains.kotlin.kapt")
         pluginManager.apply("org.jetbrains.kotlin.plugin.allopen")
         pluginManager.apply("org.jetbrains.kotlin.plugin.spring")
+
+        // JVM 21 설정
+        extensions.configure<KotlinJvmProjectExtension> {
+            jvmToolchain(21)
+        }
+
         // 필요한 최소 의존성만 추가
         dependencies {
             add("implementation", "org.springframework.boot:spring-boot-starter-validation")

build-logic/src/main/kotlin/transentia/SpringBootAppConventionPlugin.kt

@@ -43,9 +43,10 @@ class SpringBootAppConventionPlugin : Plugin<Project> {
             add("implementation", "org.springframework.boot:spring-boot-starter-json")
             add("implementation", "org.jetbrains.kotlin:kotlin-reflect")
 
-            // Observability - Actuator + Prometheus
+            // Observability - Actuator + Prometheus + Tracing
             add("implementation", "org.springframework.boot:spring-boot-starter-actuator")
             add("implementation", "io.micrometer:micrometer-registry-prometheus")
+            add("implementation", "io.micrometer:micrometer-tracing-bridge-otel")
 
             add("testImplementation", "org.springframework.boot:spring-boot-starter-test")
         }

infrastructure/kafka/kafka-consumer/src/main/kotlin/io/github/hyungkishin/transentia/infrastructure/kafka/consumer/config/KafkaConsumerConfig.kt

@@ -101,6 +101,7 @@ class KafkaConsumerConfig<K : Serializable, V : SpecificRecordBase>(
         factory.containerProperties.apply {
             pollTimeout = kafkaConsumerConfigData.pollTimeoutMs
             ackMode = ContainerProperties.AckMode.MANUAL_IMMEDIATE
+            isObservationEnabled = true  // 트레이싱 활성화
         }
 
         return factory

infrastructure/kafka/kafka-producer/build.gradle.kts

@@ -7,4 +7,4 @@ dependencies {
     implementation(project(":kafka-config"))
     implementation("org.apache.avro:avro:1.11.4")
     implementation("io.confluent:kafka-avro-serializer:7.9.2")
-}
+}

infrastructure/kafka/kafka-producer/src/main/kotlin/io/github/hyungkishin/transentia/infrastructure/kafka/producer/KafkaProducerConfig.kt

@@ -41,6 +41,8 @@ class KafkaProducerConfig<K : Serializable, V : SpecificRecordBase>(
 
     @Bean
     fun kafkaTemplate(): KafkaTemplate<K, V> {
-        return KafkaTemplate(producerFactory())
+        return KafkaTemplate(producerFactory()).apply {
+            setObservationEnabled(true)
+        }
     }
-}
+}

infrastructure/kafka/kafka-producer/src/main/kotlin/io/github/hyungkishin/transentia/infrastructure/kafka/producer/service/impl/KafkaProducerImpl.kt

@@ -50,4 +50,4 @@ class KafkaProducerImpl<K : Serializable, V : SpecificRecordBase>(
         return kafkaTemplate.send(topicName, message)
     }
 
-}
+}

services/fds/infra/src/main/kotlin/io/github/hyungkishin/transentia/infra/adapter/in/messaging/TransferEventConsumer.kt

@@ -2,10 +2,12 @@ package io.github.hyungkishin.transentia.infra.adapter.`in`.messaging
 
 import com.fasterxml.jackson.databind.ObjectMapper
 import io.github.hyungkishin.transentia.application.service.AnalyzeTransferService
+import io.github.hyungkishin.transentia.infra.config.TracingTransformerSupplier
 import io.github.hyungkishin.transentia.infra.event.TransferEventMapper
 import io.github.hyungkishin.transentia.infrastructure.kafka.model.TransferEventAvroModel
 import org.apache.kafka.streams.kstream.KStream
 import org.slf4j.LoggerFactory
+import org.slf4j.MDC
 import org.springframework.context.annotation.Bean
 import org.springframework.context.annotation.Configuration
 import java.util.function.Function
@@ -22,6 +24,7 @@ import java.util.function.Function
  * 특징:
  * - Stateless 처리 (이전 이벤트 참조 불필요)
  * - 실시간 차단/리뷰 판정
+ * - TracingTransformer로 분산 트레이싱 컨텍스트 전파
  */
 @Configuration
 class TransferEventConsumer(
@@ -35,15 +38,16 @@ class TransferEventConsumer(
     fun processTransferEvents(): Function<KStream<String, TransferEventAvroModel>, KStream<String, String>> {
         return Function { input ->
             input
+                .transformValues(TracingTransformerSupplier())
                 .peek { key, event ->
                     log.info(
                         "[FDS단일분석] 이벤트 수신 - key={} eventId={} accountId={} amount={}",
                         key, event.eventId, event.receiverId, event.amount
                     )
                 }
-                .mapValues { _, event ->
+                .mapValues { event ->
                     try {
-                        // 1. Avro → Domain Event 변환
+                        // 1. Domain Event 변환
                         val domainEvent = transferEventMapper.toDomain(event)
 
                         // 2. FDS 분석 실행 (Application Layer)
@@ -92,6 +96,9 @@ class TransferEventConsumer(
                         )
 
                         objectMapper.writeValueAsString(errorResult)
+                    } finally {
+                        // MDC 정리
+                        MDC.clear()
                     }
                 }
         }

services/fds/infra/src/main/kotlin/io/github/hyungkishin/transentia/infra/config/TracingTransformer.kt

@@ -0,0 +1,57 @@
+package io.github.hyungkishin.transentia.infra.config
+
+import io.github.hyungkishin.transentia.infrastructure.kafka.model.TransferEventAvroModel
+import org.apache.kafka.streams.kstream.ValueTransformerWithKey
+import org.apache.kafka.streams.kstream.ValueTransformerWithKeySupplier
+import org.apache.kafka.streams.processor.ProcessorContext
+import org.slf4j.MDC
+
+/**
+ * Kafka Record 헤더에서 traceparent를 추출하여 MDC에 설정하는 Transformer
+ * 
+ * Kafka Streams는 일반 KafkaConsumer와 달리 observationEnabled가 적용되지 않아서
+ * 수동으로 W3C Trace Context 헤더를 파싱해야 함
+ * 
+ * traceparent 형식: {version}-{traceId}-{spanId}-{flags}
+ * 예: 00-130c0e23e150eb0ec69d4a4774cc1f03-47684cf7bc701ad3-01
+ */
+class TracingTransformer : ValueTransformerWithKey<String, TransferEventAvroModel, TransferEventAvroModel> {
+
+    private lateinit var context: ProcessorContext
+
+    override fun init(context: ProcessorContext) {
+        this.context = context
+    }
+
+    override fun transform(key: String?, value: TransferEventAvroModel): TransferEventAvroModel {
+        // 이전 MDC 정리
+        MDC.clear()
+
+        val headers = context.headers()
+        val traceparent = headers.lastHeader("traceparent")?.value()?.let { String(it) }
+
+        if (traceparent != null) {
+            // traceparent 형식: 00-{traceId}-{spanId}-{flags}
+            val parts = traceparent.split("-")
+            if (parts.size >= 3) {
+                MDC.put("traceId", parts[1])
+                MDC.put("spanId", parts[2])
+            }
+        }
+
+        return value
+    }
+
+    override fun close() {
+        MDC.clear()
+    }
+}
+
+/**
+ * TracingTransformer를 생성하는 Supplier
+ */
+class TracingTransformerSupplier : ValueTransformerWithKeySupplier<String, TransferEventAvroModel, TransferEventAvroModel> {
+    override fun get(): ValueTransformerWithKey<String, TransferEventAvroModel, TransferEventAvroModel> {
+        return TracingTransformer()
+    }
+}

services/fds/instances/api/src/main/resources/application-local.yml

@@ -0,0 +1,20 @@
+# ============================================
+# Local 환경용 설정 (Docker 호스트명 → localhost)
+# ============================================
+spring:
+  kafka:
+    bootstrap-servers: localhost:9094
+  cloud:
+    stream:
+      kafka:
+        binder:
+          brokers: localhost:9094
+        streams:
+          binder:
+            brokers: localhost:9094
+            configuration:
+              schema.registry.url: http://localhost:8085
+
+kafka-config:
+  bootstrap-servers: localhost:9094
+  schema-registry-url: http://localhost:8085

services/fds/instances/api/src/main/resources/application.yml

@@ -4,6 +4,9 @@ server:
 spring:
   application:
     name: fds-service
+  autoconfigure:
+    exclude:
+      - org.springframework.boot.actuate.autoconfigure.tracing.BraveAutoConfiguration
 
   # ============================================
   # Spring Cloud Stream 설정
@@ -120,7 +123,6 @@ management:
     tags:
       application: ${spring.application.name}
   tracing:
-    enabled: false
-  zipkin:
-    tracing:
-      endpoint: http://localhost:9999/disabled
+    enabled: true
+    sampling:
+      probability: 1.0  # 개발환경 100% 샘플링

services/fds/instances/api/src/main/resources/logback-spring.xml

@@ -1,9 +1,7 @@
 <configuration>
     <appender name="CONSOLE" class="ch.qos.logback.core.ConsoleAppender">
         <encoder>
-            <pattern>
-                [%d{yyyy-MM-dd HH:mm:ss.SSS}] [%-5level] [trace:%X{X-Trace-Id}] %logger{36} - %msg%n
-            </pattern>
+            <pattern>%d{HH:mm:ss.SSS} [%5level] [%X{traceId:-}] %logger{25} - %msg%n</pattern>
         </encoder>
     </appender>
 

services/transfer/infra/src/main/kotlin/io/github/hyungkishin/transentia/infra/adapter/KafkaTransferEventPublisher.kt

@@ -9,11 +9,9 @@ import io.github.hyungkishin.transentia.infrastructure.kafka.model.TransferEvent
 import io.github.hyungkishin.transentia.infrastructure.kafka.model.TransferStatus
 import io.github.hyungkishin.transentia.infrastructure.kafka.producer.service.KafkaProducer
 import org.slf4j.LoggerFactory
-import org.slf4j.MDC
 import org.springframework.beans.factory.annotation.Value
 import org.springframework.stereotype.Component
 import java.time.Instant
-import java.util.*
 
 @Component
 class KafkaTransferEventPublisher(
@@ -41,7 +39,6 @@ class KafkaTransferEventPublisher(
                         mapOf(
                             "eventType" to "TRANSFER_COMPLETED",
                             "eventVersion" to "v1",
-                            "traceId" to (MDC.get("traceId") ?: UUID.randomUUID().toString()),
                             "producer" to "transfer-api",
                             "contentType" to "application/json"
                         )
@@ -79,7 +76,6 @@ class KafkaTransferEventPublisher(
                     mapOf(
                         "eventType" to "TRANSFER_COMPLETED",
                         "eventVersion" to "v1",
-                        "traceId" to (MDC.get("traceId") ?: UUID.randomUUID().toString()),
                         "producer" to "transfer-api-fallback",
                         "contentType" to "application/json"
                     )

services/transfer/instances/api/src/main/kotlin/io/github/hyungkishin/transentia/api/config/AsyncConfig.kt

@@ -1,9 +1,8 @@
 package io.github.hyungkishin.transentia.api.config
 
-import org.slf4j.MDC
 import org.springframework.context.annotation.Bean
 import org.springframework.context.annotation.Configuration
-import org.springframework.core.task.TaskDecorator
+import org.springframework.core.task.support.ContextPropagatingTaskDecorator
 import org.springframework.scheduling.annotation.EnableAsync
 import org.springframework.scheduling.concurrent.ThreadPoolTaskExecutor
 import java.util.concurrent.Executor
@@ -24,28 +23,12 @@ class AsyncConfig {
         executor.setWaitForTasksToCompleteOnShutdown(true)
         executor.setAwaitTerminationSeconds(30)
 
-        // TaskDecorator 적용
-        executor.setTaskDecorator(mdcTaskDecorator())
+        // Spring Boot 3.0+ ContextPropagatingTaskDecorator
+        // MDC + Micrometer Observation Context 모두 전파
+        executor.setTaskDecorator(ContextPropagatingTaskDecorator())
 
         executor.initialize()
         return executor
     }
 
-    // MDC 정보 전파를 위한 TaskDecorator
-    private fun mdcTaskDecorator(): TaskDecorator {
-        return TaskDecorator { runnable ->
-            val contextMap = MDC.getCopyOfContextMap()
-            Runnable {
-                try {
-                    if (contextMap != null) {
-                        MDC.setContextMap(contextMap)
-                    }
-                    runnable.run()
-                } finally {
-                    MDC.clear()
-                }
-            }
-        }
-    }
-
-}
+}

services/transfer/instances/api/src/main/resources/application-local.yml

@@ -0,0 +1,6 @@
+# ============================================
+# Local 환경용 설정 (Docker 호스트명 → localhost)
+# ============================================
+kafka-config:
+  bootstrap-servers: localhost:9094
+  schema-registry-url: http://localhost:8085

services/transfer/instances/api/src/main/resources/application.yml

@@ -1,6 +1,9 @@
 spring:
   application:
     name: transfer-api
+  autoconfigure:
+    exclude:
+      - org.springframework.boot.actuate.autoconfigure.tracing.BraveAutoConfiguration
   datasource:
     url: jdbc:postgresql://localhost:5432/transfer
     username: postgres
@@ -79,7 +82,6 @@ management:
     tags:
       application: ${spring.application.name}
   tracing:
-    enabled: false
-  zipkin:
-    tracing:
-      endpoint: http://localhost:9999/disabled
+    enabled: true
+    sampling:
+      probability: 1.0  # 개발환경 100% 샘플링