# Spring Boot Logging im Jahr 2026: strukturierte Logs in Produktion mit Logback und JSON

> Vollständiger Leitfaden zu strukturiertem Logging in Spring Boot. Logback-JSON-Konfiguration, MDC für Tracing, Best Practices in Produktion und ELK-Stack-Integration.

- Published: 2026-03-27
- Updated: 2026-05-04
- Author: SharpSkill
- Tags: spring boot logging, logback json, structured logs, elk stack, observability
- Reading time: 14 min

---

Klassische Text-Logs werden in Produktion schnell unbeherrschbar. Bei Hunderten von Instanzen, die Tausende Zeilen pro Sekunde erzeugen, wird die Suche nach einem bestimmten Fehler zum Albtraum. Strukturierte JSON-Logs verändern diese Lage grundlegend, da jedes Ereignis abfragbar und automatisch auswertbar wird.

> **Kernaussage**
>
> Spring Boot 3.4+ unterstützt strukturiertes JSON-Logging nativ ohne externe Abhängigkeiten. Für ältere Versionen bleibt der Logback Logstash Encoder die Referenzlösung.

## Warum strukturierte Logs einsetzen

### Grenzen klassischer Text-Logs

Ein typischer Text-Log sieht so aus:

```
2026-03-27 10:15:32.456 INFO  [order-service,abc123] c.e.s.OrderService - Order created for user john@example.com, amount: 150.00€, items: 3
```

Dieses Format wirft in Produktion mehrere Probleme auf. Das Extrahieren bestimmter Informationen erfordert komplexe und fragile reguläre Ausdrücke. Die Korrelation zwischen Services setzt strenge Konventionen voraus, die jedes Team unterschiedlich auslegt. Analysewerkzeuge wie Elasticsearch tun sich schwer, diese unstrukturierten Zeichenketten effizient zu indexieren.

### Vorteile des JSON-Formats

Dasselbe Ereignis im JSON-Format wird sofort nutzbar:

```json
{
  "@timestamp": "2026-03-27T10:15:32.456Z",
  "level": "INFO",
  "logger": "com.example.service.OrderService",
  "message": "Order created",
  "service": "order-service",
  "traceId": "abc123",
  "userId": "john@example.com",
  "orderId": "ORD-789456",
  "amount": 150.00,
  "currency": "EUR",
  "itemCount": 3
}
```

Jedes Feld wird filterbar und aggregierbar. Eine Elasticsearch-Abfrage findet sofort alle Bestellungen über 100 € der letzten fünfzehn Minuten. Kibana-Dashboards visualisieren Trends ohne manuelles Parsing.

## Native Konfiguration in Spring Boot 3.4+

### Strukturierte JSON-Logs aktivieren

Spring Boot 3.4 führt nativen Support für strukturiertes Logging über die Eigenschaft `logging.structured` ein. Dieser Ansatz benötigt keine zusätzliche Abhängigkeit.

```yaml
# application.yml
# Native structured logging configuration for Spring Boot 3.4+
logging:
  structured:
    # Output format: ecs (Elastic), logstash, gelf
    format:
      console: ecs
      file: ecs
  file:
    name: /var/log/app/application.log
  level:
    root: INFO
    com.example: DEBUG
```

Das ECS-Format (Elastic Common Schema) garantiert direkte Kompatibilität mit Elasticsearch und Kibana ohne weitere Konfiguration.

### JSON-Felder anpassen

Um jedem Log fachliche Felder hinzuzufügen, lässt Spring Boot die Konfiguration zusätzlicher Attribute zu.

```yaml
# application.yml
# Custom fields in structured logs
logging:
  structured:
    format:
      console: ecs
    ecs:
      # Service information added to every log
      service:
        name: ${spring.application.name}
        version: ${app.version:1.0.0}
        environment: ${spring.profiles.active:default}
        node-name: ${HOSTNAME:unknown}
```

```java
// LoggingConfig.java
// Programmatic configuration for additional fields
package com.example.logging.config;

import org.springframework.boot.logging.structured.StructuredLogFormatterCustomizer;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;

@Configuration
public class LoggingConfig {

    @Bean
    StructuredLogFormatterCustomizer<EcsStructuredLogFormatter> ecsCustomizer() {
        return formatter -> formatter
            // Adds static fields to all logs
            .addStaticField("team", "backend")
            .addStaticField("region", System.getenv("AWS_REGION"))
            // Customizes exception formatting
            .setIncludeStacktrace(true)
            .setStacktraceMaxLength(5000);
    }
}
```

Diese Felder erscheinen in jeder Log-Zeile und erleichtern die Filterung nach Team oder Region in den Dashboards.

## Klassische Logback-Konfiguration mit JSON-Encoder

### Abhängigkeit Logstash Encoder

Für Spring-Boot-Versionen vor 3.4 oder bei Bedarf an erweiterter Anpassung bleibt der Logstash Logback Encoder die Referenzlösung.

```xml
<!-- pom.xml -->
<!-- Dependency for JSON logging with Logback -->
<dependency>
    <groupId>net.logstash.logback</groupId>
    <artifactId>logstash-logback-encoder</artifactId>
    <version>7.4</version>
</dependency>
```

### Vollständige Logback-Konfiguration

Die Datei `logback-spring.xml` bietet vollständige Kontrolle über das Ausgabeformat.

```xml
<!-- src/main/resources/logback-spring.xml -->
<!-- Logback configuration for structured JSON logs -->
<?xml version="1.0" encoding="UTF-8"?>
<configuration>
    <!-- Spring Boot properties -->
    <springProperty scope="context" name="appName" source="spring.application.name" defaultValue="app"/>
    <springProperty scope="context" name="appVersion" source="app.version" defaultValue="1.0.0"/>

    <!-- JSON console appender for production -->
    <appender name="JSON_CONSOLE" class="ch.qos.logback.core.ConsoleAppender">
        <encoder class="net.logstash.logback.encoder.LogstashEncoder">
            <!-- Custom fields added to every log -->
            <customFields>{"service":"${appName}","version":"${appVersion}"}</customFields>
            <!-- Includes MDC (tracing context) -->
            <includeMdcKeyName>traceId</includeMdcKeyName>
            <includeMdcKeyName>spanId</includeMdcKeyName>
            <includeMdcKeyName>userId</includeMdcKeyName>
            <includeMdcKeyName>requestId</includeMdcKeyName>
            <!-- ISO8601 timestamp format -->
            <timestampPattern>yyyy-MM-dd'T'HH:mm:ss.SSSZ</timestampPattern>
            <!-- Complete stack traces -->
            <throwableConverter class="net.logstash.logback.stacktrace.ShortenedThrowableConverter">
                <maxDepthPerThrowable>30</maxDepthPerThrowable>
                <maxLength>4096</maxLength>
                <shortenedClassNameLength>36</shortenedClassNameLength>
                <rootCauseFirst>true</rootCauseFirst>
            </throwableConverter>
        </encoder>
    </appender>

    <!-- Rolling JSON file appender -->
    <appender name="JSON_FILE" class="ch.qos.logback.core.rolling.RollingFileAppender">
        <file>/var/log/${appName}/application.json</file>
        <rollingPolicy class="ch.qos.logback.core.rolling.TimeBasedRollingPolicy">
            <fileNamePattern>/var/log/${appName}/application.%d{yyyy-MM-dd}.%i.json.gz</fileNamePattern>
            <maxHistory>30</maxHistory>
            <maxFileSize>100MB</maxFileSize>
            <totalSizeCap>3GB</totalSizeCap>
        </rollingPolicy>
        <encoder class="net.logstash.logback.encoder.LogstashEncoder">
            <customFields>{"service":"${appName}","version":"${appVersion}"}</customFields>
        </encoder>
    </appender>

    <!-- Text appender for development -->
    <appender name="TEXT_CONSOLE" class="ch.qos.logback.core.ConsoleAppender">
        <encoder>
            <pattern>%d{HH:mm:ss.SSS} %highlight(%-5level) [%thread] %cyan(%logger{36}) - %msg%n</pattern>
        </encoder>
    </appender>

    <!-- Activation by Spring profile -->
    <springProfile name="prod,staging">
        <root level="INFO">
            <appender-ref ref="JSON_CONSOLE"/>
            <appender-ref ref="JSON_FILE"/>
        </root>
    </springProfile>

    <springProfile name="dev,local">
        <root level="DEBUG">
            <appender-ref ref="TEXT_CONSOLE"/>
        </root>
    </springProfile>
</configuration>
```

Diese Konfiguration aktiviert JSON-Logs nur in Produktion und behält in der Entwicklung lesbare Logs bei.

> **Spring-Profile**
>
> Der Einsatz von `<springProfile>` ermöglicht das automatische Umschalten zwischen Text- und JSON-Format je nach Umgebung, ohne die Konfiguration zu ändern.

## MDC für verteiltes Tracing

### Weitergabe des Trace-Kontexts

MDC (Mapped Diagnostic Context) reichert jeden Log mit Kontextinformationen wie Request- oder Trace-IDs an.

```java
// TracingFilter.java
// Filter for automatic trace context injection
package com.example.logging.filter;

import jakarta.servlet.FilterChain;
import jakarta.servlet.ServletException;
import jakarta.servlet.http.HttpServletRequest;
import jakarta.servlet.http.HttpServletResponse;
import org.slf4j.MDC;
import org.springframework.core.Ordered;
import org.springframework.core.annotation.Order;
import org.springframework.stereotype.Component;
import org.springframework.web.filter.OncePerRequestFilter;

import java.io.IOException;
import java.util.UUID;

@Component
@Order(Ordered.HIGHEST_PRECEDENCE)
public class TracingFilter extends OncePerRequestFilter {

    // Standard MDC keys for tracing
    private static final String TRACE_ID_KEY = "traceId";
    private static final String SPAN_ID_KEY = "spanId";
    private static final String REQUEST_ID_KEY = "requestId";
    private static final String USER_ID_KEY = "userId";

    @Override
    protected void doFilterInternal(
            HttpServletRequest request,
            HttpServletResponse response,
            FilterChain filterChain) throws ServletException, IOException {

        try {
            // Retrieve or generate trace identifiers
            String traceId = extractOrGenerate(request, "X-Trace-Id", TRACE_ID_KEY);
            String spanId = generateSpanId();
            String requestId = extractOrGenerate(request, "X-Request-Id", REQUEST_ID_KEY);
            String userId = request.getHeader("X-User-Id");

            // Inject into MDC to appear in all logs
            MDC.put(TRACE_ID_KEY, traceId);
            MDC.put(SPAN_ID_KEY, spanId);
            MDC.put(REQUEST_ID_KEY, requestId);
            if (userId != null) {
                MDC.put(USER_ID_KEY, userId);
            }

            // Propagate to responses for inter-service chaining
            response.setHeader("X-Trace-Id", traceId);
            response.setHeader("X-Request-Id", requestId);

            filterChain.doFilter(request, response);

        } finally {
            // Clean MDC after each request
            MDC.clear();
        }
    }

    private String extractOrGenerate(HttpServletRequest request, String header, String key) {
        String value = request.getHeader(header);
        return value != null ? value : UUID.randomUUID().toString().replace("-", "").substring(0, 16);
    }

    private String generateSpanId() {
        return UUID.randomUUID().toString().replace("-", "").substring(0, 8);
    }
}
```

Jeder während der Request-Verarbeitung erzeugte Log enthält automatisch diese Kennungen.

### Einsatz von MDC im Fachcode

```java
// OrderService.java
// Business service with enriched contextual logging
package com.example.service;

import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.slf4j.MDC;
import org.springframework.stereotype.Service;

@Service
public class OrderService {

    private static final Logger log = LoggerFactory.getLogger(OrderService.class);

    public Order createOrder(CreateOrderRequest request) {
        // Add business information to MDC context
        MDC.put("orderId", request.getOrderId());
        MDC.put("customerId", request.getCustomerId());

        try {
            log.info("Creating order with {} items", request.getItems().size());

            // Business logic...
            Order order = processOrder(request);

            log.info("Order created successfully, total: {} {}",
                order.getTotal(), order.getCurrency());

            return order;

        } catch (Exception e) {
            // Exception appears with full MDC context
            log.error("Failed to create order", e);
            throw e;
        } finally {
            // Clean business keys added
            MDC.remove("orderId");
            MDC.remove("customerId");
        }
    }
}
```

Der entstehende JSON-Log enthält alle für das Debugging notwendigen Informationen:

```json
{
  "@timestamp": "2026-03-27T10:15:32.456Z",
  "level": "INFO",
  "logger": "com.example.service.OrderService",
  "message": "Order created successfully, total: 150.00 EUR",
  "traceId": "a1b2c3d4e5f67890",
  "spanId": "12345678",
  "requestId": "req-abc-123",
  "userId": "user-456",
  "orderId": "ORD-789",
  "customerId": "CUST-321"
}
```

## Asynchrones Logging für Performance

### Konfiguration des Thread-Pools

In Produktion beeinträchtigen synchrone Log-Schreibvorgänge die Latenz der Anfragen. Der asynchrone Appender entkoppelt das Logging vom Haupt-Thread.

```xml
<!-- logback-spring.xml -->
<!-- High-performance asynchronous appender configuration -->
<appender name="ASYNC_JSON" class="ch.qos.logback.classic.AsyncAppender">
    <!-- Pending log buffer size -->
    <queueSize>1024</queueSize>
    <!-- Never block the calling thread -->
    <neverBlock>true</neverBlock>
    <!-- Threshold before dropping DEBUG/TRACE logs -->
    <discardingThreshold>20</discardingThreshold>
    <!-- Include caller information (expensive) -->
    <includeCallerData>false</includeCallerData>
    <!-- Actual appender for writing -->
    <appender-ref ref="JSON_FILE"/>
</appender>

<springProfile name="prod">
    <root level="INFO">
        <appender-ref ref="ASYNC_JSON"/>
    </root>
</springProfile>
```

### Metriken des Logging-Systems

Das Monitoring des Logging-Systems selbst verhindert stillen Log-Verlust.

```java
// LoggingMetrics.java
// Exposing Logback metrics via Micrometer
package com.example.logging.metrics;

import ch.qos.logback.classic.Logger;
import ch.qos.logback.classic.LoggerContext;
import ch.qos.logback.classic.spi.ILoggingEvent;
import ch.qos.logback.core.Appender;
import ch.qos.logback.classic.AsyncAppender;
import io.micrometer.core.instrument.Gauge;
import io.micrometer.core.instrument.MeterRegistry;
import org.slf4j.LoggerFactory;
import org.springframework.stereotype.Component;

import jakarta.annotation.PostConstruct;
import java.util.Iterator;

@Component
public class LoggingMetrics {

    private final MeterRegistry registry;

    public LoggingMetrics(MeterRegistry registry) {
        this.registry = registry;
    }

    @PostConstruct
    void registerMetrics() {
        LoggerContext context = (LoggerContext) LoggerFactory.getILoggerFactory();
        Logger rootLogger = context.getLogger(Logger.ROOT_LOGGER_NAME);

        // Iterate through appenders to find AsyncAppenders
        Iterator<Appender<ILoggingEvent>> it = rootLogger.iteratorForAppenders();
        while (it.hasNext()) {
            Appender<ILoggingEvent> appender = it.next();
            if (appender instanceof AsyncAppender asyncAppender) {
                registerAsyncMetrics(asyncAppender);
            }
        }
    }

    private void registerAsyncMetrics(AsyncAppender appender) {
        String appenderName = appender.getName();

        // Current queue size
        Gauge.builder("logback.async.queue.size", appender, AsyncAppender::getQueueSize)
            .tag("appender", appenderName)
            .description("Current async appender queue size")
            .register(registry);

        // Remaining capacity
        Gauge.builder("logback.async.queue.remaining", appender, AsyncAppender::getRemainingCapacity)
            .tag("appender", appenderName)
            .description("Remaining capacity in async queue")
            .register(registry);

        // Number of dropped logs
        Gauge.builder("logback.async.discarded", appender, AsyncAppender::getNumberOfElementsInQueue)
            .tag("appender", appenderName)
            .description("Number of discarded log events")
            .register(registry);
    }
}
```

Eine Prometheus-Alarmierung auf `logback.async.queue.remaining < 100` warnt vor drohendem Log-Verlust.

## ELK-Stack-Integration

### Filebeat-Konfiguration

Filebeat sammelt die JSON-Dateien und sendet sie ohne Transformation an Elasticsearch.

```yaml
# filebeat.yml
# Filebeat configuration for Spring Boot JSON logs
filebeat.inputs:
  - type: log
    enabled: true
    paths:
      - /var/log/*/application.json
    # Automatic JSON parsing
    json:
      keys_under_root: true
      overwrite_keys: true
      add_error_key: true
      message_key: message

processors:
  # Add Kubernetes metadata if available
  - add_kubernetes_metadata:
      host: ${NODE_NAME}
      matchers:
        - logs_path:
            logs_path: "/var/log/containers/"
  # Parse timestamp
  - timestamp:
      field: "@timestamp"
      layouts:
        - '2006-01-02T15:04:05.000Z'
        - '2006-01-02T15:04:05.000-07:00'
      test:
        - '2026-03-27T10:15:32.456Z'

output.elasticsearch:
  hosts: ["elasticsearch:9200"]
  index: "logs-%{[service]}-%{+yyyy.MM.dd}"
  pipeline: "spring-boot-logs"

setup.template:
  name: "logs"
  pattern: "logs-*"
```

### Elasticsearch-Pipeline zur Anreicherung

```json
// PUT _ingest/pipeline/spring-boot-logs
{
  "description": "Spring Boot logs enrichment",
  "processors": [
    {
      "geoip": {
        "field": "client.ip",
        "target_field": "client.geo",
        "ignore_missing": true
      }
    },
    {
      "user_agent": {
        "field": "user_agent.original",
        "target_field": "user_agent",
        "ignore_missing": true
      }
    },
    {
      "set": {
        "field": "event.ingested",
        "value": "{{_ingest.timestamp}}"
      }
    },
    {
      "script": {
        "description": "Classify log level severity",
        "source": """
          def level = ctx.level;
          if (level == 'ERROR') ctx.severity = 4;
          else if (level == 'WARN') ctx.severity = 3;
          else if (level == 'INFO') ctx.severity = 2;
          else ctx.severity = 1;
        """
      }
    }
  ]
}
```

## Best Practices in Produktion

### Systematisch einzubindende Informationen

Jeder Log sollte mindestens die für Debugging und Korrelation notwendigen Informationen enthalten.

```java
// StructuredLogger.java
// Helper for consistent structured logs
package com.example.logging;

import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.slf4j.MDC;

import java.util.Map;
import java.util.function.Supplier;

public final class StructuredLogger {

    private final Logger delegate;

    private StructuredLogger(Class<?> clazz) {
        this.delegate = LoggerFactory.getLogger(clazz);
    }

    public static StructuredLogger getLogger(Class<?> clazz) {
        return new StructuredLogger(clazz);
    }

    // Log with temporary business context
    public void info(String message, Map<String, String> context) {
        try {
            context.forEach(MDC::put);
            delegate.info(message);
        } finally {
            context.keySet().forEach(MDC::remove);
        }
    }

    // Log with supplier for lazy evaluation
    public void debug(Supplier<String> messageSupplier, Map<String, String> context) {
        if (delegate.isDebugEnabled()) {
            try {
                context.forEach(MDC::put);
                delegate.debug(messageSupplier.get());
            } finally {
                context.keySet().forEach(MDC::remove);
            }
        }
    }

    // Error log with full context
    public void error(String message, Throwable t, Map<String, String> context) {
        try {
            context.forEach(MDC::put);
            delegate.error(message, t);
        } finally {
            context.keySet().forEach(MDC::remove);
        }
    }
}
```

```java
// Usage in business code
private static final StructuredLogger log = StructuredLogger.getLogger(PaymentService.class);

public void processPayment(Payment payment) {
    log.info("Processing payment", Map.of(
        "paymentId", payment.getId(),
        "amount", String.valueOf(payment.getAmount()),
        "currency", payment.getCurrency(),
        "method", payment.getMethod().name()
    ));
}
```

### Sensible Informationen, die ausgeschlossen werden müssen

Logs dürfen niemals personenbezogene oder sensible Daten enthalten.

```java
// SensitiveDataFilter.java
// Sensitive data masking filter
package com.example.logging.filter;

import ch.qos.logback.classic.spi.ILoggingEvent;
import ch.qos.logback.core.filter.Filter;
import ch.qos.logback.core.spi.FilterReply;

import java.util.regex.Pattern;

public class SensitiveDataFilter extends Filter<ILoggingEvent> {

    // Sensitive data patterns to mask
    private static final Pattern EMAIL_PATTERN =
        Pattern.compile("[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\\.[a-zA-Z]{2,}");
    private static final Pattern CREDIT_CARD_PATTERN =
        Pattern.compile("\\b\\d{4}[- ]?\\d{4}[- ]?\\d{4}[- ]?\\d{4}\\b");
    private static final Pattern PASSWORD_PATTERN =
        Pattern.compile("(?i)(password|pwd|secret|token)[\"']?\\s*[:=]\\s*[\"']?[^\\s,}\"']+");
    private static final Pattern PHONE_PATTERN =
        Pattern.compile("\\+?\\d{1,3}[- ]?\\d{6,14}");

    @Override
    public FilterReply decide(ILoggingEvent event) {
        // Accept all logs but modify the message
        // Note: for real masking, use a custom converter
        return FilterReply.NEUTRAL;
    }

    // Utility method to mask data
    public static String maskSensitiveData(String input) {
        if (input == null) return null;

        String result = input;
        result = EMAIL_PATTERN.matcher(result).replaceAll("[EMAIL_MASKED]");
        result = CREDIT_CARD_PATTERN.matcher(result).replaceAll("[CARD_MASKED]");
        result = PASSWORD_PATTERN.matcher(result).replaceAll("$1=[REDACTED]");
        result = PHONE_PATTERN.matcher(result).replaceAll("[PHONE_MASKED]");

        return result;
    }
}
```

> **DSGVO und Compliance**
>
> Logs mit personenbezogenen Daten unterliegen der DSGVO. IP-Adressen, E-Mails und Nutzer-IDs erfordern eine Aufbewahrungsrichtlinie und gegebenenfalls eine Einwilligung.

### Geeignete Log-Level

```java
// LogLevelGuidelines.java
// Appropriate log level guidelines
package com.example.logging;

public class LogLevelGuidelines {

    // ERROR: Failure requiring intervention
    // - Unrecoverable exceptions
    // - Critical transaction failures
    // - External service unavailability
    log.error("Payment gateway unreachable after 3 retries", exception);

    // WARN: Abnormal but handled situation
    // - Retry in progress
    // - Performance degradation
    // - Resources near limits
    log.warn("Database connection pool at 85% capacity");

    // INFO: Significant business events
    // - Transaction start/end
    // - Important state changes
    // - Key user actions
    log.info("Order {} shipped to customer {}", orderId, customerId);

    // DEBUG: Diagnostic information
    // - Execution details
    // - Important variable values
    // - Branching decisions
    log.debug("Cache miss for key {}, fetching from database", cacheKey);

    // TRACE: Very fine details
    // - Method entry/exit
    // - Complete object contents
    // - Loops and iterations
    log.trace("Processing item {} of {}", index, total);
}
```

## Tests und Validierung der Logs

### Unit-Tests zur JSON-Struktur

```java
// StructuredLoggingTest.java
// Structured log validation tests
package com.example.logging;

import ch.qos.logback.classic.Logger;
import ch.qos.logback.classic.spi.ILoggingEvent;
import ch.qos.logback.core.read.ListAppender;
import com.fasterxml.jackson.databind.JsonNode;
import com.fasterxml.jackson.databind.ObjectMapper;
import org.junit.jupiter.api.BeforeEach;
import org.junit.jupiter.api.Test;
import org.slf4j.LoggerFactory;
import org.slf4j.MDC;

import static org.assertj.core.api.Assertions.assertThat;

class StructuredLoggingTest {

    private ListAppender<ILoggingEvent> listAppender;
    private Logger logger;
    private ObjectMapper objectMapper;

    @BeforeEach
    void setUp() {
        logger = (Logger) LoggerFactory.getLogger(StructuredLoggingTest.class);
        listAppender = new ListAppender<>();
        listAppender.start();
        logger.addAppender(listAppender);
        objectMapper = new ObjectMapper();
    }

    @Test
    void shouldIncludeMdcFieldsInLog() {
        // Given
        MDC.put("traceId", "test-trace-123");
        MDC.put("userId", "user-456");

        // When
        logger.info("Test message with MDC context");

        // Then
        ILoggingEvent event = listAppender.list.get(0);
        assertThat(event.getMDCPropertyMap())
            .containsEntry("traceId", "test-trace-123")
            .containsEntry("userId", "user-456");

        MDC.clear();
    }

    @Test
    void shouldLogExceptionWithStackTrace() {
        // Given
        Exception testException = new RuntimeException("Test error");

        // When
        logger.error("Operation failed", testException);

        // Then
        ILoggingEvent event = listAppender.list.get(0);
        assertThat(event.getThrowableProxy()).isNotNull();
        assertThat(event.getThrowableProxy().getMessage()).isEqualTo("Test error");
    }
}
```

## Fazit

Strukturierte JSON-Logs verändern die Observability von Spring-Boot-Anwendungen grundlegend:

✅ **Abfragbar**: jedes Feld wird in Elasticsearch oder CloudWatch filterbar

✅ **Korrelierbar**: MDC propagiert Trace-IDs zwischen Services

✅ **Performant**: der asynchrone Appender entkoppelt das Logging von der Verarbeitung

✅ **Sicher**: das Maskieren sensibler Daten sichert die DSGVO-Konformität

✅ **Integriert**: native Kompatibilität mit ELK Stack, Datadog, Splunk

✅ **Alarmfähig**: strukturierte Felder ermöglichen präzise Alert-Regeln

✅ **Wartbar**: das JSON-Format eliminiert fragile Parsing-Regex

Dieser Ansatz bildet zusammen mit Metriken (Micrometer) und verteiltem Tracing (OpenTelemetry) das Fundament moderner Observability.

---

Source: SharpSkill (https://sharpskill.dev), tech interview preparation for your real stack.
HTML version of this page: https://sharpskill.dev/de/blog/spring-boot/spring-boot-structured-logging-logback-json