Uploaded image for project: 'Hadoop Map/Reduce'
  1. Hadoop Map/Reduce
  2. MAPREDUCE-3713

Incorrect headroom reported to jobs

VotersWatch issueWatchersCreate sub-taskLinkCloneUpdate Comment AuthorReplace String in CommentUpdate Comment VisibilityDelete Comments
    XMLWordPrintableJSON

Details

    • Bug
    • Status: Closed
    • Blocker
    • Resolution: Fixed
    • 0.23.0
    • 0.23.1
    • mrv2, resourcemanager
    • None
    • Reviewed
    • Fixed the way head-room is allocated to applications by CapacityScheduler so that it deducts current-usage per user and not per-application.

    Description

      With multiple jobs submitted per user, and multiple users submitting jobs - the headroom reported to the AppMasters is incorrect (very high).
      Leads to a deadlock - reduces started, map tasks not complete... and reduces are not preempted by the AM due to the incorrect headroom.

      Attachments

        1. MAPREDUCE-3713.patch
          5 kB
          Arun Murthy
        2. MAPREDUCE-3713.patch
          14 kB
          Arun Murthy
        3. MAPREDUCE-3713.patch
          14 kB
          Arun Murthy

        Issue Links

        Activity

          This comment will be Viewable by All Users Viewable by All Users
          Cancel

          People

            acmurthy Arun Murthy
            sseth Siddharth Seth
            Votes:
            0 Vote for this issue
            Watchers:
            1 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved:

              Slack

                Issue deployment