WriteStream到csv无法正常工作出现错误

Question

val spark = SparkSession.builder().appName("app2").getOrCreate()

val df = spark
  .readStream
  .format("kafka")
  .option("kafka.bootstrap.servers", "host:port")
  .option("subscribe", args(0))
  .option("startingOffsets", "earliest")
  .load()
val checkPoint = "outputKafka/checkpoint/"

val newDf = df.selectExpr("CAST(value AS STRING)")

val df2 = newDf.withColumn("Primary_Type",split(col("value"),",")(5))
               .withColumn("Create_Date",split(col("value"),",")(2))

val pattern = "MM/dd/yyyy hh:mm:ss a"

val finalDf = df2.withColumn("DateNew", unix_timestamp(df2("Create_Date"), pattern)
                 .cast("timestamp"))

finalDf.withWatermark("DateNew","500 milliseconds")
  .groupBy(window(finalDf("DateNew"),"2 minutes"),col("Primary_Type"))
  .agg(count("*").as("Crime_Count"))
  .select("window.start","window.end","Primary_Type","Crime_Count")
  .writeStream.format("csv")
  .option("format", "append")
  .option("path",args(1)+"kafka.csv")
  .option("checkpointLocation",checkPoint)
  .option("header", "true")
  .outputMode("append")
  .trigger(Trigger.ProcessingTime(20.seconds)).start().awaitTermination()

摆脱错误

  

线程“ main” org.apache.spark.sql.AnalysisException中的异常：   流聚合时不支持追加输出模式   在不带水印的流式DataFrames / DataSet上;;

     

at org.apache.spark.sql.catalyst.analysis.UnsupportedOperationChecker $ .org $ apache $ spark $ sql $ catalyst $ analysis $ UnsupportedOperationChecker $$ throwError（UnsupportedOperationChecker.scala：374）           在org.apache.spark.sql.catalyst.analysis.UnsupportedOperationChecker $ .checkForStreaming（UnsupportedOperationChecker.scala：110）处           在org.apache.spark.sql.streaming.StreamingQueryManager.createQuery（StreamingQueryManager.scala：235）           在org.apache.spark.sql.streaming.StreamingQueryManager.startQuery（StreamingQueryManager.scala：299）           在org.apache.spark.sql.streaming.DataStreamWriter.start（DataStreamWriter.scala：296）           在com.mindtree.chicago.Question2 $ .main（Question2.scala：74）           在com.mindtree.chicago.Question2.main（Question2.scala）           在sun.reflect.NativeMethodAccessorImpl.invoke0（本机方法）处           在sun.reflect.NativeMethodAccessorImpl.invoke（NativeMethodAccessorImpl.java:62）           在sun.reflect.DelegatingMethodAccessorImpl.invoke（DelegatingMethodAccessorImpl.java:43）           在java.lang.reflect.Method.invoke（Method.java:498）           在org.apache.spark.deploy.JavaMainApplication.start（SparkApplication.scala：52）           在org.apache.spark.deploy.SparkSubmit $ .org $ apache $ spark $ deploy $ SparkSubmit $$ runMain（SparkSubmit.scala：894）中           在org.apache.spark.deploy.SparkSubmit $ .doRunMain $ 1（SparkSubmit.scala：198）           在org.apache.spark.deploy.SparkSubmit $ .submit（SparkSubmit.scala：228）           在org.apache.spark.deploy.SparkSubmit $ .main（SparkSubmit.scala：137）           在org.apache.spark.deploy.SparkSubmit.main（SparkSubmit.scala）