Uploaded image for project: 'Spark'
  1. Spark
  2. SPARK-26708

Incorrect result caused by inconsistency between a SQL cache's cached RDD and its physical plan

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Resolved
    • Blocker
    • Resolution: Fixed
    • 2.4.0
    • 2.4.1, 3.0.0
    • SQL

    Description

      When performing non-cascading cache invalidation, recache is called on the other cache entries which are dependent on the cache being invalidated. It leads to the the physical plans of those cache entries being re-compiled. For those cache entries, if the cache RDD has already been persisted, chances are there will be inconsistency between the data and the new plan. It can cause a correctness issue if the new plan's outputPartitioning or outputOrdering is different from the that of the actual data, and meanwhile the cache is used by another query that asks for specific outputPartitioning or outputOrdering which happens to match the new plan but not the actual data.

      Attachments

        Activity

          People

            maryannxue Wei Xue
            smilegator Xiao Li
            Votes:
            0 Vote for this issue
            Watchers:
            5 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: