java - 从磁盘读取 Spark 流错误 - java.io.NotSerializedException : org. apache.spark.streaming.api.java.JavaStreamingContext

Question

 Caused by: java.io.NotSerializableException: org.apache.spark.streaming.api.java.JavaStreamingContext
Serialization stack:
        - object not serializable (class: org.apache.spark.streaming.api.java.JavaStreamingContext, value: org.apache.spark.streaming.api.java.JavaStreamingContext@694c3f17)
        - field (class: com.emc.network.sparkanalysis.logic.ProcessVideoStreamData$2, name: val$jssc, type: class org.apache.spark.streaming.api.java.JavaStreamingContext)
        - object (class com.emc.network.sparkanalysis.logic.ProcessVideoStreamData$2, com.emc.network.sparkanalysis.logic.ProcessVideoStreamData$2@45abf747)
        - field (class: com.emc.network.sparkanalysis.logic.ProcessVideoStreamData$2$1, name: this$1, type: class com.emc.network.sparkanalysis.logic.ProcessVideoStreamData$2)
        - object (class com.emc.network.sparkanalysis.logic.ProcessVideoStreamData$2$1, com.emc.network.sparkanalysis.logic.ProcessVideoStreamData$2$1@5b02d2b0)
        - field (class: org.apache.spark.api.java.JavaRDDLike$$anonfun$fn$1$1, name: f$3, type: interface org.apache.spark.api.java.function.FlatMapFunction)
        - object (class org.apache.spark.api.java.JavaRDDLike$$anonfun$fn$1$1, <function1>)
        at org.apache.spark.serializer.SerializationDebugger$.improveException(SerializationDebugger.scala:40)
        at org.apache.spark.serializer.JavaSerializationStream.writeObject(JavaSerializer.scala:47)
        at org.apache.spark.serializer.JavaSerializerInstance.serialize(JavaSerializer.scala:84)
        at org.apache.spark.util.ClosureCleaner$.ensureSerializable(ClosureCleaner.scala:301)

我在这一行遇到错误:

JavaRDD<String> words = javarddPerIp.flatMap(new FlatMapFunction<String, String>()

<小时/>

JavaPairDStream<String, String> ones = stream
                .mapToPair(new PairFunction<String, String, String>() {
                    @Override
                    public Tuple2<String, String> call(String s) {
                        String[] lineArr = SPACE.split(s);
                        return new Tuple2<String, String>(lineArr[0], s);
                    }
                });

        JavaPairDStream<String, Iterable<String>> ipmapping = ones.groupByKey();

        ipmapping
                .foreachRDD(new Function2<JavaPairRDD<String, Iterable<String>>, Time, Void>() {
                    JavaRDD<String> finalLogs = jssc.sparkContext().emptyRDD();

                    @Override
                    public Void call(JavaPairRDD<String, Iterable<String>> v1,Time v2) throws Exception {

                        JavaRDD<Iterable<String>> stringValues = v1.values();
                        List<Iterable<String>> stringList = stringValues.collect();

                        for (Iterable<String> it : stringList) {
                            List<String> rddlist = new ArrayList<String>();
                            Iterator<String> values = it.iterator();
                            while (values.hasNext()) {
                                rddlist.add(values.next());
                            }

                            JavaRDD<String> javarddPerIp = jssc.sparkContext().parallelize(rddlist);

                            final Long numberOfrows;
                            numberOfrows = javarddPerIp.count();

                            System.out.println("javarddPerIp.count()"+ javarddPerIp.count());

                            JavaRDD<String> words = javarddPerIp
                                    .flatMap(new FlatMapFunction<String, String>() {
                                        @Override
                                        public Iterable<String> call(String s) {
                                            String[] splitstring = SPACE
                                                    .split(s);
                                            Double emcbitrate = Double
                                                    .parseDouble(splitstring[20])
                                                    / Double.parseDouble(splitstring[30]);
                                            StringBuffer sf = new StringBuffer();
                                            sf.append(emcbitrate.toString())
                                                    .append(SPACE)
                                                    .append(splitstring[44])
                                                    .append(SPACE)
                                                    .append(splitstring[51]);
                                            return Arrays.asList(s + SPACE
                                                    + emcbitrate.toString());
                                        }
                                    });

Answer 1

最后，下面的帖子帮助我解决了这个问题:删除内部函数并创建新类并提供实现:

org.apache.spark.SparkException: Task not serializable