Kafka生产者与消费者练习测试题

需求数据如下：

student1,11100
student2,11200
student3,11300
student4,11400
student5,11500
student6,11600
student7,11700
student8,11800
student9,11900
student10,12000
student11,12100
student12,12200

一、通过Producer API发送到kafka中的【topicHW】
注：topic自行创建
二、创建一个Consumer API程序，对kafka集群中的【topicHW】进行消费。
处理消费到的数据，将消费到的数据发送到另外一个名为topicDEAL 的topic中 ，要求如下：

1. 创建名为【topicDEAL】的topic ，其拥有3分区2副本，将消费到的数据中以逗号为分隔的第二个值+1500，其余不变，放入到topicDEAL这个topic中
   最终要求结果如下 ：

student1,12600
student2,12700
student3,12800
student4,12900
...

注意，如果没有启动consumer就直接开始生产数据，则无法读取到刚刚生产到的数据，如果出现了这种情况，需要重新运行一下生产者的代码即可。
topicDEAL 这个topic中的数据消费在moba中处理。

一、启动环境

关闭Kafka

kafka-server-stop.sh stop

开启Kafka

[root@hadoop11 kafka0.11]# kafka-server-start.sh -daemon config/server.properties

注意：启动目录的绝对路径和相对路径

---

环境为三台Kafka集群

二、创建

1.创建topicHW，成功如下：

[root@hadoop11 kafka0.11]# kafka-topics.sh  --zookeeper hadoop11:2181  --create   --topic topicHW --partitions 3 --replication-factor 2
Created topic "topicHW".

【报错】创建topicHW时出现如下报错：

[root@hadoop11 ~]# kafka-topics.sh  --zookeeper hadoop11:2181  --create   --topic topicHW --partitions 3 --replication-factor 2
Error while executing topic command : replication factor: 2 larger than available brokers: 1

错误描述：副本数大于brokers数，排除了因为三台节点未完全启动的可能。
解决方案：删除已经存在的所有topic，将jps中已经运行的ConsoleConsumer进程kill -9。

[root@hadoop11 ~]# kafka-topics.sh  --zookeeper hadoop11:2181  --list
__consumer_offsets
topicA
topicHW
[root@hadoop11 ~]# kafka-topics.sh  --zookeeper hadoop11:2181  --delete   --topic topicHW
Topic topicHW is marked for deletion.
Note: This will have no impact if delete.topic.enable is not set to true.
[root@hadoop11 ~]# kafka-topics.sh  --zookeeper hadoop11:2181  --delete   --topic topicA
Topic topicA is marked for deletion.
Note: This will have no impact if delete.topic.enable is not set to true.
[root@hadoop11 ~]# kafka-topics.sh  --zookeeper hadoop11:2181  --list
__consumer_offsets
topicA - marked for deletion

[root@hadoop11 ~]# xcall.sh jps
要执行的命令是:jps
-----------------------hadoop11---------------------
51696 Jps
39028 ConsoleConsumer
42309 ConsoleConsumer
42118 QuorumPeerMain
41098 ConsoleProducer
45691 ConsoleConsumer
-----------------------hadoop12---------------------
10040 QuorumPeerMain
12872 Jps
-----------------------hadoop13---------------------
3512 QuorumPeerMain
5373 Jps
[root@hadoop11 ~]# kill -9 ^C
[root@hadoop11 ~]# kill -9 39028
[root@hadoop11 ~]# kill -9 42309
[1]   已杀死               kafka-console-consumer.sh --bootstrap-server hadoop11:9092 --topic topicHW
[root@hadoop11 ~]# kill -9 41098
[3]-  已杀死               kafka-console-consumer.sh --bootstrap-server hadoop11:9092 --topic topicHW
[root@hadoop11 ~]# kill -9 45691
[2]-  已杀死               kafka-console-producer.sh --broker-list hadoop11:9092 --topic topicHW

二、通过Producer API发送到kafka中的【topicHW】

package hw;

import org.apache.kafka.clients.producer.KafkaProducer;
import org.apache.kafka.clients.producer.ProducerConfig;
import org.apache.kafka.clients.producer.ProducerRecord;
import org.apache.kafka.common.serialization.StringSerializer;

import java.util.HashMap;
import java.util.Map;

public class CustomProducer {
    public static void main(String[] args) throws Exception {
        Map<String, Object> configs = new HashMap<>();
        configs.put(ProducerConfig.BOOTSTRAP_SERVERS_CONFIG, "hadoop11:9092,hadoop12:9092,hadoop13:9092");
        configs.put(ProducerConfig.KEY_SERIALIZER_CLASS_CONFIG, StringSerializer.class);
        configs.put(ProducerConfig.VALUE_SERIALIZER_CLASS_CONFIG, StringSerializer.class);
        //2. 创建生产者
        KafkaProducer<String, String> producer = new KafkaProducer<>(configs);
        //3. 发送数据
        int j = 11000;
        for (int i = 1; i <= 12; i++) {
            producer.send(new ProducerRecord<>("topicHW", "student" + i + "," + (j += 100)));
        }
        producer.close();
    }
}

linux终端显示生产者发送的数据

[root@hadoop11 kafka0.11]# kafka-console-consumer.sh --bootstrap-server hadoop11:9092 --topic topicHW
student1,11100
student2,11200
student3,11300
student4,11400
student5,11500
student6,11600
student7,11700
student8,11800
student9,11900
student10,12000
student11,12100
student12,12200

三、创建一个Consumer API程序，对kafka集群中的【topicHW】进行消费。

处理消费到的数据，将消费到的数据发送到另外一个名为topicDEAL 的topic中 ，要求如下：

1. 创建名为【topicDEAL】的topic ，其拥有3分区2副本，将消费到的数据中以逗号为分隔的第二个值+1500，其余不变，放入到topicDEAL这个topic中

结果如下 ：
student1,12600
student2,12700
student3,12800
student4,12900

[root@hadoop11 kafka0.11]# kafka-topics.sh  --zookeeper hadoop11:2181  --create   --topic topicDEAL --partitions 3 --replication-factor 2
Created topic "topicDEAL".

2.消费者接收并发送

package hw;

import com.sun.scenario.effect.impl.sw.sse.SSEBlend_SRC_OUTPeer;
import org.apache.kafka.clients.consumer.ConsumerConfig;
import org.apache.kafka.clients.consumer.ConsumerRecord;
import org.apache.kafka.clients.consumer.ConsumerRecords;
import org.apache.kafka.clients.consumer.KafkaConsumer;
import org.apache.kafka.clients.producer.KafkaProducer;
import org.apache.kafka.clients.producer.ProducerConfig;
import org.apache.kafka.clients.producer.ProducerRecord;
import org.apache.kafka.common.serialization.StringDeserializer;
import org.apache.kafka.common.serialization.StringSerializer;

import java.util.Arrays;
import java.util.HashMap;
import java.util.Map;

public class CustomConsumer {
    public static void main(String[] args) {
        //1. 初始化配置信息
        Map<String, Object> map = new HashMap<>();
        map.put(ConsumerConfig.BOOTSTRAP_SERVERS_CONFIG, "hadoop11:9092,hadoop12:9092,hadoop13:9092");
        map.put(ConsumerConfig.KEY_DESERIALIZER_CLASS_CONFIG, StringDeserializer.class);
        map.put(ConsumerConfig.VALUE_DESERIALIZER_CLASS_CONFIG, StringDeserializer.class);
        map.put(ConsumerConfig.GROUP_ID_CONFIG, "g00000");

        Map<String, Object> configs = new HashMap<>();
        configs.put(ProducerConfig.BOOTSTRAP_SERVERS_CONFIG, "hadoop11:9092,hadoop12:9092,hadoop13:9092");
        configs.put(ProducerConfig.KEY_SERIALIZER_CLASS_CONFIG, StringSerializer.class);
        configs.put(ProducerConfig.VALUE_SERIALIZER_CLASS_CONFIG, StringSerializer.class);

        //2. 创建生产者
        KafkaProducer<String, String> producer = new KafkaProducer<>(configs);

        //2. 创建Consumer
        KafkaConsumer<String, String> kafkaConsumer = new KafkaConsumer(map);
        //订阅 topic-user的数据
        kafkaConsumer.subscribe(Arrays.asList("topicHW"));

        while (true) {
            //3. 消费数据
            ConsumerRecords<String, String> consumerRecords = kafkaConsumer.poll(1000);
            for (ConsumerRecord<String, String> consumerRecord : consumerRecords) {
                String[] arr = consumerRecord.value().split(","); //切分
                int a = Integer.parseInt(arr[1]) + 1500;
                String data = arr[0] + "," + a;

                producer.send(new ProducerRecord<>("topicDEAL", data));
                System.out.println(data);
            }

        }
    }
}

如果出现数据格式化错误，需要改一下GROUP_ID，g00000改个名字
map.put(ConsumerConfig.GROUP_ID_CONFIG, "g00000");