Spark-kafka：从Spark写入流时的org.apache.kafka.common.errors.TimeoutException

Question

我在撰写关于Spark主题的流时遇到了一个问题。

import org.apache.spark.sql.types._

val mySchema = StructType(Array(
  StructField("ID", IntegerType),
  StructField("ACCOUNT_NUMBER", StringType)
))

val streamingDataFrame = spark.readStream.schema(mySchema).option("delimiter",",")
                              .csv("file:///opt/files")


streamingDataFrame.selectExpr("CAST(id AS STRING) AS key", "to_json(struct(*)) AS value")
                  .writeStream.format("kafka")
                  .option("topic", "testing")
                  .option("kafka.bootstrap.servers", "10.55.55.55:9092")
                  .option("checkpointLocation", "file:///opt/")
                  .start().awaitTermination()

错误：

 2018-09-12 11:09:04,344 ERROR executor.Executor: Exception in task 0.0 in stage 0.0 (TID 0)
org.apache.kafka.common.errors.TimeoutException: Expiring 38 record(s) for testing-0: 30016 ms has passed since batch creation plus linger time
2018-09-12 11:09:04,358 WARN scheduler.TaskSetManager: Lost task 0.0 in stage 0.0 (TID 0, localhost, executor driver):                                            org.apache.kafka.common.errors.TimeoutException: Expiring 38 record(s) for testing-0: 30016 ms has passed since batch creation plus linger time
2018-09-12 11:09:04,359 ERROR scheduler.TaskSetManager: Task 0 in stage 0.0 failed 1 times; aborting job
2018-09-12 11:09:04,370 ERROR streaming.StreamExecution: Query [id = 866e4416-138a-42b6-82fd-04b6ee1aa638, runId = 4dd10740-29dd-4275-97e2-a43104d71cf5] terminated with error
 org.apache.spark.SparkException: Job aborted due to stage failure: Task 0 in stage 0.0 failed 1 times, most recent failure: Lost task 0.0 in stage 0.0 (TID 0, localhost, executor driver):   org.apache.kafka.common.errors.TimeoutException: Expiring 38 record(s) for testing-0: 30016 ms has passed since batch creation plus linger time

Driver stacktrace:
    at org.apache.spark.scheduler.DAGScheduler.org$apache$spark$scheduler$DAGScheduler$$failJobAndIndependentStages(DAGScheduler.scala:1499)
    at org.apache.spark.scheduler.DAGScheduler$$anonfun$abortStage$1.apply(DAGScheduler.scala:1487)
    at org.apache.spark.scheduler.DAGScheduler$$anonfun$abortStage$1.apply(DAGScheduler.scala:1486)
    at scala.collection.mutable.ResizableArray$class.foreach(ResizableArray.scala:59)

我的sbt详细信息：

libraryDependencies += "org.apache.spark" %% "spark-sql-kafka-0-10" % "2.2.0"

libraryDependencies += "org.apache.kafka" % "kafka-clients" % "0.10.0.0"

但是当我使用bin/kafka-console-producer.sh和bin/kafka-console-consumer.sh通过服务器发送消息时，我可以发送和接收消息

Answer 1

您需要在客户端增加request.timeout.ms的值。

Kafka将记录分组成批次以提高吞吐量。将新记录添加到批处理中时，必须在时间限制内发送。 request.timeout.ms是一个可配置参数（默认值为30秒），用于控制此时间限制。

当批处理排队更长时间时，将抛出一个TimeoutException并从队列中删除记录（因此将不会传递消息）。