本记录为Springboot框架整合本地成功部署的模拟3节点kafka集群

编码语言为Java

Java版本为jdk21

1. 确保 Kafka 集群已正确启动​​

• ZooKeeper 集群（端口 2181、2182、2183）和 Kafka 集群（端口 9092、9093、9094）已正常运行。

• 验证集群状态：

  python
   代码解读
  复制代码
  # 查看 Broker 列表
  bin\windows\kafka-broker-api-versions.bat --bootstrap-server localhost:9092
  

​2. 添加 Kafka 依赖​​

在 pom.xml 中添加 Kafka 依赖：

xml
 代码解读
复制代码
<dependency>
    <groupId>org.springframework.kafkagroupId>
    <artifactId>spring-kafkaartifactId>
dependency>

3. 配置 Kafka 连接​​

在 application.yml（或 application.properties）中配置 Kafka：

yaml
 代码解读
复制代码
spring:
  kafka:
    bootstrap-servers: localhost:9092,localhost:9093,localhost:9093
    producer:
      key-serializer: org.apache.kafka.common.serialization.StringSerializer # Key的序列化器类
      value-serializer: org.apache.kafka.common.serialization.StringSerializer # Value的序列化器类
      retries: 3  # 生产失败重试次数
      acks: all   # 确保所有副本确认  0 最快但不持久  1 等待leader确认  all 等待所有副本确认
      batch-size: 30720000 # 30MB # 批量发送时的批次大小（字节）
      buffer-memory: 33554432 # 32MB # 生产者的内存缓冲区大小（字节）
    consumer:
      auto-offset-reset: earliest # 偏移量重置策略： earliest：从最早的记录开始消费  latest：从最新的记录开始消费
      key-deserializer: org.apache.kafka.common.serialization.StringDeserializer # Key的反序列化器类
      value-deserializer: org.apache.kafka.common.serialization.StringDeserializer # Value的反序列化器类
      enable-auto-commit: false # 手动提交 Offset
      # 每次poll()调用返回的最大消息条数
      max-poll-records: 10
      session:
        # 消费者会话超时时间，超时未发送心跳将被认为失联（毫秒）
        timeout:
          ms: 300000 # 5分钟

    listener:
      # 如果指定的主题不存在，是否让应用启动失败，false表示不会报错
      missing-topics-fatal: false
      # 消费模式：single=单条消息，batch=批量消费
      type: batch
      # 消费确认模式：
      # manual_immediate：手动确认消息，立即提交offset
      ack-mode: manual_immediate

4. 生产者代码​​

typescript
 代码解读
复制代码
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.kafka.core.KafkaTemplate;
import org.springframework.kafka.support.SendResult;
import org.springframework.stereotype.Service;

import java.util.concurrent.CompletableFuture;

/**
 * @author kangk
 * @description 消息生产者
 */
@Service
public class KafkaProducerService {

    @Autowired
    private KafkaTemplate<String, String> kafkaTemplate;

    // 没有key，在集群下，也只会到默认分区
    public void sendMessage(String topic, String message) {
        CompletableFuture<SendResult<String, String>> future =
                kafkaTemplate.send(topic, message);

        future.whenComplete((result, ex) -> {
            if (ex == null) {
                System.out.println("Sent to partition " +
                        result.getRecordMetadata().partition());
            } else {
                System.out.println("Failed: " + ex.getMessage());
            }
        });
    }

    public void sendMessage(String topic, String key, String message) {
        CompletableFuture<SendResult<String, String>> future =
                kafkaTemplate.send(topic, key, message);

        future.whenComplete((result, ex) -> {
            if (ex == null) {
                System.out.println("Sent to partition " +
                        result.getRecordMetadata().partition());
            } else {
                System.out.println("Failed: " + ex.getMessage());
            }
        });
    }
}

5. 消费者代码（手动提交）​​

ini
 代码解读
复制代码
import com.kangk.demo.comm.KafkaConstants;
import lombok.extern.slf4j.Slf4j;
import org.springframework.kafka.annotation.KafkaListener;
import org.springframework.kafka.support.Acknowledgment;
import org.springframework.stereotype.Service;

import java.util.List;

/**
 * @author kangk
 * @description 批量消息消费者
 */

@Service
@Slf4j
public class KafkaBatchConsumer {

    //监听多个 Topic  @KafkaListener(topics = {"topic1", "topic2"}, groupId = "group1")
    // autoStartup参数：是是否自动启动；=”true“：自动启动，即生产者启动，该消费者将会开始消费；=”false"：不自动启动，不开该模式的消费。
    @KafkaListener(
            topics = {KafkaConstants.TOPIC},
            groupId = KafkaConstants.GROUP_ID,
            containerFactory = "batchFactory",
            autoStartup = "true"
    )
    public void batchListen(List messages, Acknowledgment acknowledgment) {
        log.info("BatchConsumer - Received batch: " + messages);
        acknowledgment.acknowledge();
    }
}

6. 测试集群容错性​​

​​(1) 发送消息​​

kotlin
 代码解读
复制代码
import com.kangk.demo.comm.KafkaConstants;
import com.kangk.demo.kafka.KafkaProducerService;
import jakarta.annotation.Resource;
import org.springframework.web.bind.annotation.*;

import java.util.UUID;

/**
 * @author kangk
 * @description KafkaController
 */
@RestController
@RequestMapping("/api/1/kafka")
public class KafkaController {

    @Resource
    private KafkaProducerService producerService;

    @PostMapping("/send")
    public String sendMessage() {

        String key = UUID.randomUUID().toString();
        producerService.sendMessage(KafkaConstants.TOPIC, key, key);

        return "Message sent: " + message;
    }
}

arduino
 代码解读
复制代码
curl -X POST "http://localhost:8080/api/1/kafka/send"

控制台输出示例：

bash
 代码解读
复制代码
Sent to partition 1  # 消息可能被发送到任意 Broker 的分区

​​(2) 关闭一个 Broker 模拟故障​​

• 停止 localhost:9093 的 Kafka 进程。
• 再次发送消息，观察是否仍能正常生产/消费（其他 Broker 应接管流量）。

7. 代码补充配置​​

​​(1) containerFactory编写​​

typescript
 代码解读
复制代码
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.kafka.annotation.EnableKafka;
import org.springframework.kafka.config.ConcurrentKafkaListenerContainerFactory;
import org.springframework.kafka.core.ConsumerFactory;

/**
 * @author kangk
 * @description 消费监听器工厂
 */

@EnableKafka
@Configuration
public class KafkaConfig {

    // 单条消费监听器工厂，手动提交offset
    @Bean
    public ConcurrentKafkaListenerContainerFactory<String, String> singleFactory(
            ConsumerFactory<String, String> consumerFactory) {
        ConcurrentKafkaListenerContainerFactory<String, String> factory =
                new ConcurrentKafkaListenerContainerFactory<>();
        factory.setConsumerFactory(consumerFactory);
        factory.getContainerProperties().setAckMode(org.springframework.kafka.listener.ContainerProperties.AckMode.MANUAL_IMMEDIATE);
        return factory;
    }

    // 批量消费监听器工厂，手动提交offset
    @Bean
    public ConcurrentKafkaListenerContainerFactory<String, String> batchFactory(
            ConsumerFactory<String, String> consumerFactory) {
        ConcurrentKafkaListenerContainerFactory<String, String> factory =
                new ConcurrentKafkaListenerContainerFactory<>();
        factory.setConsumerFactory(consumerFactory);
        factory.setBatchListener(true); // 启用批量消费
        factory.setConcurrency(10);
        factory.getContainerProperties().setAckMode(org.springframework.kafka.listener.ContainerProperties.AckMode.MANUAL_IMMEDIATE);
        return factory;
    }

}

​​(2) 常量提取​​

java
 代码解读
复制代码

/**
 * @author kangk
 * @description kafka常量类
 */
public class KafkaConstants {

    private KafkaConstants() {}  // 防止实例化

    public static final String GROUP_ID = "my-demo-group";
    public static final String TOPIC = "test02";
}

​8.完整目录结构示例​​

bash
 代码解读
复制代码
src/main/java/
├── com/kangk/demo/
│   ├── config/
│   │   └── KafkaConfig.java       # 消费监听器工厂
│   ├── constants/
│   │   └── KafkaConstants.java    # 常量类
│   ├── controller/
│   │   ├── KafkaController.java
│   ├── kafka/
│   │   ├── KafkaBatchConsumer.java
│   │   └── KafkaConsumerService.java
│   └── DemoApplication.java
src/main/resources/
├── application.yml                # Kafka 配置