Uploaded image for project: 'Spark'
  1. Spark
  2. SPARK-38960

Spark should fail fast if initial memory too large(set by "spark.executor.extraJavaOptions") for executor to start

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Resolved
    • Minor
    • Resolution: Won't Fix
    • 3.4.0
    • None
    • Spark Core, Spark Submit, YARN
    • None

    Description

      if you set initial memory(set by "spark.executor.extraJavaOptions=-Xms{XXX}G" ) larger than maximum memory(set by "spark.executor.memory")

      Eg.

           spark.executor.memory=1G

           spark.executor.extraJavaOptions=-Xms2G

       

      from the driver process you just see executor failures with no warning, since the more meaningful errors are buried in the executor logs. 

      Eg., on Yarn, you see:

      Error occurred during initialization of VM
      Initial heap size set to a larger value than the maximum heap size

      Instead we should just fail fast with a clear error message in the driver logs.

      Attachments

        Activity

          People

            Unassigned Unassigned
            panbingkun BingKun Pan
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: