Flink将数据写入到hdfs中

1. 场景

随机产生数据然后将产生的数据写入到hdfs 中。

2. 随机数据源

代码：

package com.wudl.flink.hdfs.source;

import org.apache.flink.api.java.tuple.Tuple;
import org.apache.flink.api.java.tuple.Tuple4;
import org.apache.flink.streaming.api.functions.source.SourceFunction;

import java.text.SimpleDateFormat;
import java.util.Date;
import java.util.Random;

/**
 * @author ：wudl
 * @date ：Created in 2021-12-27 0:29
 * @description：
 * @modified By：
 * @version: 1.0
 */

public  class MySource implements SourceFunction<String> {
    private boolean isRunning = true;

    String[] citys = {"北京","广东","山东","江苏","河南","上海","河北","浙江","香港","山西","陕西","湖南","重庆","福建","天津","云南","四川","广西","安徽","海南","江西","湖北","山西","辽宁","内蒙古"};
    int i = 0;
    @Override
    public void run(SourceContext<String> ctx) throws Exception {

        Random random = new Random();
        SimpleDateFormat df = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss");
        while (isRunning) {
            int number = random.nextInt(4) + 1;
            Integer id =  i++;
            String eventTime = df.format(new Date());
            String address = citys[random.nextInt(citys.length)];
            int productId = random.nextInt(25);
            ctx.collect(id+","+eventTime +","+ address +","+ productId);
            Thread.sleep(500);
        }
    }
    @Override
    public void cancel() {

        isRunning = false;
    }
}

3. hdfssink

需要注意的怎么设置文件的前缀和后缀以及 文件的大小 。

package com.wudl.flink.hdfs.sink;

import org.apache.flink.api.common.serialization.SimpleStringEncoder;
import org.apache.flink.api.connector.sink.Sink;
import org.apache.flink.api.java.tuple.Tuple4;
import org.apache.flink.connector.file.sink.FileSink;
import org.apache.flink.core.fs.Path;
import org.apache.flink.streaming.api.functions.sink.filesystem.OutputFileConfig;
import org.apache.flink.streaming.api.functions.sink.filesystem.bucketassigners.DateTimeBucketAssigner;
import org.apache.flink.streaming.api.functions.sink.filesystem.rollingpolicies.DefaultRollingPolicy;

import java.util.concurrent.TimeUnit;

/**
 * @author ：wudl
 * @date ：Created in 2021-12-26 23:49
 * @description：HdfsSink
 * @modified By：
 * @version: 1.0
 */

public class HdfsSink {
    public static FileSink<String> getHdfsSink() {
        OutputFileConfig config = OutputFileConfig
                .builder()
                .withPartPrefix("wudl")
                .withPartSuffix(".txt")
                .build();
        FileSink<String> finkSink = FileSink.forRowFormat(new Path("hdfs://192.168.1.161:8020/FlinkFileSink/wudlfile"), new SimpleStringEncoder<String>("UTF-8"))
                .withRollingPolicy(DefaultRollingPolicy.builder()
                        每隔15分钟生成一个新文件
                        .withRolloverInterval(TimeUnit.MINUTES.toMinutes(1))
                        //每隔5分钟没有新数据到来,也把之前的生成一个新文件
                        .withInactivityInterval(TimeUnit.MINUTES.toMinutes(5))
                        .withMaxPartSize(1024 * 1024 * 1024)
                        .build())
                .withOutputFileConfig(config)
                .withBucketAssigner(new DateTimeBucketAssigner("yyyy-MM-dd"))
                .build();

        return  finkSink;
    }
}

4.主类

package com.wudl.flink.hdfs;

import com.wudl.flink.hdfs.sink.HdfsSink;
import com.wudl.flink.hdfs.source.MySource;
import org.apache.commons.lang3.SystemUtils;
import org.apache.flink.api.common.restartstrategy.RestartStrategies;
import org.apache.flink.api.common.time.Time;
import org.apache.flink.api.java.tuple.Tuple4;
import org.apache.flink.connector.file.sink.FileSink;
import org.apache.flink.runtime.state.filesystem.FsStateBackend;
import org.apache.flink.streaming.api.CheckpointingMode;
import org.apache.flink.streaming.api.datastream.DataStreamSource;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;

import java.util.concurrent.TimeUnit;

/**
 * @author ：wudl
 * @date ：Created in 2021-12-27 0:12
 * @description：
 * @modified By：
 * @version: 1.0
 */

public class Application {
    public static void main(String[] args) throws Exception {

        StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
        env.setParallelism(1);
        env.getCheckpointConfig().setCheckpointStorage("hdfs://192.168.1.161:8020/flink-checkpoint/checkpoint");
        //设置两个Checkpoint 之间最少等待时间,如设置Checkpoint之间最少是要等 500ms(为了避免每隔1000ms做一次Checkpoint的时候,前一次太慢和后一次重叠到一起去了  --//默认是0
        env.getCheckpointConfig().setMinPauseBetweenCheckpoints(500);
        //设置Checkpoint的时间间隔为1000ms做一次Checkpoint/其实就是每隔1000ms发一次Barrier!
        env.getCheckpointConfig().setCheckpointingMode(CheckpointingMode.EXACTLY_ONCE);
        //设置checkpoint的超时时间,如果 Checkpoint在 60s内尚未完成说明该次Checkpoint失败,则丢弃。
        env.getCheckpointConfig().setCheckpointTimeout(10000L);
        //设置同一时间有多少个checkpoint可以同时执行
        env.getCheckpointConfig().setMaxConcurrentCheckpoints(2);
        // 设置重启策略
        // 一个时间段内的最大失败次数
        env.setRestartStrategy(RestartStrategies.failureRateRestart(3,
                // 衡量失败次数的是时间段
                Time.of(5, TimeUnit.MINUTES),
                // 间隔
                Time.of(10, TimeUnit.SECONDS)
        ));


        DataStreamSource<String> mySouceData = env.addSource(new MySource());
        FileSink<String> hdfsSink = HdfsSink.getHdfsSink();

        mySouceData.sinkTo(hdfsSink );
        mySouceData.print();


        env.execute();

    }
}

结果