kafka使用（1）

1ConsumerRecord

@Component
public class ConsumerDemo {
    @KafkaListener(topics = "topicTEST")
    public void listen (ConsumerRecord<?, ?> record){
        System.out.printf("topic is %s, offset is %d, timestamp is %s, value is %s \n", record.topic(), record.offset(), record.timestamp(),record.value());
    }
}

假设我们有3个kafka broker分别brokerA、brokerB、brokerC.

1. 当我们创建的topic有3个分区partition时并且replication-factor为1，基本上一个broker上一个分区。挡一个broker宕机了，该topic就无法使用了，因为三个分区只有两个能用，

2. 当我们创建的topic有3个分区partition时并且replication-factor为2时，可能分区数据分布情况是

2创建主题

2.1注解方式

    //创建TopicName为topic.quick.initial的Topic并设置分区数为8以及副本数为1
    @Bean
    public NewTopic initialTopic() {
        return new NewTopic("topic.quick.initial",8, (short) 1 );
    }

2.2KafkaAdmin和AdminClient

    @Bean
    public KafkaAdmin kafkaAdmin() {
        Map<String, Object> props = new HashMap<>();
        //配置Kafka实例的连接地址
        props.put(AdminClientConfig.BOOTSTRAP_SERVERS_CONFIG, "192.168.228.128:9090");
        KafkaAdmin admin = new KafkaAdmin(props);
        return admin;
    }
?
    @Bean
    public AdminClient adminClient() {
        return AdminClient.create(kafkaAdmin().getConfigurationProperties());
    }

    @Autowired
    private AdminClient adminClient;
?
    @Test
    public void testCreateTopic() throws InterruptedException {
        NewTopic topic = new NewTopic("topic.quick.initial2", 1, (short) 1);
        adminClient.createTopics(Arrays.asList(topic));
        Thread.sleep(1000);
    }

• 结果：

2.3为什么要分区

分区提高系统的吞吐量，但是分区的参数只能增大，不能减小。

2.4获得主题的信息

    /**
     * 获得主题的信息
     */
    @Test
    public void testSelectTopicInfo() throws ExecutionException, InterruptedException {
        DescribeTopicsResult topicTest = adminClient.describeTopics(Arrays.asList("topic.quick.initial"));
        topicTest.all().get().forEach((k,v)->{
            System.out.println("k: "+k+" ,v: "+v.toString()+"\n");
        });
    }

k: topic.quick.initial ,v: (name=topic.quick.initial, internal=false, partitions=(partition=0, leader=kafka:9090 (id: 1 rack: null), replicas=kafka:9090 (id: 1 rack: null), isr=kafka:9090 (id: 1 rack: null)),(partition=1, leader=kafka:9090 (id: 1 rack: null), replicas=kafka:9090 (id: 1 rack: null), isr=kafka:9090 (id: 1 rack: null)),(partition=2, leader=kafka:9090 (id: 1 rack: null), replicas=kafka:9090 (id: 1 rack: null), isr=kafka:9090 (id: 1 rack: null)),(partition=3, leader=kafka:9090 (id: 1 rack: null), replicas=kafka:9090 (id: 1 rack: null), isr=kafka:9090 (id: 1 rack: null)),(partition=4, leader=kafka:9090 (id: 1 rack: null), replicas=kafka:9090 (id: 1 rack: null), isr=kafka:9090 (id: 1 rack: null)),(partition=5, leader=kafka:9090 (id: 1 rack: null), replicas=kafka:9090 (id: 1 rack: null), isr=kafka:9090 (id: 1 rack: null)),(partition=6, leader=kafka:9090 (id: 1 rack: null), replicas=kafka:9090 (id: 1 rack: null), isr=kafka:9090 (id: 1 rack: null)),(partition=7, leader=kafka:9090 (id: 1 rack: null), replicas=kafka:9090 (id: 1 rack: null), isr=kafka:9090 (id: 1 rack: null)), authorizedOperations=null)
?

3自定义发送消息监听器

@Configuration
public class KafkaSendListenerHandler implements ProducerListener {
    private static final Logger log = LoggerFactory.getLogger(KafkaSendListenerHandler.class);
?
?
    @Override
    public void onSuccess(ProducerRecord producerRecord, RecordMetadata recordMetadata) {
        log.info("监听消息发送成功...");
        String key = (String)producerRecord.key();
        log.info("key : " + key);
        log.info("Message send success : " + producerRecord.toString());
        log.info("-----------------------------");
    }
?
    @Override
    public void onError(ProducerRecord producerRecord, Exception exception) {
?
    }
}

4kafka事务管理

• KafkaTemplate 的 executeInTransaction 方法来声明事务

    @Test
    public void testExecuteInTransaction() throws InterruptedException {
        kafkaTemplate.executeInTransaction(new KafkaOperations.OperationsCallback() {
            @Override
            public Object doInOperations(KafkaOperations kafkaOperations) {
                kafkaOperations.send("topicTEST", "test executeInTransaction");
                throw new RuntimeException("fail");
                //return true;
            }
        });
    }

• 使用@Transactional注解方式使用注解方式开启事务，首先需要配置KafkaTransactionManager，这个类是Kafka提供事务管理类，需要使用生产者工厂来创建这个事务管理类。需要注意的是，在producerFactory中开启事务功能，并设置TransactionIdPrefix，TransactionIdPrefix是用来生成Transactional.id的前缀。

 @Bean
    public ProducerFactory<String, Object> producerFactory() {
        DefaultKafkaProducerFactory<String, Object> factory = new DefaultKafkaProducerFactory<>(producerConfigs());
        factory.transactionCapable();
        factory.setTransactionIdPrefix("tran-");
        return factory;
    }
?
    @Bean
    public Map<String, Object> producerConfigs() {
        Map<String, Object> props = new HashMap<>();
        props.put(ProducerConfig.BOOTSTRAP_SERVERS_CONFIG, "192.168.228.128:9090");
        props.put("acks", "all");
        props.put("key.serializer", "org.apache.kafka.common.serialization.StringSerializer");
        props.put("value.serializer", "org.apache.kafka.common.serialization.StringSerializer");
        return props;
    }
?
    @Bean
    public KafkaTemplate<String, Object> kafkaTemplate() {
        return new KafkaTemplate<String, Object>(producerFactory());
    }
?
    @Bean
    public KafkaTransactionManager transactionManager(ProducerFactory producerFactory) {
        KafkaTransactionManager manager = new KafkaTransactionManager(producerFactory);
        return manager;
    }

使用事务结果：

org.apache.kafka.common.KafkaException: Failing batch since transaction was aborted
at org.apache.kafka.clients.producer.internals.Sender.maybeSendAndPollTransactionalRequest(Sender.java:422) [kafka-clients-2.5.1.jar:na]
at org.apache.kafka.clients.producer.internals.Sender.runOnce(Sender.java:312) [kafka-clients-2.5.1.jar:na]
at org.apache.kafka.clients.producer.internals.Sender.run(Sender.java:239) [kafka-clients-2.5.1.jar:na]
at java.lang.Thread.run(Thread.java:748) ~[na:1.8.0_152]
?
2020-09-15 18:50:54.064  INFO 14636 --- [           main] o.s.t.c.transaction.TransactionContext   : Rolled back transaction for test: ....

kafka使用（1）

原文：https://www.cnblogs.com/Lambquan/p/13675100.html