Spring Cloud微服务课程设计 第三章:LoadBalancer负载均衡
第三章:LoadBalancer负载均衡
理论讲解
什么是负载均衡?
负载均衡就像是餐厅里的多个服务员,当有很多顾客同时来就餐时,经理会把顾客均匀地分配给不同的服务员,避免某个服务员忙不过来,而其他服务员却闲着。在微服务中,负载均衡就是把用户请求合理地分配到多个相同的服务实例上。
应用场景举例:
想象一下双11购物节:
- 成千上万的用户同时访问电商网站
- 如果所有请求都打到同一台商品服务服务器,服务器会崩溃
- 通过负载均衡,把请求分散到10台相同的商品服务服务器上
- 这样每台服务器只处理1/10的请求,系统就能稳定运行
项目结构
chapter-03-loadbalancer/
├── commons-service/ # 公共基础服务
├── eureka-server/ # 注册中心服务端
├── user-service/ # 用户服务
├── product-service/ # 商品服务(多实例)
├── order-service/ # 订单服务(使用LoadBalancer)
└── inventory-service/ # 库存服务(新增)
完整代码实现
1. 父工程pom.xml更新
在父工程中添加LoadBalancer依赖管理:
2. Product Service商品服务(多实例配置)
商品服务配置文件1 - application-8082.properties
#=============================================
# 0. 必改/高危配置(上线前一定要检查)
#=============================================
# 商品服务实例1端口,集群部署时切忌重复
server.port=8082
# Eureka 注册名,一旦上线禁止变更(消费者会硬编码)
spring.application.name=product-service
#=============================================
# 1. 服务器级配置
#=============================================
# 如需统一加前缀可修改,默认根路径
server.servlet.context-path=/
#=============================================
# 2. Spring 基础配置
#=============================================
# 开启 Spring Security 时的登录账号
spring.security.user.name=admin
#=============================================
# 3. Eureka 客户端实例配置
#=============================================
# 在 Eureka 控制台显示的主机名,可读性优先
eureka.instance.hostname=product-service
# 注册时优先使用 IP,防止 hostname 解析失败
eureka.instance.prefer-ip-address=true
# 唯一标识,格式 IP:端口
eureka.instance.instance-id=${spring.cloud.client.ip-address}:${server.port}
# 心跳间隔 5s,缩短感知时间
eureka.instance.lease-renewal-interval-in-seconds=5
# 10s 内收不到心跳即剔除,开发调试可设短
eureka.instance.lease-expiration-duration-in-seconds=10
# 自定义元数据,可做灰度、路由、权重
eureka.instance.metadata-map.zone=zone-a
# 权重,负载均衡策略可读取
eureka.instance.metadata-map.weight=1
# 版本号,方便 A/B 发布
eureka.instance.metadata-map.version=v1.1
#=============================================
# 4. Eureka 客户端行为配置
#=============================================
# 是否把自己注册到 Eureka(默认 true,显式写出可提醒运维)
eureka.client.register-with-eureka=true
# 是否拉取注册表(默认 true,同上)
eureka.client.fetch-registry=true
#=============================================
# 5. Actuator 监控配置
#=============================================
# 按需暴露,生产勿暴露 env/beans
management.endpoints.web.exposure.include=health,info,metrics
# 默认 never,always 方便排查
management.endpoint.health.show-details=always
#=============================================
# 6. 日志级别
#=============================================
# 业务包日志级别,上线后改为 INFO 或 WARN
logging.level.com.lihaozhe.productservice=DEBUG
商品服务配置文件2 - application-8084.properties
#=============================================
# 0. 必改/高危配置(上线前一定要检查)
#=============================================
# 商品服务实例2端口,集群部署时切忌重复
server.port=8084
# Eureka 注册名,一旦上线禁止变更(消费者会硬编码)
spring.application.name=product-service
#=============================================
# 1. 服务器级配置
#=============================================
# 如需统一加前缀可修改,默认根路径
server.servlet.context-path=/
#=============================================
# 2. Spring 基础配置
#=============================================
# 开启 Spring Security 时的登录账号
spring.security.user.name=admin
#=============================================
# 3. Eureka 客户端实例配置
#=============================================
# 在 Eureka 控制台显示的主机名,可读性优先
eureka.instance.hostname=product-service
# 注册时优先使用 IP,防止 hostname 解析失败
eureka.instance.prefer-ip-address=true
# 唯一标识,格式 IP:端口
eureka.instance.instance-id=${spring.cloud.client.ip-address}:${server.port}
# 心跳间隔 5s,缩短感知时间
eureka.instance.lease-renewal-interval-in-seconds=5
# 10s 内收不到心跳即剔除,开发调试可设短
eureka.instance.lease-expiration-duration-in-seconds=10
# 自定义元数据,可做灰度、路由、权重
eureka.instance.metadata-map.zone=zone-b
# 权重,负载均衡策略可读取
eureka.instance.metadata-map.weight=2
# 版本号,方便 A/B 发布
eureka.instance.metadata-map.version=v1.1
#=============================================
# 4. Eureka 客户端行为配置
#=============================================
# 是否把自己注册到 Eureka(默认 true,显式写出可提醒运维)
eureka.client.register-with-eureka=true
# 是否拉取注册表(默认 true,同上)
eureka.client.fetch-registry=true
#=============================================
# 5. Actuator 监控配置
#=============================================
# 按需暴露,生产勿暴露 env/beans
management.endpoints.web.exposure.include=health,info,metrics
# 默认 never,always 方便排查
management.endpoint.health.show-details=always
#=============================================
# 6. 日志级别
#=============================================
# 业务包日志级别,上线后改为 INFO 或 WARN
logging.level.com.lihaozhe.productservice=DEBUG
商品服务控制器增强
package com.lihaozhe.productservice.controller;
import com.lihaozhe.productservice.dto.ProductDTO;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.beans.factory.annotation.Value;
import org.springframework.web.bind.annotation.*;
import java.util.ArrayList;
import java.util.List;
import java.util.Map;
import java.util.Optional;
/**
* 商品服务控制器
* 提供商品管理的REST API
*
* @author 李昊哲
* @version 1.0.0
*/
@RestController
@RequestMapping("/api/products")
public class ProductController {
// 模拟商品数据
private final List<ProductDTO> products = new ArrayList<>();
private final Logger logger = LoggerFactory.getLogger(this.getClass());
// 注入当前服务端口,用于标识实例
@Value("${server.port}")
private String serverPort;
// 实例标识
private final String instanceId;
/**
* 构造函数,初始化模拟数据
*/
public ProductController() {
// 初始化测试商品
products.add(new ProductDTO(1L, "iPhone 15", "手机", 5999.00, 100, "最新款iPhone"));
products.add(new ProductDTO(2L, "MacBook Pro", "电脑", 12999.00, 50, "专业级笔记本电脑"));
products.add(new ProductDTO(3L, "AirPods Pro", "耳机", 1899.00, 200, "无线降噪耳机"));
products.add(new ProductDTO(4L, "iPad Air", "平板", 4399.00, 80, "轻薄便携平板"));
// 生成实例ID(基于时间戳)
this.instanceId = "product-service-" + System.currentTimeMillis() % 1000;
}
/**
* 根据ID获取商品 - 增强版本,返回实例信息
* GET /api/products/{id}
*/
@GetMapping("/{id}")
public Map<String, Object> getProduct(@PathVariable("id") Long id) {
logger.info("实例 {} (端口:{}) 处理商品查询请求: {}", instanceId, serverPort, id);
simulateProcessingTime();
Optional<ProductDTO> productOpt = products.stream()
.filter(p -> p.getId().equals(id))
.findFirst();
if (productOpt.isPresent()) {
ProductDTO product = productOpt.get();
return Map.of(
"product", product,
"instanceInfo", Map.of(
"instanceId", instanceId,
"serverPort", serverPort,
"timestamp", System.currentTimeMillis()
)
);
}
return Map.of(
"product", null,
"instanceInfo", Map.of(
"instanceId", instanceId,
"serverPort", serverPort,
"timestamp", System.currentTimeMillis()
)
);
}
/**
* 获取所有商品 - 增强版本
* GET /api/products
*/
@GetMapping
public Map<String, Object> getAllProducts() {
logger.info("实例 {} (端口:{}) 处理获取所有商品请求", instanceId, serverPort);
simulateProcessingTime();
return Map.of(
"products", products,
"instanceInfo", Map.of(
"instanceId", instanceId,
"serverPort", serverPort,
"totalProducts", products.size(),
"timestamp", System.currentTimeMillis()
)
);
}
/**
* 根据分类获取商品
* GET /api/products/category/{category}
*/
@GetMapping("/category/{category}")
public Map<String, Object> getProductsByCategory(@PathVariable("category") String category) {
logger.info("实例 {} (端口:{}) 处理分类查询: {}", instanceId, serverPort, category);
simulateProcessingTime();
List<ProductDTO> categoryProducts = products.stream()
.filter(p -> p.getCategory().equalsIgnoreCase(category))
.toList();
return Map.of(
"products", categoryProducts,
"instanceInfo", Map.of(
"instanceId", instanceId,
"serverPort", serverPort,
"category", category,
"count", categoryProducts.size(),
"timestamp", System.currentTimeMillis()
)
);
}
/**
* 更新商品库存
* PUT /api/products/{id}/stock
*/
@PutMapping("/{id}/stock")
public Map<String, Object> updateStock(@PathVariable("id") Long id, @RequestBody Map<String, Integer> request) {
Integer quantity = request.get("quantity");
logger.info("实例 {} (端口:{}) 更新商品库存: {}, 数量: {}", instanceId, serverPort, id, quantity);
// 查找商品并更新库存
Optional<ProductDTO> productOpt = products.stream()
.filter(p -> p.getId().equals(id))
.findFirst();
if (productOpt.isPresent()) {
ProductDTO product = productOpt.get();
int newStock = product.getStock() + quantity;
product.setStock(Math.max(newStock, 0)); // 库存不能为负数
return Map.of(
"success", true,
"message", "库存更新成功",
"newStock", product.getStock(),
"instanceInfo", Map.of(
"instanceId", instanceId,
"serverPort", serverPort,
"timestamp", System.currentTimeMillis()
)
);
}
return Map.of(
"success", false,
"message", "商品不存在",
"instanceInfo", Map.of(
"instanceId", instanceId,
"serverPort", serverPort,
"timestamp", System.currentTimeMillis()
)
);
}
/**
* 获取实例信息
* GET /api/products/instance-info
*/
@GetMapping("/instance-info")
public Map<String, Object> getInstanceInfo() {
return Map.of(
"instanceId", instanceId,
"serverPort", serverPort,
"serviceName", "product-service",
"startupTime", System.currentTimeMillis() - (System.currentTimeMillis() % 1000000),
"currentTime", System.currentTimeMillis()
);
}
/**
* 健康检查端点
* GET /api/products/health
*/
@GetMapping("/health")
public Map<String, Object> health() {
return Map.of(
"status", "UP",
"service", "product-service",
"timestamp", System.currentTimeMillis(),
"totalProducts", products.size()
);
}
/**
* 模拟处理时间
* 不同实例有不同的处理时间,便于观察负载均衡效果
*/
private void simulateProcessingTime() {
try {
// 基于端口号模拟不同的处理时间
int port = Integer.parseInt(serverPort);
int delay = port == 8082 ? 50 : 30; // 8082端口延迟50ms,8084端口延迟30ms
Thread.sleep(delay);
} catch (InterruptedException e) {
Thread.currentThread().interrupt();
}
}
}
3. Order Service订单服务(使用LoadBalancer)
pom.xml更新
<?xml version="1.0" encoding="UTF-8"?>
<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>com.lihaozhe</groupId>
<artifactId>springcloud-course</artifactId>
<version>1.0.0</version>
<relativePath>../../pom.xml</relativePath>
</parent>
<artifactId>order-service</artifactId>
<properties>
<maven.compiler.source>25</maven.compiler.source>
<maven.compiler.target>25</maven.compiler.target>
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
<!-- 替换为你的启动类全路径 -->
<start-class>com.lihaozhe.orderservice.OrderServiceApplication</start-class>
</properties>
<dependencies>
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-web</artifactId>
</dependency>
<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-actuator</artifactId>
</dependency>
<!-- 引出 commons-service 公共基础服务 -->
<dependency>
<groupId>com.lihaozhe</groupId>
<artifactId>commons-service</artifactId>
<version>1.0.0</version>
</dependency>
<!-- Spring Cloud loadbalancer依赖管理 -->
<dependency>
<groupId>org.springframework.cloud</groupId>
<artifactId>spring-cloud-starter-loadbalancer</artifactId>
</dependency>
<!-- Apache HttpClient 5 -->
<dependency>
<groupId>org.apache.httpcomponents.client5</groupId>
<artifactId>httpclient5</artifactId>
</dependency>
<!-- 可选:如果需要连接池等高级功能 -->
<dependency>
<groupId>org.apache.httpcomponents.core5</groupId>
<artifactId>httpcore5</artifactId>
</dependency>
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-validation</artifactId>
</dependency>
</dependencies>
<build>
<plugins>
<plugin>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-maven-plugin</artifactId>
<configuration>
<!-- 替换为你的启动类全路径 -->
<mainClass>com.lihaozhe.orderservice.OrderServiceApplication</mainClass>
</configuration>
<!-- 可选:如果需要打包为可执行jar,添加此配置 -->
<executions>
<execution>
<goals>
<goal>repackage</goal>
</goals>
</execution>
</executions>
</plugin>
</plugins>
</build>
</project>
应用启动类更新
package com.lihaozhe.orderservice;
import org.springframework.boot.SpringApplication;
import org.springframework.boot.autoconfigure.SpringBootApplication;
import org.springframework.cloud.client.discovery.EnableDiscoveryClient;
import org.springframework.cloud.client.loadbalancer.LoadBalanced;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Primary;
import org.springframework.web.client.RestClient;
/**
* 订单服务应用启动类 - 负载均衡版本
* 使用@LoadBalanced注解启用负载均衡功能
* 订单服务是服务消费者,会调用用户服务和商品服务
*
* @author 李昊哲
* @version 1.0.0
*/
@SpringBootApplication
@EnableDiscoveryClient
public class OrderServiceApplication {
public static void main(String[] args) {
SpringApplication.run(OrderServiceApplication.class, args);
}
/**
* 创建普通的RestClient(用于直接URL调用)
* 首选的默认RestClient(添加@Primary)
*
* @return 普通的RestClient实例
*/
@Primary
@Bean("restClient")
public RestClient restClient() {
return RestClient.builder().build();
}
/**
* 创建普通的RestClient(用于直接URL调用)
* 其他RestClient(非首选)指定名称
*
* @return 普通的RestClient实例
*/
@Bean("normalRestClient")
public RestClient normalRestClient() {
return RestClient.builder().build();
}
/**
* 创建负载均衡的RestClient
* 使用@LoadBalanced注解后,RestClient会自动使用服务名进行服务发现和负载均衡
* 这个方法本身不实现负载均衡逻辑,只是“启用”负载均衡功能。
* 实际的负载均衡策略由另一个组件(即 ReactorLoadBalancer<ServiceInstance>)提供。
* 它是客户端入口,用于发起带服务发现 + 负载均衡的请求。
* 其他RestClient(非首选)指定名称
*
* @return 用于发起带服务发现 + 负载均衡请求的客户端入口
*/
@Bean("loadBalancedRestClient")
@LoadBalanced
public RestClient loadBalancedRestClient() {
return RestClient.builder().build();
}
}
工具类
package com.lihaozhe.orderservice.util;
import org.springframework.beans.factory.annotation.Qualifier;
import org.springframework.cloud.client.ServiceInstance;
import org.springframework.cloud.client.discovery.DiscoveryClient;
import org.springframework.cloud.client.loadbalancer.LoadBalancerClient;
import org.springframework.stereotype.Component;
import org.springframework.web.client.RestClient;
import java.util.List;
import java.util.concurrent.ThreadLocalRandom;
/**
* Loadbalancer工具
*
* @author 李昊哲
* @version 1.0.0
*/
@Component
public class LoadbalancerUtil {
private final DiscoveryClient discoveryClient;
private final LoadBalancerClient loadBalancerClient;
@Qualifier("loadBalancedRestClient")
private final RestClient loadBalancedRestClient;
/**
* 构造函数,注入依赖
*/
public LoadbalancerUtil(DiscoveryClient discoveryClient,
LoadBalancerClient loadBalancerClient, RestClient loadBalancedRestClient) {
this.discoveryClient = discoveryClient;
this.loadBalancerClient = loadBalancerClient;
this.loadBalancedRestClient = loadBalancedRestClient;
}
/**
* 获取当前服务实例信息
*
* @return 当前服务实例列表
*/
public List<ServiceInstance> getDiscoveryClient(String serviceId) {
return discoveryClient.getInstances(serviceId);
}
/**
* 获取当前选择的服务实例
*
* @return 当前选择的服务实例
*/
public ServiceInstance getCurrentServiceInstance(String serviceId) {
return loadBalancerClient.choose(serviceId);
}
/**
* 随机获取某服务实例主机
*
* @param serviceId 服务实例
* @return 服务实例主机 URI 地址
*/
public String randomServiceUrl(String serviceId) {
List<ServiceInstance> instances = getDiscoveryClient(serviceId);
ServiceInstance instance = instances.get(ThreadLocalRandom.current().nextInt(instances.size()));
return instance.getUri().toString();
}
/**
* 获取某服务实例主机
*
* @param serviceId 服务实例
* @return 服务实例主机 URI 地址
*/
public String getCurrentServiceUrl(String serviceId) {
return getCurrentServiceInstance(serviceId).getUri().toString();
}
public LoadBalancerClient getLoadBalancerClient() {
return loadBalancerClient;
}
public RestClient getLoadBalancedRestClient() {
return loadBalancedRestClient;
}
}
负载均衡服务类
package com.lihaozhe.orderservice.service;
import com.lihaozhe.orderservice.util.LoadbalancerUtil;
import com.lihaozhe.productservice.dto.ProductDTO;
import com.lihaozhe.userservice.dto.UserDTO;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.cloud.client.ServiceInstance;
import org.springframework.core.ParameterizedTypeReference;
import org.springframework.stereotype.Service;
import java.util.ArrayList;
import java.util.List;
import java.util.Map;
/**
* 负载均衡服务类
* 演示如何使用LoadBalancer进行服务调用和负载均衡
*
* @author 李昊哲
* @version 1.0.0
*/
@Service
public class LoadBalancerService {
private final LoadbalancerUtil loadbalancerUtil;
private final Logger logger = LoggerFactory.getLogger(LoadBalancerService.class);
/**
* 构造函数,注入依赖
*/
public LoadBalancerService(LoadbalancerUtil loadbalancerUtil) {
this.loadbalancerUtil = loadbalancerUtil;
}
/**
* 方式1:使用@LoadBalanced的RestClient进行服务调用(推荐)
* 这种方式会自动进行负载均衡
*/
public ProductDTO getProductWithLoadBalancer(Long productId) {
// 注意:这里使用服务名而不是具体的URL
// String url = "http://product-service/api/products/" + productId;
String productServiceId = "product-service";
String url = String.format("%s/api/products/%d", loadbalancerUtil.getCurrentServiceUrl(productServiceId), productId);
try {
// 定义泛型类型:Map<String, User>
ParameterizedTypeReference<Map<String, Object>> typeRef = new ParameterizedTypeReference<Map<String, Object>>() {
};
Map<String, Object> response = loadbalancerUtil.getLoadBalancedRestClient().get()
.uri(url)
.retrieve()
.body(typeRef);
if (response != null && response.get("product") != null) {
// 将Map转换为ProductDTO
Map<String, Object> productMap = (Map<String, Object>) response.get("product");
Map<String, Object> instanceInfo = (Map<String, Object>) response.get("instanceInfo");
ProductDTO product = new ProductDTO();
product.setId(Long.valueOf(productMap.get("id").toString()));
product.setName((String) productMap.get("name"));
product.setCategory((String) productMap.get("category"));
product.setPrice(Double.valueOf(productMap.get("price").toString()));
product.setStock(Integer.valueOf(productMap.get("stock").toString()));
product.setDescription((String) productMap.get("description"));
logger.info("通过负载均衡调用商品服务,实例信息: {}", instanceInfo);
return product;
}
return null;
} catch (Exception e) {
throw new RuntimeException("通过负载均衡调用商品服务失败: " + e.getMessage());
}
}
/**
* 获取所有商品服务实例信息
*/
public List<ServiceInstance> getProductServiceInstances() {
return loadbalancerUtil.getDiscoveryClient("product-service");
}
/**
* 获取当前选择的商品服务实例
*/
public ServiceInstance getCurrentProductServiceInstance() {
return loadbalancerUtil.getLoadBalancerClient().choose("product-service");
}
/**
* 使用负载均衡调用用户服务
*/
public UserDTO getUserWithLoadBalancer(Long userId) {
// String url = "http://user-service/api/users/" + userId;
String userServiceId = "user-service";
String url = String.format("%s/api/users/%s", loadbalancerUtil.getCurrentServiceUrl(userServiceId), userId);
try {
return loadbalancerUtil.getLoadBalancedRestClient().get()
.uri(url)
.retrieve()
.body(UserDTO.class);
} catch (Exception e) {
throw new RuntimeException("通过负载均衡调用用户服务失败: " + e.getMessage());
}
}
/**
* 测试负载均衡 - 连续调用多次观察分布
*/
public Map<String, Object> testLoadBalancing(int callCount) {
Map<String, Integer> distribution = new java.util.HashMap<>();
List<Map<String,Object>> details = new ArrayList<>();
String productServiceId = "product-service";
for (int i = 0; i < callCount; i++) {
// String url = "http://product-service/api/products/instance-info";
String url = String.format("%s/api/products/instance-info", loadbalancerUtil.getCurrentServiceUrl(productServiceId));
try {
// 定义泛型类型:Map<String, Object>
ParameterizedTypeReference<Map<String, Object>> typeRef = new ParameterizedTypeReference<Map<String, Object>>() {
};
Map<String, Object> response = loadbalancerUtil
.getLoadBalancedRestClient()
.get()
.uri(url)
.retrieve()
.body(typeRef);
if (response != null) {
String instanceId = (String) response.get("instanceId");
distribution.put(instanceId, distribution.getOrDefault(instanceId, 0) + 1);
details.add(response);
}
// 短暂延迟,避免请求过快
Thread.sleep(10);
} catch (Exception e) {
logger.error("负载均衡测试调用失败: {}", e.getMessage());
}
}
return Map.of(
"totalCalls", callCount,
"distribution", distribution,
"details", details,
"instances", loadbalancerUtil.getDiscoveryClient(productServiceId).size()
);
}
}
订单控制器增强
package com.lihaozhe.orderservice.controller;
import com.lihaozhe.orderservice.dto.OrderDTO;
import com.lihaozhe.orderservice.dto.OrderItemDTO;
import com.lihaozhe.orderservice.service.LoadBalancerService;
import com.lihaozhe.orderservice.util.LoadbalancerUtil;
import com.lihaozhe.productservice.dto.ProductDTO;
import com.lihaozhe.userservice.dto.UserDTO;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.cloud.client.ServiceInstance;
import org.springframework.web.bind.annotation.*;
import java.util.ArrayList;
import java.util.List;
import java.util.Map;
import java.util.stream.Collectors;
import java.util.stream.IntStream;
/**
* 订单服务控制器 - 负载均衡版本
* 演示负载均衡的各种使用方式
*
* @author 李昊哲
* @version 1.0.0
*/
@RestController
@RequestMapping("/api/orders")
public class OrderController {
private final LoadBalancerService loadBalancerService;
private final List<OrderDTO> orders = new ArrayList<>();
private Long orderIdCounter = 1L;
private final Logger logger = LoggerFactory.getLogger(this.getClass());
private final LoadbalancerUtil loadbalancerUtil;
/**
* 构造函数,注入服务发现工具类
*/
public OrderController(LoadBalancerService loadBalancerService, LoadbalancerUtil loadbalancerUtil) {
this.loadBalancerService = loadBalancerService;
this.loadbalancerUtil = loadbalancerUtil;
}
/**
* 创建订单 - 使用负载均衡版本
* POST /api/orders
* 演示调用用户服务和商品服务
*/
@PostMapping
public Map<String, Object> createOrder(@RequestBody Map<String, Object> orderRequest) {
Long userId = Long.valueOf(orderRequest.get("userId").toString());
Long productId = Long.valueOf(orderRequest.get("productId").toString());
Integer quantity = Integer.valueOf(orderRequest.get("quantity").toString());
logger.info("创建订单 - 用户ID: {}, 商品ID: {}, 数量: {}", userId, productId, quantity);
try {
// 1. 使用负载均衡调用用户服务
UserDTO user = loadBalancerService.getUserWithLoadBalancer(userId);
if (user == null) {
return Map.of("success", false, "message", "用户不存在");
}
// 2. 使用负载均衡调用商品服务
ProductDTO product = loadBalancerService.getProductWithLoadBalancer(productId);
if (product == null) {
return Map.of("success", false, "message", "商品不存在");
}
// 3. 检查库存
if (product.getStock() < quantity) {
return Map.of("success", false, "message", "库存不足");
}
// 4. 创建订单
OrderItemDTO orderItem = new OrderItemDTO(productId, product.getName(), quantity, product.getPrice());
List<OrderItemDTO> items = List.of(orderItem);
Double totalAmount = product.getPrice() * quantity;
OrderDTO order = new OrderDTO(orderIdCounter++, userId, items, totalAmount, "CREATED");
orders.add(order);
return Map.of(
"success", true,
"message", "订单创建成功",
"orderId", order.getId(),
"totalAmount", totalAmount,
"user", user.getUsername(),
"product", product.getName()
);
} catch (Exception e) {
return Map.of("success", false, "message", "创建订单失败: " + e.getMessage());
}
}
/**
* 根据订单ID获取订单详情
* GET /api/orders/{orderId}
*/
@GetMapping("/{orderId}")
public OrderDTO getOrder(@PathVariable("orderId") Long orderId) {
return orders.stream()
.filter(o -> o.getId().equals(orderId))
.findFirst()
.orElse(null);
}
/**
* 获取用户的所有订单
* GET /api/orders/user/{userId}
*/
@GetMapping("/user/{userId}")
public List<OrderDTO> getUserOrders(@PathVariable("userId") Long userId) {
return orders.stream()
.filter(o -> o.getUserId().equals(userId))
.toList();
}
/**
* 负载均衡测试端点
* GET /api/orders/loadbalance-test?calls=20
*/
@GetMapping("/loadbalance-test")
public Map<String, Object> loadBalanceTest(@RequestParam(name = "calls", defaultValue = "20") int calls) {
return loadBalancerService.testLoadBalancing(calls);
}
/**
* GET /api/orders/testServiceInstance
*
* @return 商品服务地址
*/
@GetMapping("/testServiceInstance")
public List<String> testServiceInstance() {
return IntStream.range(0, 10)
.mapToObj(i -> loadbalancerUtil.getCurrentServiceUrl("product-service"))
.collect(Collectors.toCollection(ArrayList::new));
}
/**
* GET /api/orders/randomServiceInstance
*
* @return 随机商品服务地址
*/
@GetMapping("/randomServiceInstance")
public List<String> randomServiceInstance() {
return IntStream.range(0, 10)
.mapToObj(i -> loadbalancerUtil.randomServiceUrl("product-service"))
.collect(Collectors.toCollection(ArrayList::new));
}
/**
* 获取商品服务实例信息
* GET /api/orders/service-instances
*/
@GetMapping("/service-instances")
public Map<String, Object> getServiceInstances() {
List<ServiceInstance> instances = loadBalancerService.getProductServiceInstances();
ServiceInstance currentInstance = loadBalancerService.getCurrentProductServiceInstance();
List<Map<String, Object>> instanceInfo = instances.stream()
.map(instance -> Map.of(
"instanceId", instance.getInstanceId(),
"host", instance.getHost(),
"port", instance.getPort(),
"uri", instance.getUri().toString(),
"metadata", instance.getMetadata(),
"isCurrent", instance.equals(currentInstance)
)).toList();
return Map.of(
"serviceName", "product-service",
"totalInstances", instances.size(),
"currentInstance", currentInstance != null ? currentInstance.getInstanceId() : "none",
"instances", instanceInfo
);
}
/**
* 健康检查端点
* GET /api/orders/health
*/
@GetMapping("/health")
public Map<String, Object> health() {
return Map.of(
"status", "UP",
"service", "order-service",
"timestamp", System.currentTimeMillis(),
"totalOrders", orders.size(),
"loadBalancerEnabled", true
);
}
}
4. 自定义负载均衡配置
自定义负载均衡配置类
注意:
会覆盖配置文件中的策略,如有更复杂需求(如基于标签、延迟感知等),再考虑实现
ReactorServiceInstanceLoadBalancer接口。但对于 90% 的场景,配置文件 + 标准客户端已足够。
本案例没有使用。
如果是用 httpclient5 实现 RestClient 需要引入以下依赖
<!-- Apache HttpClient 5 -->
<dependency>
<groupId>org.apache.httpcomponents.client5</groupId>
<artifactId>httpclient5</artifactId>
</dependency>
<!-- 可选:如果需要连接池等高级功能 -->
<dependency>
<groupId>org.apache.httpcomponents.core5</groupId>
<artifactId>httpcore5</artifactId>
</dependency>
package com.lihaozhe.orderservice.config;
import org.springframework.cloud.client.ServiceInstance;
import org.springframework.cloud.client.loadbalancer.LoadBalanced;
import org.springframework.cloud.loadbalancer.core.ReactorLoadBalancer;
import org.springframework.cloud.loadbalancer.core.RoundRobinLoadBalancer;
import org.springframework.cloud.loadbalancer.core.ServiceInstanceListSupplier;
import org.springframework.cloud.loadbalancer.support.LoadBalancerClientFactory;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.core.env.Environment;
import org.springframework.web.client.RestClient;
import org.apache.hc.client5.http.config.ConnectionConfig;
import org.apache.hc.client5.http.impl.classic.CloseableHttpClient;
import org.apache.hc.client5.http.impl.classic.HttpClients;
import org.apache.hc.client5.http.impl.io.PoolingHttpClientConnectionManager;
import org.apache.hc.core5.util.Timeout;
import org.springframework.http.client.HttpComponentsClientHttpRequestFactory;
import org.springframework.http.client.JdkClientHttpRequestFactory;
import java.net.http.HttpClient;
import java.time.Duration;
/**
* 自定义负载均衡配置
* 可以配置不同的负载均衡策略
*
* @author 李昊哲
* @version 1.0.0
*/
@Configuration
public class LoadBalancerConfiguration {
/**
* 创建负载均衡的RestClient
* 使用@LoadBalanced注解后,RestClient会自动使用服务名进行服务发现和负载均衡
* 这个方法本身不实现负载均衡逻辑,只是“启用”负载均衡功能。
* 实际的负载均衡策略由另一个组件(即 ReactorLoadBalancer<ServiceInstance>)提供。
* 它是客户端入口,用于发起带服务发现 + 负载均衡的请求。
* 其他RestClient(非首选)指定名称
*
* @return 用于发起带服务发现 + 负载均衡请求的客户端入口
*/
@Bean("loadBalancedRestClient")
@LoadBalanced
public RestClient loadBalancedRestClient() {
return RestClient.builder().build();
}
/**
* 定义一个自定义的负载均衡器 Bean。
* Spring Cloud LoadBalancer 在需要为某个服务做实例选择时,
* 会尝试查找类型为 ReactorLoadBalancer<ServiceInstance> 的 Bean。
* 可以根据需要返回不同的负载均衡策略
* 此方法返回一个具体的负载均衡策略实现(如轮询或随机)。
*
* @param environment Spring 环境对象,用于获取当前上下文中的属性(如服务名)
* @param loadBalancerClientFactory 负载均衡客户端工厂,用于获取服务实例列表提供者(ServiceInstanceListSupplier)
* @return 一个实现了负载均衡逻辑的 ReactorLoadBalancer 实例(这里是轮询策略)
*/
@Bean
public ReactorLoadBalancer<ServiceInstance> customLoadBalancer(
Environment environment,
LoadBalancerClientFactory loadBalancerClientFactory) {
// 从当前 Spring 环境中获取正在被调用的服务名称。
// LoadBalancerClientFactory.PROPERTY_NAME 的值是 "loadbalancer.client.name"
// 这个属性由 Spring Cloud LoadBalancer 在内部自动设置,
// 表示当前正在处理哪个微服务的负载均衡请求(例如 "user-service")。
String name = environment.getProperty(LoadBalancerClientFactory.PROPERTY_NAME);
// 【可选】返回一个随机负载均衡器(每次从可用实例中随机选一个)
// 注释掉了,未启用
// return new RandomLoadBalancer(
// // getLazyProvider(name, ServiceInstanceListSupplier.class) 返回一个延迟初始化的 Provider,
// // 它能在需要时获取该服务(name)对应的服务实例列表(比如从 Eureka/Nacos 获取)
// loadBalancerClientFactory.getLazyProvider(name, ServiceInstanceListSupplier.class),
// name // 当前服务名,用于标识和日志等
// );
// 返回一个轮询(Round Robin)负载均衡器 —— 这是 Spring Cloud LoadBalancer 的默认策略
return new RoundRobinLoadBalancer(
// 同上:通过工厂获取该服务名对应的服务实例列表的延迟提供者(Lazy Provider)
// 这样可以避免在启动时就拉取所有服务列表,按需加载,提升性能
loadBalancerClientFactory.getLazyProvider(name, ServiceInstanceListSupplier.class),
name // 将服务名传入,用于内部状态管理(如轮询计数器按服务隔离)
);
}
/**
* - 连接超时时间
* - 读取超时时间
*/
@Bean("restClientWithJDK")
@LoadBalanced
public RestClient restClientWithJDK() {
IO.println("restClientWithJDK 创建成功");
// 配置JDK HttpClient
HttpClient jdkHttpClient = HttpClient.newBuilder().connectTimeout(Duration.ofSeconds(5)).build();
// 构建RestClient
return RestClient.builder().requestFactory(new JdkClientHttpRequestFactory(jdkHttpClient)).build();
}
/**
* - 连接超时时间
* - 读取超时时间
*/
@Bean("restClientWithH5")
@LoadBalanced
public RestClient restClientWithH5() {
IO.println("restClientWithH5 创建成功");
// 1. 配置Apache HttpClient的连接池和超时
PoolingHttpClientConnectionManager connectionManager = new PoolingHttpClientConnectionManager();
ConnectionConfig connectionConfig = ConnectionConfig.custom().setConnectTimeout(Timeout.ofSeconds(5)).build();
connectionManager.setDefaultConnectionConfig(connectionConfig);
CloseableHttpClient httpClient = HttpClients.custom().setConnectionManager(connectionManager).build();
// 2. 包装为RestClient所需的RequestFactory
HttpComponentsClientHttpRequestFactory requestFactory = new HttpComponentsClientHttpRequestFactory(httpClient);
// 3. 构建RestClient
return RestClient.builder()
.requestFactory(requestFactory) // 关联Apache HttpClient的配置
.build();
}
}
在订单服务配置中启用自定义负载均衡
# application.properties
#=============================================
# 0. 必改/高危配置(上线前一定要检查)
#=============================================
# 订单服务实例端口,集群部署时切忌重复
server.port=8083
#=============================================
# 1. Spring 基础配置
#=============================================
# Eureka 注册名,一旦上线禁止变更(消费者会硬编码)
spring.application.name=order-service
#=============================================
# 2. Eureka 实例级配置
#=============================================
# 在 Eureka 控制台显示的主机名,可读性优先
eureka.instance.hostname=order-service
# 注册时优先使用 IP,防止 hostname 解析失败
eureka.instance.prefer-ip-address=true
# 唯一标识,格式 IP:端口
eureka.instance.instance-id=${spring.cloud.client.ip-address}:${server.port}
# 心跳间隔 5s,缩短感知时间
eureka.instance.lease-renewal-interval-in-seconds=5
# 10s 内收不到心跳即剔除,开发调试可设短
eureka.instance.lease-expiration-duration-in-seconds=10
#=============================================
# 3. Eureka 客户端行为配置
#=============================================
# 是否把自己注册到 Eureka(默认 true,显式写出可提醒运维)
eureka.client.register-with-eureka=true
# 是否拉取注册表(默认 true,同上)
eureka.client.fetch-registry=true
# 注册中心地址(带安全认证)
eureka.client.service-url.defaultZone=http://admin:lihaozhe@localhost:8761/eureka/
#=============================================
# 4. LoadBalancer配置
#=============================================
# 为所有服务启用 Spring Cloud LoadBalancer
spring.cloud.loadbalance.enabled=true
# 禁用缓存,每次请求都从注册中心获取最新实例(生产环境不建议)
spring.cloud.loadbalance.cache.ttl=0
# 启用自定义负载均衡配置
# spring.cloud.loadbalancer.configurations=custom
#=============================================
# 5. Actuator 监控配置
#=============================================
# 按需暴露,生产勿暴露 env/beans
management.endpoints.web.exposure.include=health,info,metrics
# 默认 never,always 方便排查
management.endpoint.health.show-details=always
#=============================================
# 6. 日志级别
#=============================================
# 业务包日志级别,上线后改为 INFO 或 WARN
logging.level.com.lihaozhe.orderservice=DEBUG
# application.yml
#=============================================
# 0. 必改/高危配置(上线前一定要检查)
#=============================================
# 订单服务实例端口,集群部署时切忌重复
server:
port: 8083
#=============================================
# 1. Spring 基础配置
#=============================================
# Eureka 注册名,一旦上线禁止变更(消费者会硬编码)
spring:
application:
name: order-service
cloud:
loadbalancer:
# 为所有服务启用 Spring Cloud LoadBalancer
enabled: true
cache:
# 禁用缓存,每次请求都从注册中心获取最新实例(生产环境不建议)
ttl: 0
#=============================================
# 2. Eureka 实例级配置
#=============================================
# 在 Eureka 控制台显示的主机名,可读性优先
eureka:
instance:
hostname: order-service
# 注册时优先使用 IP,防止 hostname 解析失败
prefer-ip-address: true
# 唯一标识,格式 IP:端口
instance-id: ${spring.cloud.client.ip-address}:${server.port}
# 心跳间隔 5s,缩短感知时间
lease-renewal-interval-in-seconds: 5
# 10s 内收不到心跳即剔除,开发调试可设短
lease-expiration-duration-in-seconds: 10
#=============================================
# 3. Eureka 客户端行为配置
#=============================================
# 是否把自己注册到 Eureka(默认 true,显式写出可提醒运维)
client:
register-with-eureka: true
# 是否拉取注册表(默认 true,同上)
fetch-registry: true
service-url:
# 注册中心地址(带安全认证)
defaultZone: http://admin:lihaozhe@localhost:8761/eureka/
#=============================================
# 4. Actuator 监控配置
#=============================================
# 按需暴露,生产勿暴露 env/beans
management:
endpoints:
web:
exposure:
include: health,info,metrics
endpoint:
health:
# 默认 never,always 方便排查
show-details: always
#=============================================
# 5. 日志级别
#=============================================
# 业务包日志级别,上线后改为 INFO 或 WARN
logging:
level:
com.lihaozhe.orderservice: DEBUG
org.springframework.cloud.loadbalancer: DEBUG
5. Inventory Service库存服务(新增)
pom.xml
<?xml version="1.0" encoding="UTF-8"?>
<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">
<parent>
<artifactId>springcloud-course</artifactId>
<groupId>com.lihaozhe</groupId>
<version>1.0.0</version>
</parent>
<modelVersion>4.0.0</modelVersion>
<artifactId>inventory-service</artifactId>
<dependencies>
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-web</artifactId>
</dependency>
<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-loadbalancer</artifactId>
</dependency>
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-actuator</artifactId>
</dependency>
</dependencies>
<build>
<plugins>
<plugin>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-maven-plugin</artifactId>
<configuration>
<!-- 替换为你的启动类全路径 -->
<mainClass>com.lihaozhe.inventoryservice.InventoryServiceApplication</mainClass>
</configuration>
<!-- 可选:如果需要打包为可执行jar,添加此配置 -->
<executions>
<execution>
<goals>
<goal>repackage</goal>
</goals>
</execution>
</executions>
</plugin>
</plugins>
</build>
</project>
应用启动类
package com.lihaozhe.inventoryservice;
import org.springframework.boot.SpringApplication;
import org.springframework.boot.autoconfigure.SpringBootApplication;
import org.springframework.cloud.client.discovery.EnableDiscoveryClient;
import org.springframework.cloud.client.loadbalancer.LoadBalanced;
import org.springframework.context.annotation.Bean;
import org.springframework.web.client.RestClient;
/**
* 库存服务应用启动类
* 演示如何作为服务消费者使用负载均衡
*
* @author 李昊哲
* @version 1.0.0
*/
@SpringBootApplication
@EnableDiscoveryClient
public class InventoryServiceApplication {
public static void main(String[] args) {
SpringApplication.run(InventoryServiceApplication.class, args);
}
@Bean("loadBalancedRestClient")
@LoadBalanced
public RestClient loadBalancedRestClient() {
return RestClient.builder().build();
}
}
库存控制器
package com.lihaozhe.inventoryservice.controller;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.beans.factory.annotation.Value;
import org.springframework.cloud.client.ServiceInstance;
import org.springframework.cloud.client.loadbalancer.LoadBalancerClient;
import org.springframework.core.ParameterizedTypeReference;
import org.springframework.web.bind.annotation.*;
import org.springframework.web.client.RestClient;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
/**
* 库存服务控制器
* 演示库存服务如何通过负载均衡调用商品服务
*
* @author 李昊哲
* @version 1.0.0
*/
@RestController
@RequestMapping("/api/inventory")
public class InventoryController {
private final RestClient loadBalancedRestClient;
private final LoadBalancerClient loadBalancerClient;
private final Logger logger = LoggerFactory.getLogger(InventoryController.class);
@Value("${server.port}")
private String serverPort;
public InventoryController(RestClient loadBalancedRestClient, LoadBalancerClient loadBalancerClient) {
this.loadBalancedRestClient = loadBalancedRestClient;
this.loadBalancerClient = loadBalancerClient;
}
/**
* 检查商品库存 - 通过负载均衡调用商品服务
* GET /api/inventory/check/{productId}
*/
@GetMapping("/check/{productId}")
public Map<String, Object> checkInventory(@PathVariable("productId") Long productId) {
ServiceInstance instance = loadBalancerClient.choose("product-service");
// 使用服务名进行调用,LoadBalancer会自动处理负载均衡
String url = instance.getUri() + "/api/products/" + productId;
try {
// 定义泛型类型:Map<String, User>
ParameterizedTypeReference<Map<String, Object>> typeRef = new ParameterizedTypeReference<Map<String, Object>>() {
};
Map<String, Object> response = loadBalancedRestClient.get()
.uri(url)
.retrieve()
.body(typeRef);
if (response != null && response.get("product") != null) {
Map<String, Object> productMap = (Map<String, Object>) response.get("product");
Map<String, Object> instanceInfo = (Map<String, Object>) response.get("instanceInfo");
return Map.of(
"productId", productId,
"productName", productMap.get("name"),
"stock", productMap.get("stock"),
"price", productMap.get("price"),
"checkedBy", "inventory-service:" + serverPort,
"productServiceInstance", instanceInfo.get("instanceId"),
"timestamp", System.currentTimeMillis()
);
}
return Map.of(
"productId", productId,
"error", "商品不存在",
"checkedBy", "inventory-service:" + serverPort
);
} catch (Exception e) {
return Map.of(
"productId", productId,
"error", "检查库存失败: " + e.getMessage(),
"checkedBy", "inventory-service:" + serverPort
);
}
}
/**
* 批量检查库存
* POST /api/inventory/batch-check
*/
@PostMapping("/batch-check")
public Map<String, Object> batchCheckInventory(@RequestBody Map<String, List<Long>> request) {
List<Long> productIds = request.get("productIds");
Map<String, Object> results = new HashMap<>();
for (Long productId : productIds) {
Map<String, Object> checkResult = checkInventory(productId);
results.put("product_" + productId, checkResult);
}
return Map.of(
"totalChecked", productIds.size(),
"results", results,
"checkedBy", "inventory-service:" + serverPort
);
}
@GetMapping("/health")
public Map<String, Object> health() {
return Map.of(
"status", "UP",
"service", "inventory-service",
"port", serverPort,
"timestamp", System.currentTimeMillis()
);
}
}
库存服务配置
# application.properties
#=============================================
# 0. 必改/高危配置(上线前一定要检查)
#=============================================
# 库存服务实例端口,集群部署时切忌重复
server.port=8085
#=============================================
# 1. Spring 基础配置
#=============================================
# Eureka 注册名,一旦上线禁止变更(消费者会硬编码)
spring.application.name=inventory-service
#=============================================
# 2. Eureka 实例级配置
#=============================================
# 在 Eureka 控制台显示的主机名,可读性优先
eureka.instance.hostname=inventory-service
# 注册时优先使用 IP,防止 hostname 解析失败
eureka.instance.prefer-ip-address=true
# 唯一标识,格式 IP:端口
eureka.instance.instance-id=${spring.cloud.client.ip-address}:${server.port}
# 心跳间隔 5s,缩短感知时间
eureka.instance.lease-renewal-interval-in-seconds=5
# 10s 内收不到心跳即剔除,开发调试可设短
eureka.instance.lease-expiration-duration-in-seconds=10
#=============================================
# 3. Eureka 客户端行为配置
#=============================================
# 是否把自己注册到 Eureka(默认 true,显式写出可提醒运维)
eureka.client.register-with-eureka=true
# 是否拉取注册表(默认 true,同上)
eureka.client.fetch-registry=true
# 注册中心地址(带安全认证)
eureka.client.service-url.defaultZone=http://admin:lihaozhe@localhost:8761/eureka/
#=============================================
# 4. LoadBalancer配置
#=============================================
# 为所有服务启用 Spring Cloud LoadBalancer
spring.cloud.loadbalance.enabled=true
# 禁用缓存,每次请求都从注册中心获取最新实例(生产环境不建议)
spring.cloud.loadbalance.cache.ttl=0
# 启用自定义负载均衡配置
# spring.cloud.loadbalancer.configurations=custom
#=============================================
# 5. Actuator 监控配置
#=============================================
# 按需暴露,生产勿暴露 env/beans
management.endpoints.web.exposure.include=health,info,metrics
# 默认 never,always 方便排查
management.endpoint.health.show-details=always
#=============================================
# 6. 日志级别
#=============================================
# 业务包日志级别,上线后改为 INFO 或 WARN
logging.level.com.lihaozhe.inventoryservice=DEBUG
# application.yml
#=============================================
# 0. 必改/高危配置(上线前一定要检查)
#=============================================
# 库存服务实例端口,集群部署时切忌重复
server:
port: 8085
#=============================================
# 1. Spring 基础配置
#=============================================
# Eureka 注册名,一旦上线禁止变更(消费者会硬编码)
spring:
application:
name: inventory-service
security:
user:
# 开启 Spring Security 时的登录账号
name: admin
# 生产环境务必换成 16 位以上随机复杂密码
password: lihaozhe
cloud:
loadbalancer:
# 为所有服务启用 Spring Cloud LoadBalancer
enabled: true
# 启用自定义负载均衡配置
# configurations: custom
cache:
# 禁用缓存,每次请求都从注册中心获取最新实例(生产环境不建议)
ttl: 0
#=============================================
# 2. Eureka 实例级配置
#=============================================
# 在 Eureka 控制台显示的主机名,可读性优先
eureka:
instance:
hostname: inventory-service
# 注册时优先使用 IP,防止 hostname 解析失败
prefer-ip-address: true
# 唯一标识,格式 IP:端口
instance-id: ${spring.cloud.client.ip-address}:${server.port}
# 心跳间隔 5s,缩短感知时间
lease-renewal-interval-in-seconds: 5
# 10s 内收不到心跳即剔除,开发调试可设短
lease-expiration-duration-in-seconds: 10
#=============================================
# 3. Eureka 客户端行为配置
#=============================================
# 是否把自己注册到 Eureka(默认 true,显式写出可提醒运维)
client:
register-with-eureka: true
# 是否拉取注册表(默认 true,同上)
fetch-registry: true
service-url:
# 注册中心地址(带安全认证)
defaultZone: http://admin:lihaozhe@localhost:8761/eureka/
#=============================================
# 4. Actuator 监控配置
#=============================================
# 按需暴露,生产勿暴露 env/beans
management:
endpoints:
web:
exposure:
include: health,info,metrics
endpoint:
health:
# 默认 never,always 方便排查
show-details: always
#=============================================
# 5. 日志级别
#=============================================
# 业务包日志级别,上线后改为 INFO 或 WARN
logging:
level:
com.lihaozhe.inventoryservice: DEBUG
开发思路和过程
-
负载均衡原理:
- 客户端从注册中心获取所有可用服务实例
- 根据负载均衡策略选择一个实例
- 向选中的实例发送请求
-
实现策略:
- 使用
@LoadBalanced注解启用负载均衡 - 配置多个商品服务实例模拟真实环境
- 实现不同的负载均衡测试场景
- 使用
-
负载均衡策略:
- 轮询(Round Robin):依次选择每个实例
- 随机(Random):随机选择实例
- 权重(Weighted):根据实例权重分配
- 最少连接(Least Connections):选择连接数最少的实例
-
监控和调试:
- 添加实例信息输出便于观察
- 实现负载均衡测试端点
- 配置详细的日志输出
运行测试
启动顺序:
-
启动Eureka Server (端口8761)
-
启动User Service (端口8081)
-
启动Product Service实例1:
java -jar product-service-1.0.0.jar --spring.config.location=classpath:/application-8082.properties -
启动Product Service实例2:
java -jar .\product-service-1.0.0.jar --spring.config.location=classpath:/application-8084.properties -
启动Order Service (端口8083)
-
启动Inventory Service (端口8085)
测试步骤:
# 1. 查看Eureka注册中心,确认所有实例已注册
打开浏览器访问: http://localhost:8761
# 2. 测试负载均衡 - 连续调用观察实例分布
curl "http://localhost:8083/api/orders/loadbalance-test?calls=10"
# 3. 查看服务实例信息
curl http://localhost:8083/api/orders/service-instances
# 4. 创建订单(会自动负载均衡)
curl -X POST http://localhost:8083/api/orders \
-H "Content-Type: application/json" \
-d '{"userId": 1, "productId": 1, "quantity": 2}'
# 5. 测试手动实例选择
curl -X POST http://localhost:8083/api/orders/manual \
-H "Content-Type: application/json" \
-d '{"userId": 1, "productId": 1, "quantity": 1}'
# 6. 测试库存服务的负载均衡调用
curl http://localhost:8085/api/inventory/check/1
# 7. 批量检查库存
curl -X POST http://localhost:8085/api/inventory/batch-check \
-H "Content-Type: application/json" \
-d '{"productIds": [1, 2, 3]}'
观察负载均衡效果:
- 多次调用
/loadbalance-test端点,观察请求在不同实例间的分布 - 查看控制台日志,观察哪个实例处理了请求
- 在Eureka界面确认所有实例状态正常
这一章我们学习了LoadBalancer负载均衡的使用,实现了请求在多个服务实例间的智能分配。
在下一章中,我们将学习Feign客户端,它提供了更简洁的服务调用方式。
更多推荐
所有评论(0)