Uploaded image for project: 'Hadoop Common'
  1. Hadoop Common
  2. HADOOP-5551

Namenode permits directory destruction on overwrite

Log workAgile BoardRank to TopRank to BottomAttach filesAttach ScreenshotVotersWatch issueWatchersCreate sub-taskConvert to sub-taskMoveLinkCloneLabelsUpdate Comment AuthorReplace String in CommentUpdate Comment VisibilityDelete Comments
    XMLWordPrintableJSON

    Details

    • Type: Bug
    • Status: Closed
    • Priority: Critical
    • Resolution: Fixed
    • Affects Version/s: 0.19.1
    • Fix Version/s: 0.19.2, 0.20.0
    • Component/s: None
    • Labels:
      None
    • Hadoop Flags:
      Reviewed

      Description

      The FSNamesystem's startFileInternal allows overwriting of directories. That is, if you have a directory named /foo/bar and you try to write a file named /foo/bar, the file is written and the directory disappears.

      This is most apparent for folks using libhdfs directly, as overwriting is always turned on. Therefore, if libhdfs applications do not check the existence of a directory first, then they will permit new files to destroy directories.

        Attachments

        1. HADOOP-5551-v2.patch
          2 kB
          Brian Bockelman
        2. HADOOP-5551-v3.patch
          2 kB
          Brian Bockelman
        3. HADOOP-5551-v4.patch
          2 kB
          Brian Bockelman

          Activity

          $i18n.getText('security.level.explanation', $currentSelection) Viewable by All Users
          Cancel

            People

            • Assignee:
              bockelman Brian Bockelman Assign to me
              Reporter:
              bockelman Brian Bockelman

              Dates

              • Created:
                Updated:
                Resolved:

                Issue deployment