[SPARK-25292] Dataframe write to csv saves part files in outputDireotry/task-xx/part-xxx instead of directly saving in outputDir

• Type: Bug
• Status: Resolved
• Priority: Critical
• Resolution: Duplicate
• Affects Version/s: 2.0.2
• Component/s: EC2, Java API, Spark Shell, Spark Submit
• Assignee: Unassigned
• Reporter: omkar puttagunta
• Created: 31/Aug/18 04:44
• Updated: 31/Aug/18 05:50
• Resolved: 31/Aug/18 05:50

Description

https://stackoverflow.com/questions/52108335/why-spark-dataframe-writes-part-files-to-temporary-in-instead-directly-creating

Running Spark 2.0.2 in Standalone Cluster Mode; 2 workers and 1 master node

Simple Test; reading pipe delimited file and writing data to csv. Commands below are executed in spark-shell with master-url set

val df = spark.sqlContext.read.option("delimiter","|").option("quote","\u0000").csv("/home/input-files/") val emailDf=df.filter("_c3='EML'") emailDf.repartition(100).write.csv("/opt/outputFile/")

After executing the cmds above in spark-shell with master url set.

In worker1 -> Each part file is created in{{/opt/outputFile/_temporary/task-xxxxx-xxx/part-xxx-xxx}}
In worker2 -> /opt/outputFile/part-xxx => part files are generated directly under outputDirectory specified during write.

Same thing happens with coalesce(100) or without specifying repartition/coalesce!!!

Quesiton

1) why worker1 /opt/outputFile/ output directory doesn't have part-xxxx files just like in worker2? why _temporary directory is created and part-xxx-xx files reside in the {{task-xxx}}directories?

Issue Links

• duplicates: SPARK-25293 Dataframe write to csv saves part files in outputDireotry/task-xx/part-xxx instead of directly saving in outputDir