Storm 消费Kafka数据及相关异常解决

问题

storm报错：Exception in thread “main” java.lang.RuntimeException: InvalidTopologyException(msg:Component: [mybolt] subscribes from non-existent stream: [default] of component [kafka_spout])

storm版本：2.3.0
kafka版本：2.11

报错：mybolt这个组件，在从kafka_sput组件上消费消息时，它所消费的default数据流是不存在的。

KafkaTopoDemo类

main方法入口类和kafkaSpout设置

package com.suhaha.storm.storm122_kafka211_demo02;

import org.apache.kafka.clients.consumer.ConsumerConfig;
import org.apache.storm.Config;
import org.apache.storm.LocalCluster;
import org.apache.storm.StormSubmitter;
import org.apache.storm.generated.AlreadyAliveException;
import org.apache.storm.generated.AuthorizationException;
import org.apache.storm.generated.InvalidTopologyException;
import org.apache.storm.kafka.spout.*;
import org.apache.storm.topology.TopologyBuilder;
import org.apache.storm.tuple.Fields;
import org.apache.storm.tuple.Values;
import org.apache.storm.kafka.spout.KafkaSpoutRetryExponentialBackoff.TimeInterval;
import static org.apache.storm.kafka.spout.KafkaSpoutConfig.FirstPollOffsetStrategy.EARLIEST;

/**
 * @comment storm消费kafka数据
 */

public class KafkaTopoDemo {
    public static void main(String[] args) {
        final TopologyBuilder topologybuilder = new TopologyBuilder();
        //简单的不可靠spout
//        topologybuilder.setSpout("kafka_spout", new KafkaSpout<>(KafkaSpoutConfig.builder("node01:9092,node02:9092,node03:9092", "topic01").build()));

        //详细的设置spout,写一个方法生成KafkaSpoutConfig
        topologybuilder.setSpout("kafka_spout", new KafkaSpout<String,String>(newKafkaSpoutConfig("topic01")));

        topologybuilder.setBolt("mybolt", new MyBolt("/tmp/storm_test.log")).shuffleGrouping("kafka_spout");

        //上面设置的是topology，现在设置storm配置
        Config stormConf=new Config();
        stormConf.setNumWorkers(1);
        stormConf.setDebug(true);

        if (args != null && args.length > 0) {//集群提交
            System.out.println("【run on cluster】");

            try {
                StormSubmitter.submitTopology(args[0], stormConf, topologybuilder.createTopology());
            } catch (AlreadyAliveException e) {
                e.printStackTrace();
            } catch (InvalidTopologyException e) {
                e.printStackTrace();
            } catch (AuthorizationException e) {
                e.printStackTrace();
            }
            System.out.println("提交完成");

        } else {//本地提交
            System.out.println("【run on local】");
            LocalCluster lc = new LocalCluster();
            lc.submitTopology("storm_kafka", stormConf, topologybuilder.createTopology());
        }
    }


    /**
     * KafkaSpoutConfig设置
    */
    private static KafkaSpoutConfig<String,String> newKafkaSpoutConfig(String topic) {
        ByTopicRecordTranslator<String, String> trans = new ByTopicRecordTranslator<>(
                (r) -> new Values(r.topic(), r.partition(), r.offset(), r.key(), r.value()),
                new Fields("topic", "partition", "offset", "key", "value"), "stream1");
        //bootstrapServer 以及topic
        return KafkaSpoutConfig.builder("node01:9092,node02:9092,node03:9092", topic)
                .setProp(ConsumerConfig.GROUP_ID_CONFIG, "kafkaSpoutTestGroup_" + System.nanoTime())//设置kafka使用者组属性"group.id"
                .setProp(ConsumerConfig.MAX_PARTITION_FETCH_BYTES_CONFIG, 200)
                .setRecordTranslator(trans)//修改spout如何将Kafka消费者message转换为tuple,以及将该tuple发布到哪个stream中
                .setRetry(getRetryService())//重试策略
                .setOffsetCommitPeriodMs(10_000)
                .setFirstPollOffsetStrategy(EARLIEST)//允许你设置从哪里开始消费数据
                .setMaxUncommittedOffsets(250)
                .build();
    }

    /**
     * 重试策略设置
    */
    protected static KafkaSpoutRetryService getRetryService() {
        return new KafkaSpoutRetryExponentialBackoff(TimeInterval.microSeconds(500),
            TimeInterval.milliSeconds(2), Integer.MAX_VALUE, TimeInterval.seconds(10));
    }
}

bolt类

异常跟此bolt类无关

package com.suhaha.storm.storm122_kafka211_demo02;

import org.apache.storm.task.OutputCollector;
import org.apache.storm.task.TopologyContext;
import org.apache.storm.topology.IRichBolt;
import org.apache.storm.topology.OutputFieldsDeclarer;
import org.apache.storm.tuple.Tuple;
import java.io.FileWriter;
import java.io.IOException;
import java.util.Map;

/**
 * @comment  该bolt中的处理逻辑非常简单，只是简单的从input中将各类数据取出来，然后简单的打印出来
 *         并且将数据打印到path指定的文件中（这里需要注意的是，最终写出的文件是在执行该bolt task的worker上的，
 *      而不在nimbus服务器路径下，也不一定在提交storm job的服务器上）
 */

public class MyBolt implements IRichBolt {
    private FileWriter fileWriter = null;
    String path = null;

    @Override
    public void prepare(Map stormConf, TopologyContext context, OutputCollector collector) {
        try {
            fileWriter = new FileWriter(path);
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

    /**
     * 构造方法
     * @param path
     */
    public MyBolt(String path) {
        this.path = path;
    }


    @Override
    public void execute(Tuple input) {
        System.out.println(input);
        try {
            /**
             * 从input中获取相应数据
             */
            System.out.println("=========================");
            String topic = input.getString(0);
            System.out.println("index 0 --> " + topic);        //topic
            System.out.println("topic   --> " + input.getStringByField("topic"));

            System.out.println("-------------------------");
            System.out.println("index 1 --> " + input.getInteger(1));        //partition
            Integer partition = input.getIntegerByField("partition");
            System.out.println("partition-> " + partition);

            System.out.println("-------------------------");
            Long offset = input.getLong(2);
            System.out.println("index 2 --> " + offset);        //offset
            System.out.println("offset----> " +input.getLongByField("offset"));

            System.out.println("-------------------------");
            String key = input.getString(3);
            System.out.println("index 3 --> " + key);        //key
            System.out.println("key-------> " + input.getStringByField("key"));

            System.out.println("-------------------------");
            String value = input.getString(4);
            System.out.println("index 4 --> " + value);        //value
            System.out.println("value--> " + input.getStringByField("value"));

            String info = "topic: " + topic + ", partiton: " +partition + ", offset: " + offset + ", key: " + key +", value: " + value + "\n";
            System.out.println("info = " + info);
            fileWriter.write(info);
            fileWriter.flush();
        } catch (Exception e) {
            e.printStackTrace();
        }
    }

    @Override
    public void cleanup() {
        // TODO Auto-generated method stub
    }

    @Override
    public void declareOutputFields(OutputFieldsDeclarer declarer) {
        // TODO Auto-generated method stub
    }

    @Override
    public Map<String, Object> getComponentConfiguration() {
        // TODO Auto-generated method stub
        return null;
    }
}

错误出现在KafkaTopoDemo类中，

错误的原因在于，在代码中对RecordTranslator进行设置时（第67行），将数据流Id设置成了stream1；而在对topologyBuilder设置bolt时（第32行），使用的分发策略是shuffleGrouping(“kafka_spout”)，其实错误跟分发策略没关系，但是跟分发策略的使用方式有关系——当使用shuffleGrouping(String componentId)这种方式设置分发策略时，mybolt组件默认是从上游组件的default 这个数据流中获取数据，而在代码中，我已将上游（kafka_spout）的数据流id设置成了stream1，故而导致了报错（InvalidTopologyException(msg:Component: [mybolt] subscribes from non-existent stream: [default] of component [kafka_spout])，说default数据流不存在）。

修改：在设置mybolt组件的分发策略时，使用shuffleGrouping(String componentId, String streamId)，手动指定要读取的数据流id为stream1

topologybuilder.setBolt("mybolt", new MyBolt("/tmp/storm_test.log")).shuffleGrouping("kafka_spout", "stream1");