Uploaded image for project: 'Hadoop Map/Reduce'
  1. Hadoop Map/Reduce
  2. MAPREDUCE-3990

MRBench allows Long-sized input-lines value but parses CLI argument as an Integer

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Resolved
    • Trivial
    • Resolution: Fixed
    • 2.0.0-alpha
    • 3.0.0-alpha1
    • benchmarks

    Description

      MRBench has the following method:

      public void generateTextFile(FileSystem fs, Path inputFile, long numLines, Order sortOrder) { ... }
      

      The method is already set to accept a long datatype for numLines, for generating very large amount of data.

      However, in MRBench#run(...), the inputLines CLI parameter is parsed via an Integer.parseInt, causing numbers passed > Integer.MAX_VALUE to throw NumberFormatExceptions as a result.

      The parsing should be Long.parseLong and the inputLines datatype should be switched to the same type as passed to the method (long).

      Attachments

        1. MAPREDUCE-3990.patch
          1 kB
          Harsh J

        Activity

          People

            qwertymaniac Harsh J
            qwertymaniac Harsh J
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: