Uploaded image for project: 'Hadoop YARN'
  1. Hadoop YARN
  2. YARN-4308

ContainersAggregated CPU resource utilization reports negative usage in first few heartbeats

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • 2.7.1
    • 2.9.0, 3.0.0-alpha1
    • nodemanager
    • None
    • Reviewed

    Description

      NodeManager reports ContainerAggregated CPU resource utilization as -ve value in first few heartbeats cycles. I added a new debug print and received below values from heartbeats.

      INFO org.apache.hadoop.yarn.server.nodemanager.containermanager.monitor.ContainersMonitorImpl: ContainersResource Utilization : CpuTrackerUsagePercent : -1.0 
      INFO org.apache.hadoop.yarn.server.nodemanager.containermanager.monitor.ContainersMonitorImpl:ContainersResource Utilization :  CpuTrackerUsagePercent : 198.94598
      

      Its better we send 0 as CPU usage rather than sending a negative values in heartbeats eventhough its happening in only first few heartbeats.

      Attachments

        1. 0010-YARN-4308.patch
          15 kB
          Sunil G
        2. 0009-YARN-4308.patch
          14 kB
          Sunil G
        3. 0008-YARN-4308.patch
          14 kB
          Sunil G
        4. 0007-YARN-4308.patch
          13 kB
          Sunil G
        5. 0006-YARN-4308.patch
          13 kB
          Sunil G
        6. 0005-YARN-4308.patch
          10 kB
          Sunil G
        7. 0004-YARN-4308.patch
          6 kB
          Sunil G
        8. 0003-YARN-4308.patch
          6 kB
          Sunil G
        9. 0002-YARN-4308.patch
          3 kB
          Sunil G
        10. 0001-YARN-4308.patch
          2 kB
          Sunil G

        Issue Links

          Activity

            People

              sunilg Sunil G
              sunilg Sunil G
              Votes:
              0 Vote for this issue
              Watchers:
              8 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: