Spark streaming 2.4.0获取org.apache.spark.sql.AnalysisException:找不到数据源: kafka

Question

尝试从Kafka读取数据时出现以下错误。我使用docker-compose来运行kafka和spark。

Exception in thread "main" org.apache.spark.sql.AnalysisException: Failed to find data source: kafka. Please deploy the application as per the deployment section of "Structured Streaming + Kafka Integration Guide".

下面是我用于阅读的代码：

object Livedata extends App with LazyLogging {
  logger.info("starting livedata...")
  val spark = SparkSession.builder().appName("livedata").master("local[*]").getOrCreate()

  val df = spark.readStream
        .format("kafka")
        .option("kafka.bootstrap.servers", "kafka:9092")
        .option("subscribe", "topic")
        .option("startingOffsets", "latest")
        .load()

  df.printSchema()

  val hadoopConfig = spark.sparkContext.hadoopConfiguration
  hadoopConfig.set("fs.hdfs.impl", classOf[org.apache.hadoop.hdfs.DistributedFileSystem].getName)
  hadoopConfig.set("fs.file.impl", classOf[org.apache.hadoop.fs.LocalFileSystem].getName)

}

在阅读了一些答案之后，我添加了用于sbt构建的所有包

这是build.sbt文件：

lazy val root = (project in file(".")).
  settings(
    inThisBuild(List(
      organization := "com.live.data",
      version := "0.1.0",
      scalaVersion := "2.12.2",
      assemblyJarName in assembly := "livedata.jar"
)),
    name := "livedata",
    libraryDependencies ++= List(
      "org.scalatest" %% "scalatest" % "3.0.5",
      "com.typesafe.scala-logging" %% "scala-logging" % "3.9.0",
      "org.apache.spark" %% "spark-sql" % "2.4.0",
      "org.apache.spark" %% "spark-sql-kafka-0-10" % "2.4.0" % "provided",
      "org.apache.kafka"           % "kafka-clients"            % "2.5.0",
      "org.apache.kafka"           % "kafka-streams"            % "2.5.0",
      "org.apache.kafka"           %% "kafka-streams-scala"     % "2.5.0"
)
)
assemblyMergeStrategy in assembly := {
  case PathList("META-INF", xs@_*) => MergeStrategy.discard
  case x => MergeStrategy.first
}

不确定这里的主要问题是什么。

更新：

主要问题是获取此org.apache.spark.sql.AnalysisException:找不到数据源: kafka异常，因为spark-sql-kafka库在类路径中不可用&它无法在META-INF/ org.apache.spark.sql.sources.DataSourceRegister文件夹中找到服务。

下面的代码块需要添加到build.sbt中。这将在最终的jar中包含org.apache.spark.sql.sources.DataSourceRegister文件。

// META-INF discarding
assemblyMergeStrategy in assembly := {
  case PathList("META-INF","services",xs @ _*) => MergeStrategy.filterDistinctLines
  case PathList("META-INF",xs @ _*) => MergeStrategy.discard
  case "application.conf" => MergeStrategy.concat
  case _ => MergeStrategy.first
}```

Answer 1

没有提供spark-sql-kafka-0-10，因此删除该部分依赖项。(虽然提供了spark-sql，因此您可以将其添加到其中)

您也不应该拉取Kafka流(因为Spark不使用它)，并且kafka客户端是由sql-kafka传递拉取的，所以也不需要它