我嘗試使用 kafka 生產者發送 java 字符串消息。字符串消息是從Java Spark JavaPairDStream中提取的。JavaPairDStream<String, String> processedJavaPairStream = inputStream.mapToPair                 (record-> new Tuple2<>(record.key(), record.value())).mapValues(message -> message.replace('>', '#'));String outTopics = "outputTopic";String broker = "localhost:9092";Properties properties = new Properties();properties.put("bootstrap.servers", broker);properties.put(ProducerConfig.KEY_SERIALIZER_CLASS_CONFIG, "org.apache.kafka.common.serialization.StringSerializer");properties.put(ProducerConfig.VALUE_SERIALIZER_CLASS_CONFIG, "org.apache.kafka.common.serialization.StringSerializer");Producer<String, String> producer = new KafkaProducer<String, String>(properties);processedJavaPairStream.foreachRDD(rdd -> rdd.foreach(tuple2 -> {    ProducerRecord<String, String> message = new ProducerRecord<String, String>(outTopics, tuple2._1, tuple2._2);    System.out.println(message.key() + " : " + message.value()); //(1)    producer.send(message).get(); //(2)}));(1) 行正確打印消息字符串。但是當我使用 kafka 生產者發送這些消息（如（2）行）時，它會拋出如下異常，我無法理解這個異常。我確認 kafaka 生產者消息是<String,String>通過第 (1) 行輸入的。但為什么第(2)行會拋出這個異常呢？我是否錯過任何流程？