Uploaded image for project: 'Hadoop HDFS'
  1. Hadoop HDFS
  2. HDFS-16040

RpcQueueTime metric counts requeued calls as unique events.

    XMLWordPrintableJSON

Details

    • Bug
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • 2.10.0, 3.3.0
    • 2.10.2, 3.2.3, 3.3.2
    • hdfs
    • None
    • Reviewed

    Description

      The RpcQueueTime metric is updated every time a call is re-queued while waiting for the server state to reach the call's client's state ID. This is in contrast to RpcProcessingTime which is only updated when the call if finally processed.

      On the Observer NameNode this can result in RpcQueueTimeNumOps being much larger than RpcProcessingTimeNumOps. The re-queueing is an internal optimization to avoid blocking and shouldn't result in an inflated metric.

      Attachments

        1. HDFS-16040.001.patch
          4 kB
          Simbarashe Dzinamarira
        2. HDFS-16040.002.patch
          4 kB
          Simbarashe Dzinamarira
        3. HDFS-16040.003.patch
          4 kB
          Simbarashe Dzinamarira

        Activity

          People

            simbadzina Simbarashe Dzinamarira
            simbadzina Simbarashe Dzinamarira
            Votes:
            0 Vote for this issue
            Watchers:
            3 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: