Uploaded image for project: 'Hadoop HDFS'
  1. Hadoop HDFS
  2. HDFS-11738

Hedged pread takes more time when block moved from initial locations

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • None
    • 2.9.0, 3.0.0-beta1, 2.8.3
    • hdfs-client
    • None

    Description

      Scenario :
      Same as HDFS-11708.

      During Hedge read,
      1. First two locations fails to read the data in hedged mode.
      2. chooseData refetches locations and adds a future to read from DN3.
      3. after adding future to DN3, main thread goes for refetching locations in loop and stucks there till all 3 retries to fetch locations exhausted, which consumes ~20 seconds with exponential retry time.

      Attachments

        1. HDFS-11738-01.patch
          28 kB
          Vinayakumar B
        2. HDFS-11738-02.patch
          14 kB
          Vinayakumar B
        3. HDFS-11738-03.patch
          13 kB
          Vinayakumar B
        4. HDFS-11738-04.patch
          14 kB
          Vinayakumar B
        5. HDFS-11738-05.patch
          12 kB
          Vinayakumar B
        6. HDFS-11738-branch-2.8-committed.patch
          12 kB
          Vinayakumar B
        7. HDFS-11738-branch-2-committed.patch
          13 kB
          Vinayakumar B

        Issue Links

          depends upon

          Bug - A problem which impairs or prevents the functions of the product. HDFS-11708 Positional read will fail if replicas moved to different DNs after stream is opened

          • Critical - Crashes, loss of data, severe memory leak.
          • Resolved

          Activity

            People

              vinayakumarb Vinayakumar B
              vinayakumarb Vinayakumar B
              Votes:
              0 Vote for this issue
              Watchers:
              8 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: