Uploaded image for project: 'Lucene - Core'
  1. Lucene - Core
  2. LUCENE-1566

Large Lucene index can hit false OOM due to Sun JRE issue

Details

    • Bug
    • Status: Closed
    • Minor
    • Resolution: Fixed
    • 2.4.1
    • 2.9
    • core/index
    • None
    • New

    Description

      This is not a Lucene issue, but I want to open this so future google
      diggers can more easily find it.

      There's this nasty bug in Sun's JRE:

      http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6478546

      The gist seems to be, if you try to read a large (eg 200 MB) number of
      bytes during a single RandomAccessFile.read call, you can incorrectly
      hit OOM. Lucene does this, with norms, since we read in one byte per
      doc per field with norms, as a contiguous array of length maxDoc().

      The workaround was a custom patch to do large file reads as several
      smaller reads.

      Background here:

      http://www.nabble.com/problems-with-large-Lucene-index-td22347854.html

      Attachments

        1. LUCENE_1566_IndexInput_Changes.patch
          21 kB
          Simon Willnauer
        2. LUCENE_1566_IndexInput_Changes.patch
          14 kB
          Simon Willnauer
        3. LUCENE_1566_IndexInput.patch
          17 kB
          Simon Willnauer
        4. LUCENE-1566.patch
          18 kB
          Michael McCandless
        5. LUCENE-1566.patch
          4 kB
          Simon Willnauer
        6. LUCENE-1566.patch
          4 kB
          Simon Willnauer

        Activity

          People

            simonw Simon Willnauer
            mikemccand Michael McCandless
            Votes:
            3 Vote for this issue
            Watchers:
            4 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: