如何在Zuul网关/代理中部署Zipkin？附配置说明

Hey there! Let me break down exactly how to set up Zipkin with your Zuul gateway/proxy—since you already know Zipkin’s perfect for tracing requests and tracking latency in microservice projects, let’s get straight to the steps.

1. Spin Up the Zipkin Server First #

Before configuring Zuul, you need a Zipkin server to collect and visualize all tracing data. The quickest way to launch one is using the official quickstart jar:

curl -sSL https://zipkin.io/quickstart.sh | bash -s
java -jar zipkin.jar

By default, this will run Zipkin at http://localhost:9411—you can visit this URL later to check the tracing UI.

2. Add Zipkin Dependency to Your Zuul Project #

Next, add the Zipkin starter dependency to your Zuul project's build file. Here's how for both Gradle and Maven:

Gradle #

compile group: 'org.springframework.cloud', name: 'spring-cloud-starter-zipkin', version: '1.3.2.RELEASE'

Maven #

<dependency>
    <groupId>org.springframework.cloud</groupId>
    <artifactId>spring-cloud-starter-zipkin</artifactId>
    <version>1.3.2.RELEASE</version>
</dependency>

This dependency includes both Spring Sleuth (which generates trace IDs) and the Zipkin client components, so you don't need to add any extra packages.

3. Configure the Sampler Bean #

To control what percentage of requests get traced, define a Sampler bean in your Zuul configuration class. Here's the expanded code example for clarity:

import org.springframework.cloud.sleuth.sampler.Sampler;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.beans.factory.annotation.Value;

@Configuration
public class SleuthTracingConfig {

    @Value("${spring.sleuth.sampler.percentage}")
    private double samplingRate;

    @Bean
    public Sampler requestSampler() {
        // Returns the sampling rate we've configured
        return () -> samplingRate;
    }
}

Then set the sampling rate in your application.properties or application.yml:

application.properties #

# Trace 100% of requests (great for testing; lower in production!)
spring.sleuth.sampler.percentage=1.0

application.yml #

spring:
  sleuth:
    sampler:
      percentage: 1.0

Pro tip: For production environments, drop this to a lower value like 0.1 (10% of requests) to avoid overload from tracing data.

4. Point Zuul to Your Zipkin Server (If Not Default) #

If your Zipkin server isn't running on the default localhost:9411, add this configuration to tell Zuul where to send tracing data:

application.properties #

spring.zipkin.base-url=http://your-zipkin-server-address:9411

application.yml #

spring:
  zipkin:
    base-url: http://your-zipkin-server-address:9411

5. Test the Setup #

Start your Zuul gateway, send a few requests through it, then head to the Zipkin UI (http://localhost:9411). You’ll see full traces of your requests, including how long Zuul took to route them, and any downstream service calls tied to the request.

Quick Notes #

• Make sure your other microservices also have Zipkin dependencies and sampling configured if you want to see the complete call chain.
• The 1.3.2.RELEASE version is a bit older—if you’re using a newer Spring Cloud release (like Greenwich or later), use a compatible Zipkin version to avoid conflicts.

内容的提问来源于stack exchange，提问作者RITESH GUPTA