Uploaded image for project: 'OpenShift Logging'
  1. OpenShift Logging
  2. LOG-5564

increased logging-collector pod CPU utilization with steadyworkload

XMLWordPrintable

    • False
    • None
    • False
    • NEW
    • NEW
    • Important
    • Proposed

      Description of problem:

      On SNO spoke cluster with Telco RAN DU profile applied cpu_util testing
      reported increased average logging collector pod cpu utilization during the steadyworload
      phase of the test compared to 4.16.0-rc.1
      
      from 38mcpus to 47mcpus (~24% increase)
          

      Version-Release number of selected component (if applicable):

          OCP: 4.16.0-rc.2
      cluster-logging-operator-bundle/images/v5.9.2-17
      
          

      How reproducible:

      Always
          

      Steps to Reproduce:

          1. Deploy spoke with Telco RAN DU profile
          2. Run cpu_util test suite for 1h
          3. Observe per pod avg cpu util metrics
          

      Actual results:

      compared to 4.16.0-rc.1
      
      from 38mcpus to 47mcpus
          

      Expected results:

      logging-collector pod cpu utilization should be the same or lower than previous build
      
          

      Additional info:

      must-gather and other info will be added in a comment once obtained
          

              Unassigned Unassigned
              rhn-support-dgonyier Dwaine Gonyier
              Anping Li Anping Li
              Votes:
              0 Vote for this issue
              Watchers:
              5 Start watching this issue

                Created:
                Updated:
                Resolved: