Uploaded image for project: 'Hadoop Map/Reduce'
  1. Hadoop Map/Reduce
  2. MAPREDUCE-5351

JobTracker memory leak caused by CleanupQueue reopening FileSystem

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

    Details

    • Type: Bug
    • Status: Closed
    • Priority: Critical
    • Resolution: Fixed
    • Affects Version/s: 1.1.2
    • Fix Version/s: 1-win, 1.2.1
    • Component/s: jobtracker
    • Labels:
      None

      Description

      When a job is completed, closeAllForUGI is called to close all the cached FileSystems in the FileSystem cache. However, the CleanupQueue may run after this occurs and call FileSystem.get() to delete the staging directory, adding a FileSystem to the cache that will never be closed.

      People on the user-list have reported this causing their JobTrackers to OOME every two weeks.

        Attachments

        1. MAPREDUCE-5351-addendum-1.patch
          2 kB
          Sandy Ryza
        2. MAPREDUCE-5351-addendum.patch
          0.6 kB
          Sandy Ryza
        3. MAPREDUCE-5351-2.patch
          6 kB
          Sandy Ryza
        4. MAPREDUCE-5351-1.patch
          6 kB
          Sandy Ryza
        5. MAPREDUCE-5351.patch
          3 kB
          Sandy Ryza
        6. JobInProgress_JobHistory.patch
          3 kB
          jandyfish

        Issue Links

          Activity

            People

            • Assignee:
              sandyr Sandy Ryza
              Reporter:
              sandyr Sandy Ryza

              Dates

              • Created:
                Updated:
                Resolved:

                Issue deployment