Using Structured Logging in Spring Boot 3.4 with Elastic Common Schema

Preface

Log tracing is an essential step in system fault diagnosis. Starting with Spring Boot 3.4.0, structured logging is supported, offering built‑in formats such as Elastic Common Schema (ECS) and Logstash, while also allowing custom log formats.

Structured Logging

ECS console output

By default, Spring Boot prints logs in a simple text format. To enable ECS‑formatted console logs, add the following property to application.properties :

logging.structured.format.console=ecs

After the change, the console displays logs in ECS JSON format (see image).

Write structured logs to file

Spring Boot 3.4.0 can also write structured logs to a file. Configure the file output with these properties:

logging.structured.format.file=ecs
logging.file.name=./demo-log.log

A log file named demo-log.log will be created, containing ECS‑formatted entries (see image).

Custom structured logging

Spring Boot 3.4.x provides the StructuredLogFormatter interface for custom log formatting. Implement the interface, for example:

package cn.com.ut.cloud.demolog.config;

import ch.qos.logback.classic.spi.ILoggingEvent;
import org.springframework.boot.logging.structured.StructuredLogFormatter;

public class MyStructuredLoggingFormatter implements StructuredLogFormatter
{
    @Override
    public String format(ILoggingEvent event) {
        return "time=" + event.getTimeStamp() + ",level=" + event.getLevel() + ",message=" + event.getMessage() + "\n";
    }
}

Register the custom formatter in application.properties :

logging.structured.format.console=cn.com.ut.cloud.demolog.config.MyStructuredLoggingFormatter

The logs will now be printed according to the custom format (see image).

Structured logging configuration

Additional ECS fields can be set via properties, for example:

logging.structured.ecs.service.name=MyApp
logging.structured.ecs.service.version=1
logging.structured.ecs.service.environment=Production
logging.structured.ecs.service.node-name=Primary

These fields appear in the structured log output (see image).

MDC support

Spring Boot 3.4 also supports MDC (Mapped Diagnostic Context). The following example adds a userId to MDC, logs a message, and then removes the entry:

package cn.com.ut.cloud.demolog.config;

import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.slf4j.MDC;
import org.springframework.boot.CommandLineRunner;
import org.springframework.stereotype.Component;
import java.util.UUID;

@Component
class MyLogger implements CommandLineRunner {
    private static final Logger logger = LoggerFactory.getLogger(MyLogger.class);

    @Override
    public void run(String... args) {
        MDC.put("userId", UUID.randomUUID().toString());
        logger.info("hello world==================");
        MDC.remove("userId");
    }
}

The resulting log entry includes the generated userId (see image).

Conclusion

Structured logging helps developers define clearer log output during development; whether to adopt it depends on the specific requirements of each system.

Source: juejin.cn/post/7447872749756989440