使用Spring Cloud Sleuth、Zipkin、Kafka、Elasticsearch实现分布式追踪

阅读更多

本例使用Spring Cloud Sleuth来实现分布式追踪，利用Kafka实现异步的分布式追踪，使用Elasticsearch存储追踪数据，使用Zipkin查看追踪内容。

1.前置条件

使用docker compose安装Kafka和Elasticsearch：

Kafka安装

version: '2'

services:
  zk:
    image: wurstmeister/zookeeper
    ports:
      - "2181:2181"


  kafka:
    image: wurstmeister/kafka:0.10.1.1
    ports:
      - "9092:9092"
    environment:
      KAFKA_ADVERTISED_HOST_NAME: kafka
      KAFKA_ZOOKEEPER_CONNECT: zk:2181
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock

Elasticsearch安装

version: '2'

services:
  elasticsearch:
    image: registry.cn-hangzhou.aliyuncs.com/wiselyman/es553:5.5.3
    restart: on-failure
    ports:
      - '9200:9200'
      - '9300:9300'

registry.cn-hangzhou.aliyuncs.com/wiselyman/es553:5.5.3是docker.elastic.co/elasticsearch/elasticsearch:5.5.3阿里云加速镜像

2.搭建Zipkin Server

依赖

<dependency>
	<groupId>org.springframework.cloudgroupId>
	<artifactId>spring-cloud-starter-sleuthartifactId>
dependency>

<dependency>
	<groupId>org.springframework.cloudgroupId>
	<artifactId>spring-cloud-sleuth-zipkin-streamartifactId>
dependency>

<dependency>
	<groupId>org.springframework.cloudgroupId>
	<artifactId>spring-cloud-stream-binder-kafkaartifactId>
dependency>

<dependency>
	<groupId>io.zipkin.javagroupId>
	<artifactId>zipkin-autoconfigure-uiartifactId>
dependency>

<dependency>
	<groupId>io.zipkin.javagroupId>
	<artifactId>zipkin-autoconfigure-storage-elasticsearch-httpartifactId>
	<version>2.4.1version>
dependency>

3.开启Zipkin支持

@SpringBootApplication
@EnableZipkinStreamServer
public class ZipkinServerApplication {

	public static void main(String[] args) {
		SpringApplication.run(ZipkinServerApplication.class, args);
	}
}

配置kafka

spring.cloud.stream.kafka.binder.zkNodes: 192.168.1.222
spring.cloud.stream.kafka.binder.brokers: 192.168.1.222

配置ES存储

zipkin:
  storage:
    type: elasticsearch
    elasticsearch:
      hosts: http://192.168.1.222:9200
      username: elastic
      password: changeme
      cluster: elasticsearch
      index: zipkin
      index-shards: 1
      index-replicas: 1

4.示例调用服务from-service

依赖

<dependency>
	<groupId>org.springframework.cloudgroupId>
	<artifactId>spring-cloud-starter-sleuthartifactId>
dependency>

<dependency>
	<groupId>org.springframework.cloudgroupId>
	<artifactId>spring-cloud-stream-binder-kafkaartifactId>
dependency>

<dependency>
	<groupId>org.springframework.cloudgroupId>
	<artifactId>spring-cloud-sleuth-streamartifactId>
dependency>

配置

spring.application.name: from-service
server.port: 8081

spring.sleuth.sampler.percentage: 1.0

spring.cloud.stream.kafka.binder.zkNodes: 192.168.1.222
spring.cloud.stream.kafka.binder.brokers: 192.168.1.222

调用控制器

@RestController
public class HelloController {
    @Autowired
    RestTemplate restTemplate;
    @GetMapping("/sync")
    public String sync(@RequestParam String param){
        return restTemplate.getForObject("http://localhost:8082/toSync?param=" + param,String.class);
    }
}

5.示例被调用服务to-service

依赖

<dependency>
	<groupId>org.springframework.cloudgroupId>
	<artifactId>spring-cloud-starter-sleuthartifactId>
dependency>

<dependency>
	<groupId>org.springframework.cloudgroupId>
	<artifactId>spring-cloud-stream-binder-kafkaartifactId>
dependency>

<dependency>
	<groupId>org.springframework.cloudgroupId>
	<artifactId>spring-cloud-sleuth-streamartifactId>
dependency>

配置

spring.application.name: to-service
server.port: 8082

spring.sleuth.sampler.percentage: 1.0

spring.cloud.stream.kafka.binder.zkNodes: 192.168.1.222
spring.cloud.stream.kafka.binder.brokers: 192.168.1.222

演示

• 启动zipkin-server，访问zipkin页面http://localhost:8080/zipkin/:

• 启动from-service,to-service,访问http://localhost:8081/sync?param=sss，这时查看zipkin页面:

源码地址:http://www.wisely.top/2018/03/02/spring_cloud_sleuth-zipkin-kafka-es/