Uploaded image for project: 'Hadoop Common'
  1. Hadoop Common
  2. HADOOP-5883

TaskMemoryMonitorThread might shoot down tasks even if their processes momentarily exceed the requested memory

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • None
    • 0.20.1
    • None
    • None
    • Reviewed

    Description

      Currently the TaskMemoryMonitorThread kills tasks as soon as it detects they are consuming more memory than the max value specified. There are valid cases (see HADOOP-5059) where if a program is executed from the task, it might momentarily occupy twice the amount of memory for a short time. Ideally the monitoring thread should handle this case.

      Attachments

        1. HADOOP-5883.patch
          29 kB
          Hemanth Yamijala
        2. HADOOP-5883.patch
          29 kB
          Hemanth Yamijala
        3. HADOOP-5883.patch
          28 kB
          Hemanth Yamijala
        4. HADOOP-5883-20.patch
          29 kB
          Hemanth Yamijala
        5. HADOOP-5883-20.patch
          29 kB
          Hemanth Yamijala

        Issue Links

          Activity

            People

              yhemanth Hemanth Yamijala
              yhemanth Hemanth Yamijala
              Votes:
              0 Vote for this issue
              Watchers:
              7 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: