如何使用Avro和Flink解碼Kafka消息

Question

我試圖使用Flink 1.0.3從Kafka主題中讀取AVRO數據。

我只知道這個特定的Kafka主題是AVRO編碼消息，我有AVRO模式文件。

我的Flink代碼：

public static void main(String[] args) throws Exception {
        StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
        Properties properties = new Properties();
        properties.setProperty("bootstrap.servers", "dojo3xxxxx:9092,dojoxxxxx:9092,dojoxxxxx:9092");
        properties.setProperty("zookeeper.connect", "dojo3xxxxx:2181,dojoxxxxx:2181,dojoxxxxx:2181");
        properties.setProperty("group.id", "Zeeshantest");
        AvroDeserializationSchema<Event> avroSchema = new AvroDeserializationSchema<>(Event.class);
        FlinkKafkaConsumer08<Event> kafkaConsumer = new FlinkKafkaConsumer08<>("myavrotopic", avroSchema, properties);
        DataStream<Event> messageStream = env.addSource(kafkaConsumer);
        messageStream.rebalance().print();
        env.execute("Flink AVRO KAFKA Test");
    }

我使用avro工具和模式“rocana.avsc”創建了我的Event.java文件

java -jar /path/to/avro-tools-1.8.1.jar compile schema rocana.avsc

這是在github上傳的rocana.avsc文件。

AvroDeserializationSchema.java

import org.apache.avro.io.BinaryDecoder;
import org.apache.avro.io.DatumReader;
import org.apache.avro.io.DecoderFactory;
import org.apache.avro.reflect.ReflectDatumReader;
import org.apache.avro.specific.SpecificDatumReader;
import org.apache.flink.api.common.typeinfo.TypeInformation;
import org.apache.flink.api.java.typeutils.TypeExtractor;
import org.apache.flink.streaming.util.serialization.DeserializationSchema;

public class AvroDeserializationSchema<T> implements DeserializationSchema<T> {

    private static final long serialVersionUID = 4330538776656642778L;

    private final Class<T> avroType;
    private transient DatumReader<T> reader;
    private transient BinaryDecoder decoder;

    public AvroDeserializationSchema(Class<T> avroType) {
        this.avroType = avroType;
    }

    @Override
    public T deserialize(byte[] message) {
        ensureInitialized();
        try {
            decoder = DecoderFactory.get().binaryDecoder(message, decoder);
            return reader.read(null, decoder);
        } catch (Exception e) {
            throw new RuntimeException(e);
        }
    }

    @Override
    public boolean isEndOfStream(T nextElement) {
        return false;
    }

    @Override
    public TypeInformation<T> getProducedType() {
        return TypeExtractor.getForClass(avroType);
    }

    private void ensureInitialized() {
        if (reader == null) {
            if (org.apache.avro.specific.SpecificRecordBase.class.isAssignableFrom(avroType)) {
                reader = new SpecificDatumReader<T>(avroType);
            } else {
                reader = new ReflectDatumReader<T>(avroType);
            }
        }
    }
}

在運行我的程序時，我收到以下錯誤：

17:25:30,759 INFO  org.apache.zookeeper.ZooKeeper                                - Session: 0x356350cb9001857 closed
17:25:30,759 INFO  org.apache.zookeeper.ClientCnxn                               - EventThread shut down
17:25:30,761 INFO  org.apache.flink.runtime.taskmanager.Task                     - Source: Custom Source (3/4) switched to FAILED with exception.
java.lang.Exception: 2
    at org.apache.flink.streaming.connectors.kafka.internals.LegacyFetcher.run(LegacyFetcher.java:222)
    at org.apache.flink.streaming.connectors.kafka.FlinkKafkaConsumer08.run(FlinkKafkaConsumer08.java:316)
    at org.apache.flink.streaming.api.operators.StreamSource.run(StreamSource.java:78)
    at org.apache.flink.streaming.runtime.tasks.SourceStreamTask.run(SourceStreamTask.java:56)
    at org.apache.flink.streaming.runtime.tasks.StreamTask.invoke(StreamTask.java:225)
    at org.apache.flink.runtime.taskmanager.Task.run(Task.java:559)
    at java.lang.Thread.run(Unknown Source)
Caused by: java.lang.ArrayIndexOutOfBoundsException: 2
    at org.apache.avro.io.parsing.Symbol$Alternative.getSymbol(Symbol.java:402)
    at org.apache.avro.io.ResolvingDecoder.doAction(ResolvingDecoder.java:290)
    at org.apache.avro.io.parsing.Parser.advance(Parser.java:88)
    at org.apache.avro.io.ResolvingDecoder.readIndex(ResolvingDecoder.java:267)
    at org.apache.avro.generic.GenericDatumReader.read(GenericDatumReader.java:155)
    at org.apache.avro.generic.GenericDatumReader.readField(GenericDatumReader.java:193)
    at org.apache.avro.reflect.ReflectDatumReader.readField(ReflectDatumReader.java:230)
    at org.apache.avro.generic.GenericDatumReader.readRecord(GenericDatumReader.java:183)
    at org.apache.avro.generic.GenericDatumReader.read(GenericDatumReader.java:151)
    at org.apache.avro.generic.GenericDatumReader.read(GenericDatumReader.java:142)
    at com.google.cloud.dataflow.sdk.coders.AvroCoder.decode(AvroCoder.java:274)
    at org.fmr.flink.AvroDeserializationSchema.deserialize(AvroDeserializationSchema.java:52)
    at org.apache.flink.streaming.util.serialization.KeyedDeserializationSchemaWrapper.deserialize(KeyedDeserializationSchemaWrapper.java:39)
    at org.apache.flink.streaming.connectors.kafka.internals.LegacyFetcher$SimpleConsumerThread.run(LegacyFetcher.java:657)
17:25:30,769 INFO  org.apache.flink.runtime.taskmanager.Task                     - Freeing task resources for Source: Custom Source (3/4)
17:25:30,776 INFO  org.apache.flink.runtime.taskmanager.Task                     - Sink: Unnamed (1/4) switched

我認為我的反序列化代碼不正確。 誰知道我做錯了什么？ 這是使用Flink從Kafka讀取AVRO數據的方法還是有更好的方法？

Answer 1

嘗試使用以下代碼反序列化Avro記錄：

Schema a; //Your Avro schema
DatumReader<GenericData.Record> reader = new GenericDatumReader<GenericData.Record>(a);
GenericData.Record a = reader.read(null, DecoderFactory.get().binaryDecoder(bytes, null));

Answer 2

無論我根據我的問題發布的代碼是完全正常的。

問題是發送到kafka主題的數據，JSON和AVRO格式的數據都發送到那里。 我訂閱了一個不同的Kafka主題，其中數據僅在AVRO中，我的代碼工作正常。