SpringCloud系列教材 (九)- 服务链路追踪


 

步骤1:什么是服务链路
步骤2:先运行,看到效果,再学习
步骤3:模仿和排错
步骤4:改造
步骤5:pom.xml
步骤6:配置信息
步骤7:sampler

步骤 1 : 什么是服务链路

在前面的例子里,我们有两个微服务,分别是数据服务和视图服务,随着业务的增加,就会有越来越多的微服务存在,他们之间也会有更加复杂的调用关系。
这个调用关系,仅仅通过观察代码,会越来越难以识别,所以就需要通过 zipkin 服务链路追踪服务器 这个东西来用图片进行识别了。

步骤 2 : 先运行,看到效果,再学习

老规矩,先下载下载区(点击进入)的可运行项目,配置运行起来,确认可用之后,再学习做了哪些步骤以达到这样的效果。 

1. 需要启动链路追踪服务器,这个启动办法是下载右上角的 zipkin-server-2.10.1-exec.jar, 然后用如下命令启动:

java -jar zipkin-server-2.10.1-exec.jar



2. 挨个启动 eureka-server, 改造后的 product-data-service 和 product-view-service-feign 。 ( product-view-service-ribbon 后续不再使用,所以既没有被改造,也不用再启动了)

3. 访问一次 http://127.0.0.1:8012/products 通过 视图微服务去访问数据微服务,这样链路追踪服务器才知道有这事儿发生~

4. 然后打开链路追踪服务器 http://localhost:9411/zipkin/dependency/ 就可以看到如图所示的 视图微服务调用数据微服务 的图形了。

SpringBoot使用skywalking实现链路追踪 spring cloud链路追踪_java

步骤 3 : 模仿和排错

在确保可运行项目能够正确无误地运行之后,再严格照着教程的步骤,对代码模仿一遍。 
模仿过程难免代码有出入,导致无法得到期望的运行结果,此时此刻通过比较正确答案 ( 可运行项目 ) 和自己的代码,来定位问题所在。 
采用这种方式,学习有效果,排错有效率,可以较为明显地提升学习速度,跨过学习路上的各个槛。 

推荐使用diffmerge软件,进行文件夹比较。把你自己做的项目文件夹,和我的可运行项目文件夹进行比较。 
这个软件很牛逼的,可以知道文件夹里哪两个文件不对,并且很明显地标记出来 
这里提供了绿色安装和使用教程:diffmerge 下载和使用教程

步骤 4 : 改造

eureka-server 不需要做改造。
product-data-service和product-view-service 需要进行改造以使其可以被追踪到。

步骤 5 : pom.xml

两个 pom.xml 里都增加 zipkin 的jar包。

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


<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<modelVersion>4.0.0</modelVersion>
<parent>
<groupId>cn.how2j.springcloud</groupId>
<artifactId>springcloud</artifactId>
<version>0.0.1-SNAPSHOT</version>
</parent>
<artifactId>product-data-service</artifactId>
 
<dependencies>
<dependency>
<groupId>org.springframework.cloud</groupId>
<artifactId>spring-cloud-starter-netflix-eureka-client</artifactId>
</dependency>
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-web</artifactId>
</dependency>
<dependency>
<groupId>org.springframework.cloud</groupId>
<artifactId>spring-cloud-starter-zipkin</artifactId>
</dependency>      
 
</dependencies>
 
</project>
<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<modelVersion>4.0.0</modelVersion>
<parent>
<groupId>cn.how2j.springcloud</groupId>
<artifactId>springcloud</artifactId>
<version>0.0.1-SNAPSHOT</version>
</parent>
<artifactId>product-view-service-feign</artifactId>
 
<dependencies>
<dependency>
<groupId>org.springframework.cloud</groupId>
<artifactId>spring-cloud-starter-netflix-eureka-client</artifactId>
</dependency>
<dependency>
<groupId>org.springframework.cloud</groupId>
<artifactId>spring-cloud-starter-openfeign</artifactId>
</dependency>
 
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-web</artifactId>
</dependency>
 
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-thymeleaf</artifactId>
</dependency>
 
<dependency>
<groupId>org.springframework.cloud</groupId>
<artifactId>spring-cloud-starter-zipkin</artifactId>
</dependency>      
 
</dependencies> 
 
</project>

步骤 6 : 配置信息

两个的配置文件都加上

spring:
zipkin:
base-url: http://localhost:9411


#   server:
#   port: 因为会启动多个 product-data-service, 所以端口号由用户自动设置,推荐 8001,8002,8003
 
spring:
application:
name: product-data-service
zipkin:
base-url: http://localhost:9411       
eureka:
client:
serviceUrl:
defaultZone: http://localhost:8761/eureka/


eureka:
client:
serviceUrl:
defaultZone: http://localhost:8761/eureka/
spring:
application:
name: product-view-service-feign
zipkin:
base-url: http://localhost:9411   
thymeleaf:
cache: false
prefix: classpath:/templates/
suffix: .html
encoding: UTF-8
content-type: text/html
mode: HTML5   


步骤 7 : sampler

在启动类里配置 Sampler 抽样策略: ALWAYS_SAMPLE 表示持续抽样

@Bean
public Sampler defaultSampler() {
return Sampler.ALWAYS_SAMPLE;
} 
package cn.how2j.springcloud;
 
import java.util.Scanner;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.Future;
import java.util.concurrent.TimeUnit;
import java.util.concurrent.TimeoutException;
import org.springframework.boot.autoconfigure.SpringBootApplication;
import org.springframework.boot.builder.SpringApplicationBuilder;
import org.springframework.cloud.netflix.eureka.EnableEurekaClient;
import org.springframework.context.annotation.Bean;
 
import brave.sampler.Sampler;
import cn.hutool.core.convert.Convert;
import cn.hutool.core.thread.ThreadUtil;
import cn.hutool.core.util.NetUtil;
import cn.hutool.core.util.NumberUtil;
 
@SpringBootApplication
@EnableEurekaClient
public class ProductDataServiceApplication {
public static void main(String[] args) {
int port = 0;
int defaultPort = 8001;
Future<Integer> future = ThreadUtil.execAsync(() ->{
int p = 0;
System.out.println("请于5秒钟内输入端口号, 推荐  8001 、 8002  或者  8003,超过5秒将默认使用 " + defaultPort);
Scanner scanner = new Scanner(System.in);
while(true) {
String strPort = scanner.nextLine();
if(!NumberUtil.isInteger(strPort)) {
System.err.println("只能是数字");
continue;
}
else {
p = Convert.toInt(strPort);
scanner.close();
break;
}
}
return p;
});
try{
port=future.get(5,TimeUnit.SECONDS);
}
catch (InterruptedException | ExecutionException | TimeoutException e){
port = defaultPort;
}
 
if(!NetUtil.isUsableLocalPort(port)) {
System.err.printf("端口%d被占用了,无法启动%n", port );
System.exit(1);
}
 
new SpringApplicationBuilder(ProductDataServiceApplication.class).properties("server.port=" + port).run(args);
}
@Bean
public Sampler defaultSampler() {
return Sampler.ALWAYS_SAMPLE;
}   
}
package cn.how2j.springcloud;
 
import java.util.Scanner;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.Future;
import java.util.concurrent.TimeUnit;
import java.util.concurrent.TimeoutException;
 
import org.springframework.boot.autoconfigure.SpringBootApplication;
import org.springframework.boot.builder.SpringApplicationBuilder;
import org.springframework.cloud.client.discovery.EnableDiscoveryClient;
import org.springframework.cloud.netflix.eureka.EnableEurekaClient;
import org.springframework.cloud.openfeign.EnableFeignClients;
import org.springframework.context.annotation.Bean;
 
import brave.sampler.Sampler;
import cn.hutool.core.convert.Convert;
import cn.hutool.core.thread.ThreadUtil;
import cn.hutool.core.util.NetUtil;
import cn.hutool.core.util.NumberUtil;
 
@SpringBootApplication
@EnableEurekaClient
@EnableDiscoveryClient
@EnableFeignClients
public class ProductViewServiceFeignApplication {
 
public static void main(String[] args) {
int port = 0;
int defaultPort = 8012;
Future<Integer> future = ThreadUtil.execAsync(() ->{
int p = 0;
System.out.println("请于5秒钟内输入端口号, 推荐  8012 、 8013  或者  8014,超过5秒将默认使用"+defaultPort);
Scanner scanner = new Scanner(System.in);
while(true) {
String strPort = scanner.nextLine();
if(!NumberUtil.isInteger(strPort)) {
System.err.println("只能是数字");
continue;
}
else {
p = Convert.toInt(strPort);
scanner.close();
break;
}
}
return p;
});
try{
port=future.get(5,TimeUnit.SECONDS);
}
catch (InterruptedException | ExecutionException | TimeoutException e){
port = defaultPort;
}
if(!NetUtil.isUsableLocalPort(port)) {
System.err.printf("端口%d被占用了,无法启动%n", port );
System.exit(1);
}
new SpringApplicationBuilder(ProductViewServiceFeignApplication.class).properties("server.port=" + port).run(args);
 
}
@Bean
public Sampler defaultSampler() {
return Sampler.ALWAYS_SAMPLE;
}
 
}

更多内容,点击了解: http://how2j.cn/k/springcloud/springcloud-zipkin/2052.html