Uploaded image for project: 'Spark'
  1. Spark
  2. SPARK-33277

Python/Pandas UDF right after off-heap vectorized reader could cause executor crash.

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • 2.4.7, 3.0.1
    • 2.4.8, 3.0.2, 3.1.0, 3.2.1
    • PySpark, SQL
    • None

    Description

      Python/Pandas UDF right after off-heap vectorized reader could cause executor crash.

      E.g.,:

      spark.range(0, 100000, 1, 1).write.parquet(path)
      
      spark.conf.set("spark.sql.columnVector.offheap.enabled", True)
      
      def f(x):
          return 0
      
      fUdf = udf(f, LongType())
      
      spark.read.parquet(path).select(fUdf('id')).head()
      

      This is because, the Python evaluation consumes the parent iterator in a separate thread and it consumes more data from the parent even after the task ends and the parent is closed. If an off-heap column vector exists in the parent iterator, it could cause segmentation fault which crashes the executor.

      Attachments

        Activity

          People

            ueshin Takuya Ueshin
            ueshin Takuya Ueshin
            Votes:
            0 Vote for this issue
            Watchers:
            6 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: