示例环境

java.version: 1.8.x
flink.version: 1.11.1
kafka:2.11

示例数据源 (项目码云下载)

Flink 系例 之 搭建开发环境与数据

示例模块 (pom.xml)

Flink 系例 之 DataStream Connectors 与 示例模块

数据流输入

DataStreamSource.java

package com.flink.examples.kafka;

import com.flink.examples.TUser;
import com.google.gson.Gson;
import org.apache.commons.lang3.StringUtils;
import org.apache.flink.api.common.functions.FilterFunction;
import org.apache.flink.api.common.functions.MapFunction;
import org.apache.flink.api.common.serialization.SimpleStringSchema;
import org.apache.flink.streaming.api.CheckpointingMode;
import org.apache.flink.streaming.api.datastream.DataStream;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import org.apache.flink.streaming.connectors.kafka.FlinkKafkaConsumer;
import org.apache.kafka.clients.consumer.ConsumerConfig;

import java.util.Properties;

/**
 * @Description 从Kafka中消费数据
 */
public class DataStreamSource {

    /**
     * 官方文档:https://ci.apache.org/projects/flink/flink-docs-release-1.11/zh/dev/connectors/kafka.html
     */

    public static void main(String[] args) throws Exception {
        final StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
        //设置并行度(使用几个CPU核心)
        env.setParallelism(1);
        //每隔2000ms进行启动一个检查点
        env.enableCheckpointing(2000);
        //设置模式为exactly-once
        env.getCheckpointConfig().setCheckpointingMode(CheckpointingMode.EXACTLY_ONCE);
        // 确保检查点之间有进行500 ms的进度
        env.getCheckpointConfig().setMinPauseBetweenCheckpoints(500);

        //1.消费者客户端连接到kafka
        Properties props = new Properties();
        props.put(ConsumerConfig.BOOTSTRAP_SERVERS_CONFIG, "192.168.110.35:9092");
        props.put(ConsumerConfig.SESSION_TIMEOUT_MS_CONFIG, 5000);
        props.put(ConsumerConfig.GROUP_ID_CONFIG, "consumer-45");
        props.put(ConsumerConfig.AUTO_OFFSET_RESET_CONFIG, "latest");
        FlinkKafkaConsumer<String> consumer = new FlinkKafkaConsumer<>("test", new SimpleStringSchema(), props);
        //setStartFromEarliest()会从最早的数据开始进行消费,忽略存储的offset信息
        //consumer.setStartFromEarliest();
        //Flink从topic中指定的时间点开始消费,指定时间点之前的数据忽略
        //consumer.setStartFromTimestamp(1559801580000L);
        //Flink从topic中最新的数据开始消费
        //consumer.setStartFromLatest();
        //Flink从topic中指定的group上次消费的位置开始消费,所以必须配置group.id参数
        //consumer.setStartFromGroupOffsets();

        //2.在算子中进行处理
        DataStream<TUser> sourceStream = env.addSource(consumer)
                .filter((FilterFunction<String>) value -> StringUtils.isNotBlank(value))
                .map((MapFunction<String, TUser>) value -> {
                    System.out.println("print:" + value);
                    //注意,因已开启enableCheckpointing容错定期检查状态机制,当算子出现错误时,
                    //会导致数据流恢复到最新checkpoint的状态,并从存储在checkpoint中的offset开始重新消费Kafka中的消息。
                    //因此会有可能导制数据重复消费,重复错误,陷入死循环。加上try|catch,捕获错误后再正确输出。
                    Gson gson = new Gson();
                    try {
                        TUser user = gson.fromJson(value, TUser.class);
                        return user;
                    }catch(Exception e){
                        System.out.println("error:" + e.getMessage());
                    }
                    return new TUser();
                })
                .returns(TUser.class);
        sourceStream.print();

        //3.执行
        env.execute("flink  kafka source");
    }

}

数据流输出

DataStreamSink.java

package com.flink.examples.kafka;

import com.flink.examples.TUser;
import com.google.gson.Gson;
import org.apache.flink.api.common.functions.MapFunction;
import org.apache.flink.api.common.serialization.SimpleStringSchema;
import org.apache.flink.streaming.api.CheckpointingMode;
import org.apache.flink.streaming.api.datastream.DataStream;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import org.apache.flink.streaming.connectors.kafka.FlinkKafkaProducer;
import org.apache.kafka.clients.producer.ProducerConfig;

import java.util.Properties;

/**
 * @Description 将生产者数据写入到kafka
 */
public class DataStreamSink {

    /**
     * 官方文档:https://ci.apache.org/projects/flink/flink-docs-release-1.11/zh/dev/connectors/kafka.html
     */

    public static void main(String[] args) throws Exception {
        StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
        //必需设置setParallelism并行度,否则不会输出
        env.setParallelism(1);
        //每隔2000ms进行启动一个检查点
        env.enableCheckpointing(2000);
        //设置模式为exactly-once
        env.getCheckpointConfig().setCheckpointingMode(CheckpointingMode.EXACTLY_ONCE);
        // 确保检查点之间有进行500 ms的进度
        env.getCheckpointConfig().setMinPauseBetweenCheckpoints(500);
        // 检查点必须在一分钟内完成,或者被丢弃
        env.getCheckpointConfig().setCheckpointTimeout(60000);
        // 同一时间只允许进行一个检查点
        env.getCheckpointConfig().setMaxConcurrentCheckpoints(1);

        //1.连接kafka
        Properties props = new Properties();
        props.put(ProducerConfig.BOOTSTRAP_SERVERS_CONFIG, "192.168.110.35:9092");
        FlinkKafkaProducer<String> producer = new FlinkKafkaProducer<String>("test", new SimpleStringSchema(), props);

        //2.创建数据,并写入数据到流中
        TUser user = new TUser();
        user.setId(8);
        user.setName("liu3");
        user.setAge(22);
        user.setSex(1);
        user.setAddress("CN");
        user.setCreateTimeSeries(1598889600000L);
        DataStream<String> sourceStream = env.fromElements(user).map((MapFunction<TUser, String>) value -> new Gson().toJson(value));

        //3.将数据流输入到kafka
        sourceStream.addSink(producer);
        sourceStream.print();
        env.execute("flink kafka sink");
    }

}
  1. 在kafka上创建名称为test的topic
  2. 先启动DataStreamSource.java获取输出流,在启动DataStreamSink.java输入流

数据展示

Flink 系例 之 Connectors 连接 Kafka-LMLPHP

05-18 00:40