[CARBONDATA-2877] CarbonDataWriterException when loading data to carbon table with large number of rows/columns from Spark-Submit

• Type: Bug
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Affects Version/s: 1.4.1
• Component/s: data-load
• Environment: Spark 2.1
• Assignee: Naman Rastogi
• Reporter: Chetan Bhat
• Due: 31/Aug/18
• Created: 23/Aug/18 07:02
• Updated: 17/Sep/18 13:29

Description

Steps :

from Spark-Submit. User creates a table with large number of columns(around 100) and tries to load around 3 lakh records to the table.

Spark-submit command - spark-submit --master yarn --num-executors 3 --executor-memory 75g --driver-memory 10g --executor-cores 12 --class

Actual Issue : Data loading fails with CarbonDataWriterException.

Executor yarn UI log-

org.apache.spark.util.TaskCompletionListenerException: org.apache.carbondata.core.datastore.exception.CarbonDataWriterException

Previous exception in task: Error while initializing data handler :
org.apache.carbondata.processing.loading.steps.DataWriterProcessorStepImpl.execute(DataWriterProcessorStepImpl.java:141)
org.apache.carbondata.processing.loading.DataLoadExecutor.execute(DataLoadExecutor.java:51)
org.apache.carbondata.spark.rdd.NewCarbonDataLoadRDD$$anon$1.<init>(NewCarbonDataLoadRDD.scala:221)
org.apache.carbondata.spark.rdd.NewCarbonDataLoadRDD.internalCompute(NewCarbonDataLoadRDD.scala:197)
org.apache.carbondata.spark.rdd.CarbonRDD.compute(CarbonRDD.scala:78)
org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:324)
org.apache.spark.rdd.RDD.iterator(RDD.scala:288)
org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:87)
org.apache.spark.scheduler.Task.run(Task.scala:99)
org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:325)
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
java.lang.Thread.run(Thread.java:748)
at org.apache.spark.TaskContextImpl.invokeListeners(TaskContextImpl.scala:138)
at org.apache.spark.TaskContextImpl.markTaskCompleted(TaskContextImpl.scala:116)
at org.apache.spark.scheduler.Task.run(Task.scala:109)
at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:325)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)

 

Expected : The dataloading should be successful from Spark-submit similar to that in Beeline.