Uploaded image for project: 'Hadoop HDFS'
  1. Hadoop HDFS
  2. HDFS-23

many filelist calls to large directory killing the namenode

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Resolved
    • Major
    • Resolution: Duplicate
    • None
    • None
    • None
    • None

    Description

      We recently had couple of clusters go down due to users' tasks calling dfs -ls frequently on a large directory.
      Namenode started taking time in GC and became unresponsive.

      I'll update once I get more info.

      Attachments

        Issue Links

          Activity

            People

              Unassigned Unassigned
              knoguchi Koji Noguchi
              Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: