[jira] [Updated] (CARBONDATA-1726) Carbon1.3.0-Streaming - Select query from spark-shell does not execute successfully for streaming table load

[Akash R Nilugal (Jira)]

     [ https://issues.apache.org/jira/browse/CARBONDATA-1726?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Chetan Bhat updated CARBONDATA-1726:
------------------------------------
    Description:
Steps :
// prepare csv file for batch loading
cd /srv/spark2.2Bigdata/install/hadoop/datanode/bin

// generate streamSample.csv

100000001,batch_1,city_1,0.1,school_1:school_11$20
100000002,batch_2,city_2,0.2,school_2:school_22$30
100000003,batch_3,city_3,0.3,school_3:school_33$40
100000004,batch_4,city_4,0.4,school_4:school_44$50
100000005,batch_5,city_5,0.5,school_5:school_55$60

// put to hdfs /tmp/streamSample.csv
./hadoop fs -put streamSample.csv /tmp

// spark-beeline
cd /srv/spark2.2Bigdata/install/spark/sparkJdbc
bin/spark-submit --master yarn-client --executor-memory 10G --executor-cores 5 --driver-memory 5G --num-executors 3 --class org.apache.carbondata.spark.thriftserver.CarbonThriftServer /srv/spark2.2Bigdata/install/spark/sparkJdbc/carbonlib/carbondata_2.11-1.3.0-SNAPSHOT-shade-hadoop2.7.2.jar "hdfs://hacluster/user/sparkhive/warehouse"

bin/beeline -u jdbc:hive2://10.18.98.34:23040

CREATE TABLE stream_table(
id INT,
name STRING,
city STRING,
salary FLOAT
)
STORED BY 'carbondata'
TBLPROPERTIES('streaming'='true', 'sort_columns'='name');

LOAD DATA LOCAL INPATH 'hdfs://hacluster/chetan/streamSample.csv' INTO TABLE stream_table OPTIONS('HEADER'='false');

// spark-shell
cd /srv/spark2.2Bigdata/install/spark/sparkJdbc
bin/spark-shell --master yarn-client --executor-memory 10G --executor-cores 5 --driver-memory 5G --num-executors 3 --jars /srv/spark2.2Bigdata/install/spark/sparkJdbc/carbonlib/carbondata_2.11-1.3.0-SNAPSHOT-shade-hadoop2.7.2.jar

import java.io.{File, PrintWriter}
import java.net.ServerSocket

import org.apache.spark.sql.{CarbonEnv, SparkSession}
import org.apache.spark.sql.hive.CarbonRelation
import org.apache.spark.sql.streaming.{ProcessingTime, StreamingQuery}

import org.apache.carbondata.core.constants.CarbonCommonConstants
import org.apache.carbondata.core.util.CarbonProperties
import org.apache.carbondata.core.util.path.{CarbonStorePath, CarbonTablePath}

CarbonProperties.getInstance().addProperty(CarbonCommonConstants.CARBON_TIMESTAMP_FORMAT, "yyyy/MM/dd")

import org.apache.spark.sql.CarbonSession._

val carbonSession = SparkSession.
  builder().
  appName("StreamExample").
  config("spark.sql.warehouse.dir", "hdfs://hacluster/user/sparkhive/warehouse").
  config("javax.jdo.option.ConnectionURL", "jdbc:mysql://10.18.98.34:3306/sparksql?characterEncoding=UTF-8").
  config("javax.jdo.option.ConnectionDriverName", "com.mysql.jdbc.Driver").
  config("javax.jdo.option.ConnectionPassword", "huawei").
  config("javax.jdo.option.ConnectionUserName", "sparksql").
  getOrCreateCarbonSession()
   
carbonSession.sparkContext.setLogLevel("ERROR")

carbonSession.sql("select * from stream_table").show

def writeSocket(serverSocket: ServerSocket): Thread = {
  val thread = new Thread() {
    override def run(): Unit = {
      // wait for client to connection request and accept
      val clientSocket = serverSocket.accept()
      val socketWriter = new PrintWriter(clientSocket.getOutputStream())
      var index = 0
      for (_ <- 1 to 1000) {
        // write 5 records per iteration
        for (_ <- 0 to 100) {
          index = index + 1
          socketWriter.println(index.toString + ",name_" + index
                               + ",city_" + index + "," + (index * 10000.00).toString +
                               ",school_" + index + ":school_" + index + index + "$" + index)
        }
        socketWriter.flush()
        Thread.sleep(2000)
      }
      socketWriter.close()
      System.out.println("Socket closed")
    }
  }
  thread.start()
  thread
}
 
def startStreaming(spark: SparkSession, tablePath: CarbonTablePath): Thread = {
  val thread = new Thread() {
    override def run(): Unit = {
      var qry: StreamingQuery = null
      try {
        val readSocketDF = spark.readStream
          .format("socket")
          .option("host", "10.18.98.34")
          .option("port", 7071)
          .load()

        // Write data from socket stream to carbondata file
        qry = readSocketDF.writeStream
          .format("carbondata")
          .trigger(ProcessingTime("5 seconds"))
          .option("checkpointLocation", tablePath.getStreamingCheckpointDir)
          .option("tablePath", tablePath.getPath)
          .start()

        qry.awaitTermination()
      } catch {
        case _: InterruptedException =>
          println("Done reading and writing streaming data")
      } finally {
        qry.stop()
      }
    }
  }
  thread.start()
  thread
}

val streamTableName = s"stream_table"

val carbonTable = CarbonEnv.getInstance(carbonSession).carbonMetastore.
  lookupRelation(Some("default"), streamTableName)(carbonSession).asInstanceOf[CarbonRelation].
  tableMeta.carbonTable

val tablePath = CarbonStorePath.getCarbonTablePath(carbonTable.getAbsoluteTableIdentifier)

val serverSocket = new ServerSocket(7071)
val socketThread = writeSocket(serverSocket)
val streamingThread = startStreaming(carbonSession, tablePath)

**Issue : There is a null pointer exception when streaming is started.*

When the executor and driver cores and memory is increased while launching the spark shell the issue still occurs.
scala> import java.io.{File, PrintWriter}
import java.io.{File, PrintWriter}

scala> import java.net.ServerSocket
import java.net.ServerSocket

scala>

scala> import org.apache.spark.sql.{CarbonEnv, SparkSession}
import org.apache.spark.sql.{CarbonEnv, SparkSession}

scala> import org.apache.spark.sql.hive.CarbonRelation
import org.apache.spark.sql.hive.CarbonRelation

scala> import org.apache.spark.sql.streaming.{ProcessingTime, StreamingQuery}
import org.apache.spark.sql.streaming.{ProcessingTime, StreamingQuery}

scala>

scala> import org.apache.carbondata.core.constants.CarbonCommonConstants
import org.apache.carbondata.core.constants.CarbonCommonConstants

scala> import org.apache.carbondata.core.util.CarbonProperties
import org.apache.carbondata.core.util.CarbonProperties

scala> import org.apache.carbondata.core.util.path.{CarbonStorePath, CarbonTablePath}
import org.apache.carbondata.core.util.path.{CarbonStorePath, CarbonTablePath}

scala>

scala> CarbonProperties.getInstance().addProperty(CarbonCommonConstants.CARBON_TIMESTAMP_FORMAT, "yyyy/MM/dd")
res0: org.apache.carbondata.core.util.CarbonProperties = org.apache.carbondata.core.util.CarbonProperties@7212b28e

scala>

scala> import org.apache.spark.sql.CarbonSession._
import org.apache.spark.sql.CarbonSession._

scala>

scala> val carbonSession = SparkSession.
     |   builder().
     |   appName("StreamExample").
     |   config("spark.sql.warehouse.dir", "hdfs://hacluster/user/sparkhive/warehouse").
     |   config("javax.jdo.option.ConnectionURL", "jdbc:mysql://10.18.98.34:3306/sparksql?characterEncoding=UTF-8").
     |   config("javax.jdo.option.ConnectionDriverName", "com.mysql.jdbc.Driver").
     |   config("javax.jdo.option.ConnectionPassword", "huawei").
     |   config("javax.jdo.option.ConnectionUserName", "sparksql").
     |   getOrCreateCarbonSession()
carbonSession: org.apache.spark.sql.SparkSession = org.apache.spark.sql.CarbonSession@7593716d

scala>
     | carbonSession.sparkContext.setLogLevel("ERROR")

scala>

scala> carbonSession.sql("select * from stream_table").show
+---------+-------+------+------+
|       id|   name|  city|salary|
+---------+-------+------+------+
|100000001|batch_1|city_1|   0.1|
|100000002|batch_2|city_2|   0.2|
|100000003|batch_3|city_3|   0.3|
|100000004|batch_4|city_4|   0.4|
|100000005|batch_5|city_5|   0.5|
+---------+-------+------+------+


scala> def writeSocket(serverSocket: ServerSocket): Thread = {
     |   val thread = new Thread() {
     |     override def run(): Unit = {
     |       // wait for client to connection request and accept
     |       val clientSocket = serverSocket.accept()
     |       val socketWriter = new PrintWriter(clientSocket.getOutputStream())
     |       var index = 0
     |       for (_ <- 1 to 1000) {
     |         // write 5 records per iteration
     |         for (_ <- 0 to 100) {
     |           index = index + 1
     |           socketWriter.println(index.toString + ",name_" + index
     |                                + ",city_" + index + "," + (index * 10000.00).toString +
     |                                ",school_" + index + ":school_" + index + index + "$" + index)
     |         }
     |         socketWriter.flush()
     |         Thread.sleep(2000)
     |       }
     |       socketWriter.close()
     |       System.out.println("Socket closed")
     |     }
     |   }
     |   thread.start()
     |   thread
     | }
writeSocket: (serverSocket: java.net.ServerSocket)Thread

scala>
     | def startStreaming(spark: SparkSession, tablePath: CarbonTablePath): Thread = {
     |   val thread = new Thread() {
     |     override def run(): Unit = {
     |       var qry: StreamingQuery = null
     |       try {
     |         val readSocketDF = spark.readStream
     |           .format("socket")
     |           .option("host", "10.18.98.34")
     |           .option("port", 7071)
     |           .load()
     |
     |         // Write data from socket stream to carbondata file
     |         qry = readSocketDF.writeStream
     |           .format("carbondata")
     |           .trigger(ProcessingTime("5 seconds"))
     |           .option("checkpointLocation", tablePath.getStreamingCheckpointDir)
     |           .option("tablePath", tablePath.getPath)
     |           .start()
     |
     |         qry.awaitTermination()
     |       } catch {
     |         case _: InterruptedException =>
     |           println("Done reading and writing streaming data")
     |       } finally {
     |         qry.stop()
     |       }
     |     }
     |   }
     |   thread.start()
     |   thread
     | }
startStreaming: (spark: org.apache.spark.sql.SparkSession, tablePath: org.apache.carbondata.core.util.path.CarbonTablePath)Thread

scala>

scala> val streamTableName = s"stream_table"
streamTableName: String = stream_table

scala>

scala> val carbonTable = CarbonEnv.getInstance(carbonSession).carbonMetastore.
     |   lookupRelation(Some("default"), streamTableName)(carbonSession).asInstanceOf[CarbonRelation].
     |   tableMeta.carbonTable
carbonTable: org.apache.carbondata.core.metadata.schema.table.CarbonTable = org.apache.carbondata.core.metadata.schema.table.CarbonTable@62cf8fda

scala>

scala> val tablePath = CarbonStorePath.getCarbonTablePath(carbonTable.getAbsoluteTableIdentifier)
tablePath: org.apache.carbondata.core.util.path.CarbonTablePath = hdfs://hacluster/user/hive/warehouse/carbon.store/default/stream_table

scala>

scala> val serverSocket = new ServerSocket(7071)
serverSocket: java.net.ServerSocket = ServerSocket[addr=0.0.0.0/0.0.0.0,localport=7071]

scala> val socketThread = writeSocket(serverSocket)
socketThread: Thread = Thread[Thread-103,5,main]

scala> val streamingThread = startStreaming(carbonSession, tablePath)
streamingThread: Thread = Thread[Thread-104,5,main]
*
*scala> Exception in thread "Thread-104" java.lang.NullPointerException
        at $line29.$read$$iw$$iw$$iw$$iw$$iw$$iw$$iw$$iw$$iw$$iw$$anon$1.run(<console>:59)**




Expected : The startstreaming should not throw exception and should be successful.

  was:
Steps :
// prepare csv file for batch loading
cd /srv/spark2.2Bigdata/install/hadoop/datanode/bin

// generate streamSample.csv

100000001,batch_1,city_1,0.1,school_1:school_11$20
100000002,batch_2,city_2,0.2,school_2:school_22$30
100000003,batch_3,city_3,0.3,school_3:school_33$40
100000004,batch_4,city_4,0.4,school_4:school_44$50
100000005,batch_5,city_5,0.5,school_5:school_55$60

// put to hdfs /tmp/streamSample.csv
./hadoop fs -put streamSample.csv /tmp

// spark-beeline
cd /srv/spark2.2Bigdata/install/spark/sparkJdbc
bin/spark-submit --master yarn-client --executor-memory 10G --executor-cores 5 --driver-memory 5G --num-executors 3 --class org.apache.carbondata.spark.thriftserver.CarbonThriftServer /srv/spark2.2Bigdata/install/spark/sparkJdbc/carbonlib/carbondata_2.11-1.3.0-SNAPSHOT-shade-hadoop2.7.2.jar "hdfs://hacluster/user/sparkhive/warehouse"

bin/beeline -u jdbc:hive2://10.18.98.34:23040

CREATE TABLE stream_table(
id INT,
name STRING,
city STRING,
salary FLOAT
)
STORED BY 'carbondata'
TBLPROPERTIES('streaming'='true', 'sort_columns'='name');

LOAD DATA LOCAL INPATH 'hdfs://hacluster/chetan/streamSample.csv' INTO TABLE stream_table OPTIONS('HEADER'='false');

// spark-shell
cd /srv/spark2.2Bigdata/install/spark/sparkJdbc
bin/spark-shell --master yarn-client

import java.io.{File, PrintWriter}
import java.net.ServerSocket

import org.apache.spark.sql.{CarbonEnv, SparkSession}
import org.apache.spark.sql.hive.CarbonRelation
import org.apache.spark.sql.streaming.{ProcessingTime, StreamingQuery}

import org.apache.carbondata.core.constants.CarbonCommonConstants
import org.apache.carbondata.core.util.CarbonProperties
import org.apache.carbondata.core.util.path.{CarbonStorePath, CarbonTablePath}

CarbonProperties.getInstance().addProperty(CarbonCommonConstants.CARBON_TIMESTAMP_FORMAT, "yyyy/MM/dd")

import org.apache.spark.sql.CarbonSession._

val carbonSession = SparkSession.
  builder().
  appName("StreamExample").
  config("spark.sql.warehouse.dir", "hdfs://hacluster/user/sparkhive/warehouse").
  config("javax.jdo.option.ConnectionURL", "jdbc:mysql://10.18.98.34:3306/sparksql?characterEncoding=UTF-8").
  config("javax.jdo.option.ConnectionDriverName", "com.mysql.jdbc.Driver").
  config("javax.jdo.option.ConnectionPassword", "huawei").
  config("javax.jdo.option.ConnectionUserName", "sparksql").
  getOrCreateCarbonSession()
   
carbonSession.sparkContext.setLogLevel("ERROR")

carbonSession.sql("select * from stream_table").show

*Issue : Select query from spark-shell does not execute successfully for streaming table load.*
When the executor and driver cores and memory is increased while launching the spark shell the issue still occurs.
bin/spark-shell --master yarn-client --executor-memory 10G --executor-cores 5 --driver-memory 5G --num-executors 3
scala> import org.apache.carbondata.core.constants.CarbonCommonConstants
import org.apache.carbondata.core.constants.CarbonCommonConstants

scala> import org.apache.carbondata.core.util.CarbonProperties
import org.apache.carbondata.core.util.CarbonProperties

scala> import org.apache.carbondata.core.util.path.{CarbonStorePath, CarbonTablePath}
import org.apache.carbondata.core.util.path.{CarbonStorePath, CarbonTablePath}

scala>

scala> CarbonProperties.getInstance().addProperty(CarbonCommonConstants.CARBON_TIMESTAMP_FORMAT, "yyyy/MM/dd")
res29: org.apache.carbondata.core.util.CarbonProperties = org.apache.carbondata.core.util.CarbonProperties@67b056e7

scala>

scala> import org.apache.spark.sql.CarbonSession._
import org.apache.spark.sql.CarbonSession._

scala>

scala> val carbonSession = SparkSession.
     |   builder().
     |   appName("StreamExample").
     |   config("spark.sql.warehouse.dir", "hdfs://hacluster/user/sparkhive/warehouse").
     |   config("javax.jdo.option.ConnectionURL", "jdbc:mysql://10.18.98.34:3306/sparksql?characterEncoding=UTF-8").
     |   config("javax.jdo.option.ConnectionDriverName", "com.mysql.jdbc.Driver").
     |   config("javax.jdo.option.ConnectionPassword", "huawei").
     |   config("javax.jdo.option.ConnectionUserName", "sparksql").
     |   getOrCreateCarbonSession()
carbonSession: org.apache.spark.sql.SparkSession = org.apache.spark.sql.CarbonSession@1d0590bc

scala>
     | carbonSession.sparkContext.setLogLevel("ERROR")

scala> carbonSession.sql("select * from stream_table").show
org.apache.spark.SparkException: Job aborted due to stage failure: Task 0 in stage 25.0 failed 4 times, most recent failure: Lost task 0.3 in stage 25.0 (TID 65, BLR1000014269, executor 8): java.lang.IllegalStateException: unread block data
        at java.io.ObjectInputStream$BlockDataInputStream.setBlockDataMode(ObjectInputStream.java:2424)
        at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1383)
        at java.io.ObjectInputStream.defaultReadFields(ObjectInputStream.java:1993)
        at java.io.ObjectInputStream.readSerialData(ObjectInputStream.java:1918)
        at java.io.ObjectInputStream.readOrdinaryObject(ObjectInputStream.java:1801)
        at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1351)
        at java.io.ObjectInputStream.readObject(ObjectInputStream.java:371)
        at org.apache.spark.serializer.JavaDeserializationStream.readObject(JavaSerializer.scala:75)
        at org.apache.spark.serializer.JavaSerializerInstance.deserialize(JavaSerializer.scala:114)
        at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:258)
        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
        at java.lang.Thread.run(Thread.java:745)

Driver stacktrace:
  at org.apache.spark.scheduler.DAGScheduler.org$apache$spark$scheduler$DAGScheduler$$failJobAndIndependentStages(DAGScheduler.scala:1435)
  at org.apache.spark.scheduler.DAGScheduler$$anonfun$abortStage$1.apply(DAGScheduler.scala:1423)
  at org.apache.spark.scheduler.DAGScheduler$$anonfun$abortStage$1.apply(DAGScheduler.scala:1422)
  at scala.collection.mutable.ResizableArray$class.foreach(ResizableArray.scala:59)
  at scala.collection.mutable.ArrayBuffer.foreach(ArrayBuffer.scala:48)
  at org.apache.spark.scheduler.DAGScheduler.abortStage(DAGScheduler.scala:1422)
  at org.apache.spark.scheduler.DAGScheduler$$anonfun$handleTaskSetFailed$1.apply(DAGScheduler.scala:802)
  at org.apache.spark.scheduler.DAGScheduler$$anonfun$handleTaskSetFailed$1.apply(DAGScheduler.scala:802)
  at scala.Option.foreach(Option.scala:257)
  at org.apache.spark.scheduler.DAGScheduler.handleTaskSetFailed(DAGScheduler.scala:802)
  at org.apache.spark.scheduler.DAGSchedulerEventProcessLoop.doOnReceive(DAGScheduler.scala:1650)
  at org.apache.spark.scheduler.DAGSchedulerEventProcessLoop.onReceive(DAGScheduler.scala:1605)
  at org.apache.spark.scheduler.DAGSchedulerEventProcessLoop.onReceive(DAGScheduler.scala:1594)
  at org.apache.spark.util.EventLoop$$anon$1.run(EventLoop.scala:48)
  at org.apache.spark.scheduler.DAGScheduler.runJob(DAGScheduler.scala:628)
  at org.apache.spark.SparkContext.runJob(SparkContext.scala:1918)
  at org.apache.spark.SparkContext.runJob(SparkContext.scala:1931)
  at org.apache.spark.SparkContext.runJob(SparkContext.scala:1944)
  at org.apache.spark.sql.execution.SparkPlan.executeTake(SparkPlan.scala:333)
  at org.apache.spark.sql.execution.CollectLimitExec.executeCollect(limit.scala:38)
  at org.apache.spark.sql.Dataset$$anonfun$org$apache$spark$sql$Dataset$$execute$1$1.apply(Dataset.scala:2371)
  at org.apache.spark.sql.execution.SQLExecution$.withNewExecutionId(SQLExecution.scala:57)
  at org.apache.spark.sql.Dataset.withNewExecutionId(Dataset.scala:2765)
  at org.apache.spark.sql.Dataset.org$apache$spark$sql$Dataset$$execute$1(Dataset.scala:2370)
  at org.apache.spark.sql.Dataset.org$apache$spark$sql$Dataset$$collect(Dataset.scala:2377)
  at org.apache.spark.sql.Dataset$$anonfun$head$1.apply(Dataset.scala:2113)
  at org.apache.spark.sql.Dataset$$anonfun$head$1.apply(Dataset.scala:2112)
  at org.apache.spark.sql.Dataset.withTypedCallback(Dataset.scala:2795)
  at org.apache.spark.sql.Dataset.head(Dataset.scala:2112)
  at org.apache.spark.sql.Dataset.take(Dataset.scala:2327)
  at org.apache.spark.sql.Dataset.showString(Dataset.scala:248)
  at org.apache.spark.sql.Dataset.show(Dataset.scala:636)
  at org.apache.spark.sql.Dataset.show(Dataset.scala:595)
  at org.apache.spark.sql.Dataset.show(Dataset.scala:604)
  ... 50 elided
Caused by: java.lang.IllegalStateException: unread block data
  at java.io.ObjectInputStream$BlockDataInputStream.setBlockDataMode(ObjectInputStream.java:2424)
  at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1383)
  at java.io.ObjectInputStream.defaultReadFields(ObjectInputStream.java:1993)
  at java.io.ObjectInputStream.readSerialData(ObjectInputStream.java:1918)
  at java.io.ObjectInputStream.readOrdinaryObject(ObjectInputStream.java:1801)
  at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1351)
  at java.io.ObjectInputStream.readObject(ObjectInputStream.java:371)
  at org.apache.spark.serializer.JavaDeserializationStream.readObject(JavaSerializer.scala:75)
  at org.apache.spark.serializer.JavaSerializerInstance.deserialize(JavaSerializer.scala:114)
  at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:258)
  at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
  at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
  at java.lang.Thread.run(Thread.java:745)



Expected : Select query from spark-shell should execute successfully for streaming table load.

> Carbon1.3.0-Streaming - Select query from spark-shell does not execute successfully for streaming table load
> ------------------------------------------------------------------------------------------------------------
>
>                 Key: CARBONDATA-1726
>                 URL: https://issues.apache.org/jira/browse/CARBONDATA-1726
>             Project: CarbonData
>          Issue Type: Bug
>          Components: data-query
>    Affects Versions: 1.3.0
>         Environment: 3 node ant cluster SUSE 11 SP4
>            Reporter: Chetan Bhat
>            Priority: Blocker
>              Labels: Functional
>
> Steps :
> // prepare csv file for batch loading
> cd /srv/spark2.2Bigdata/install/hadoop/datanode/bin
> // generate streamSample.csv
> 100000001,batch_1,city_1,0.1,school_1:school_11$20
> 100000002,batch_2,city_2,0.2,school_2:school_22$30
> 100000003,batch_3,city_3,0.3,school_3:school_33$40
> 100000004,batch_4,city_4,0.4,school_4:school_44$50
> 100000005,batch_5,city_5,0.5,school_5:school_55$60
> // put to hdfs /tmp/streamSample.csv
> ./hadoop fs -put streamSample.csv /tmp
> // spark-beeline
> cd /srv/spark2.2Bigdata/install/spark/sparkJdbc
> bin/spark-submit --master yarn-client --executor-memory 10G --executor-cores 5 --driver-memory 5G --num-executors 3 --class org.apache.carbondata.spark.thriftserver.CarbonThriftServer /srv/spark2.2Bigdata/install/spark/sparkJdbc/carbonlib/carbondata_2.11-1.3.0-SNAPSHOT-shade-hadoop2.7.2.jar "hdfs://hacluster/user/sparkhive/warehouse"
> bin/beeline -u jdbc:hive2://10.18.98.34:23040
> CREATE TABLE stream_table(
> id INT,
> name STRING,
> city STRING,
> salary FLOAT
> )
> STORED BY 'carbondata'
> TBLPROPERTIES('streaming'='true', 'sort_columns'='name');
> LOAD DATA LOCAL INPATH 'hdfs://hacluster/chetan/streamSample.csv' INTO TABLE stream_table OPTIONS('HEADER'='false');
> // spark-shell
> cd /srv/spark2.2Bigdata/install/spark/sparkJdbc
> bin/spark-shell --master yarn-client --executor-memory 10G --executor-cores 5 --driver-memory 5G --num-executors 3 --jars /srv/spark2.2Bigdata/install/spark/sparkJdbc/carbonlib/carbondata_2.11-1.3.0-SNAPSHOT-shade-hadoop2.7.2.jar
> import java.io.{File, PrintWriter}
> import java.net.ServerSocket
> import org.apache.spark.sql.{CarbonEnv, SparkSession}
> import org.apache.spark.sql.hive.CarbonRelation
> import org.apache.spark.sql.streaming.{ProcessingTime, StreamingQuery}
> import org.apache.carbondata.core.constants.CarbonCommonConstants
> import org.apache.carbondata.core.util.CarbonProperties
> import org.apache.carbondata.core.util.path.{CarbonStorePath, CarbonTablePath}
> CarbonProperties.getInstance().addProperty(CarbonCommonConstants.CARBON_TIMESTAMP_FORMAT, "yyyy/MM/dd")
> import org.apache.spark.sql.CarbonSession._
> val carbonSession = SparkSession.
>   builder().
>   appName("StreamExample").
>   config("spark.sql.warehouse.dir", "hdfs://hacluster/user/sparkhive/warehouse").
>   config("javax.jdo.option.ConnectionURL", "jdbc:mysql://10.18.98.34:3306/sparksql?characterEncoding=UTF-8").
>   config("javax.jdo.option.ConnectionDriverName", "com.mysql.jdbc.Driver").
>   config("javax.jdo.option.ConnectionPassword", "huawei").
>   config("javax.jdo.option.ConnectionUserName", "sparksql").
>   getOrCreateCarbonSession()
>    
> carbonSession.sparkContext.setLogLevel("ERROR")
> carbonSession.sql("select * from stream_table").show
> def writeSocket(serverSocket: ServerSocket): Thread = {
>   val thread = new Thread() {
>     override def run(): Unit = {
>       // wait for client to connection request and accept
>       val clientSocket = serverSocket.accept()
>       val socketWriter = new PrintWriter(clientSocket.getOutputStream())
>       var index = 0
>       for (_ <- 1 to 1000) {
>         // write 5 records per iteration
>         for (_ <- 0 to 100) {
>           index = index + 1
>           socketWriter.println(index.toString + ",name_" + index
>                                + ",city_" + index + "," + (index * 10000.00).toString +
>                                ",school_" + index + ":school_" + index + index + "$" + index)
>         }
>         socketWriter.flush()
>         Thread.sleep(2000)
>       }
>       socketWriter.close()
>       System.out.println("Socket closed")
>     }
>   }
>   thread.start()
>   thread
> }
>  
> def startStreaming(spark: SparkSession, tablePath: CarbonTablePath): Thread = {
>   val thread = new Thread() {
>     override def run(): Unit = {
>       var qry: StreamingQuery = null
>       try {
>         val readSocketDF = spark.readStream
>           .format("socket")
>           .option("host", "10.18.98.34")
>           .option("port", 7071)
>           .load()
>         // Write data from socket stream to carbondata file
>         qry = readSocketDF.writeStream
>           .format("carbondata")
>           .trigger(ProcessingTime("5 seconds"))
>           .option("checkpointLocation", tablePath.getStreamingCheckpointDir)
>           .option("tablePath", tablePath.getPath)
>           .start()
>         qry.awaitTermination()
>       } catch {
>         case _: InterruptedException =>
>           println("Done reading and writing streaming data")
>       } finally {
>         qry.stop()
>       }
>     }
>   }
>   thread.start()
>   thread
> }
> val streamTableName = s"stream_table"
> val carbonTable = CarbonEnv.getInstance(carbonSession).carbonMetastore.
>   lookupRelation(Some("default"), streamTableName)(carbonSession).asInstanceOf[CarbonRelation].
>   tableMeta.carbonTable
> val tablePath = CarbonStorePath.getCarbonTablePath(carbonTable.getAbsoluteTableIdentifier)
> val serverSocket = new ServerSocket(7071)
> val socketThread = writeSocket(serverSocket)
> val streamingThread = startStreaming(carbonSession, tablePath)
> **Issue : There is a null pointer exception when streaming is started.*
> When the executor and driver cores and memory is increased while launching the spark shell the issue still occurs.
> scala> import java.io.{File, PrintWriter}
> import java.io.{File, PrintWriter}
> scala> import java.net.ServerSocket
> import java.net.ServerSocket
> scala>
> scala> import org.apache.spark.sql.{CarbonEnv, SparkSession}
> import org.apache.spark.sql.{CarbonEnv, SparkSession}
> scala> import org.apache.spark.sql.hive.CarbonRelation
> import org.apache.spark.sql.hive.CarbonRelation
> scala> import org.apache.spark.sql.streaming.{ProcessingTime, StreamingQuery}
> import org.apache.spark.sql.streaming.{ProcessingTime, StreamingQuery}
> scala>
> scala> import org.apache.carbondata.core.constants.CarbonCommonConstants
> import org.apache.carbondata.core.constants.CarbonCommonConstants
> scala> import org.apache.carbondata.core.util.CarbonProperties
> import org.apache.carbondata.core.util.CarbonProperties
> scala> import org.apache.carbondata.core.util.path.{CarbonStorePath, CarbonTablePath}
> import org.apache.carbondata.core.util.path.{CarbonStorePath, CarbonTablePath}
> scala>
> scala> CarbonProperties.getInstance().addProperty(CarbonCommonConstants.CARBON_TIMESTAMP_FORMAT, "yyyy/MM/dd")
> res0: org.apache.carbondata.core.util.CarbonProperties = org.apache.carbondata.core.util.CarbonProperties@7212b28e
> scala>
> scala> import org.apache.spark.sql.CarbonSession._
> import org.apache.spark.sql.CarbonSession._
> scala>
> scala> val carbonSession = SparkSession.
>      |   builder().
>      |   appName("StreamExample").
>      |   config("spark.sql.warehouse.dir", "hdfs://hacluster/user/sparkhive/warehouse").
>      |   config("javax.jdo.option.ConnectionURL", "jdbc:mysql://10.18.98.34:3306/sparksql?characterEncoding=UTF-8").
>      |   config("javax.jdo.option.ConnectionDriverName", "com.mysql.jdbc.Driver").
>      |   config("javax.jdo.option.ConnectionPassword", "huawei").
>      |   config("javax.jdo.option.ConnectionUserName", "sparksql").
>      |   getOrCreateCarbonSession()
> carbonSession: org.apache.spark.sql.SparkSession = org.apache.spark.sql.CarbonSession@7593716d
> scala>
>      | carbonSession.sparkContext.setLogLevel("ERROR")
> scala>
> scala> carbonSession.sql("select * from stream_table").show
> +---------+-------+------+------+
> |       id|   name|  city|salary|
> +---------+-------+------+------+
> |100000001|batch_1|city_1|   0.1|
> |100000002|batch_2|city_2|   0.2|
> |100000003|batch_3|city_3|   0.3|
> |100000004|batch_4|city_4|   0.4|
> |100000005|batch_5|city_5|   0.5|
> +---------+-------+------+------+
> scala> def writeSocket(serverSocket: ServerSocket): Thread = {
>      |   val thread = new Thread() {
>      |     override def run(): Unit = {
>      |       // wait for client to connection request and accept
>      |       val clientSocket = serverSocket.accept()
>      |       val socketWriter = new PrintWriter(clientSocket.getOutputStream())
>      |       var index = 0
>      |       for (_ <- 1 to 1000) {
>      |         // write 5 records per iteration
>      |         for (_ <- 0 to 100) {
>      |           index = index + 1
>      |           socketWriter.println(index.toString + ",name_" + index
>      |                                + ",city_" + index + "," + (index * 10000.00).toString +
>      |                                ",school_" + index + ":school_" + index + index + "$" + index)
>      |         }
>      |         socketWriter.flush()
>      |         Thread.sleep(2000)
>      |       }
>      |       socketWriter.close()
>      |       System.out.println("Socket closed")
>      |     }
>      |   }
>      |   thread.start()
>      |   thread
>      | }
> writeSocket: (serverSocket: java.net.ServerSocket)Thread
> scala>
>      | def startStreaming(spark: SparkSession, tablePath: CarbonTablePath): Thread = {
>      |   val thread = new Thread() {
>      |     override def run(): Unit = {
>      |       var qry: StreamingQuery = null
>      |       try {
>      |         val readSocketDF = spark.readStream
>      |           .format("socket")
>      |           .option("host", "10.18.98.34")
>      |           .option("port", 7071)
>      |           .load()
>      |
>      |         // Write data from socket stream to carbondata file
>      |         qry = readSocketDF.writeStream
>      |           .format("carbondata")
>      |           .trigger(ProcessingTime("5 seconds"))
>      |           .option("checkpointLocation", tablePath.getStreamingCheckpointDir)
>      |           .option("tablePath", tablePath.getPath)
>      |           .start()
>      |
>      |         qry.awaitTermination()
>      |       } catch {
>      |         case _: InterruptedException =>
>      |           println("Done reading and writing streaming data")
>      |       } finally {
>      |         qry.stop()
>      |       }
>      |     }
>      |   }
>      |   thread.start()
>      |   thread
>      | }
> startStreaming: (spark: org.apache.spark.sql.SparkSession, tablePath: org.apache.carbondata.core.util.path.CarbonTablePath)Thread
> scala>
> scala> val streamTableName = s"stream_table"
> streamTableName: String = stream_table
> scala>
> scala> val carbonTable = CarbonEnv.getInstance(carbonSession).carbonMetastore.
>      |   lookupRelation(Some("default"), streamTableName)(carbonSession).asInstanceOf[CarbonRelation].
>      |   tableMeta.carbonTable
> carbonTable: org.apache.carbondata.core.metadata.schema.table.CarbonTable = org.apache.carbondata.core.metadata.schema.table.CarbonTable@62cf8fda
> scala>
> scala> val tablePath = CarbonStorePath.getCarbonTablePath(carbonTable.getAbsoluteTableIdentifier)
> tablePath: org.apache.carbondata.core.util.path.CarbonTablePath = hdfs://hacluster/user/hive/warehouse/carbon.store/default/stream_table
> scala>
> scala> val serverSocket = new ServerSocket(7071)
> serverSocket: java.net.ServerSocket = ServerSocket[addr=0.0.0.0/0.0.0.0,localport=7071]
> scala> val socketThread = writeSocket(serverSocket)
> socketThread: Thread = Thread[Thread-103,5,main]
> scala> val streamingThread = startStreaming(carbonSession, tablePath)
> streamingThread: Thread = Thread[Thread-104,5,main]
> *
> *scala> Exception in thread "Thread-104" java.lang.NullPointerException
>         at $line29.$read$$iw$$iw$$iw$$iw$$iw$$iw$$iw$$iw$$iw$$iw$$anon$1.run(<console>:59)**
> Expected : The startstreaming should not throw exception and should be successful.

--
This message was sent by Atlassian JIRA
(v6.4.14#64029)