Uploaded image for project: 'OpenShift Bugs'
  1. OpenShift Bugs
  2. OCPBUGS-19497

ptp clock does not stay locked in GM configuration when threshold is changed in ptpconfig

XMLWordPrintable

    • No
    • False
    • Hide

      None

      Show
      None
    • 9/19 : dev is working on handling monitoring of ptp config update.PR targeted for this week Friday

      This is a clone of issue OCPBUGS-19345. The following is the description of the original issue:

      Description of problem:

      GM clock is not locked after threshold is changed in ptpconfig
      
      
      

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

      ptp-operator.v4.14.0-202308291102

      How reproducible:

      100%

      Steps to Reproduce:

      1. Deploy SNO with PTP GM configuration
      
      2. Check ptp metrics:
      oc -n openshift-ptp exec $(oc get pods -n openshift-ptp|awk '/linuxptp-daemon.*Running/ {print $1}') -c linuxptp-daemon-container -- curl -s 127.0.0.1:9091/metrics
      [...]
      # TYPE openshift_ptp_clock_state gauge
      openshift_ptp_clock_state{iface="CLOCK_REALTIME",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="phc2sys"} 1
      openshift_ptp_clock_state{iface="ens3f0",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="GM"} 1
      openshift_ptp_clock_state{iface="ens3f0",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="gnss"} 1
      openshift_ptp_clock_state{iface="ens3fx",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="ts2phc"} 1
      openshift_ptp_clock_state{iface="ens5fx",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="ptp4l"} 1
      openshift_ptp_clock_state{iface="ens7fx",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="ptp4l"} 1
      
      3. Run regression tests
      
      4. Check ptp metrics again after ptp events and metrics tests are finished:
      
      [kni@registry.ran-vcl01 ~]$ oc -n openshift-ptp exec $(oc get pods -n openshift-ptp|awk '/linuxptp-daemon.*Running/ {print $1}') -c linuxptp-daemon-container -- curl -s 127.0.0.1:9091/metrics # HELP cne_api_events_published Metric to get number of events published by the rest api [...] # HELP openshift_ptp_clock_class  # TYPE openshift_ptp_clock_class gauge openshift_ptp_clock_class{node="cnfde4.ptp.lab.eng.bos.redhat.com",process="ptp4l"} 248 # HELP openshift_ptp_clock_state 0 = FREERUN, 1 = LOCKED, 2 = HOLDOVER # TYPE openshift_ptp_clock_state gauge openshift_ptp_clock_state{iface="CLOCK_REALTIME",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="phc2sys"} 0 openshift_ptp_clock_state{iface="ens3f0",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="GM"} 1 openshift_ptp_clock_state{iface="ens3f0",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="gnss"} 1 openshift_ptp_clock_state{iface="ens3fx",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="ts2phc"} 0 openshift_ptp_clock_state{iface="ens5fx",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="ptp4l"} 0 openshift_ptp_clock_state{iface="ens7fx",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="ptp4l"} 0
      
      

      Actual results:

      Following deployment of SNO, ptp clocks were locked:
      
      oc -n openshift-ptp exec $(oc get pods -n openshift-ptp|awk '/linuxptp-daemon.*Running/ {print $1}') -c linuxptp-daemon-container -- curl -s 127.0.0.1:9091/metrics
      [...]
      # TYPE openshift_ptp_clock_state gauge openshift_ptp_clock_state{iface="CLOCK_REALTIME",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="phc2sys"} 1 openshift_ptp_clock_state{iface="ens3f0",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="GM"} 1 openshift_ptp_clock_state{iface="ens3f0",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="gnss"} 1 openshift_ptp_clock_state{iface="ens3fx",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="ts2phc"} 1 openshift_ptp_clock_state{iface="ens5fx",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="ptp4l"} 1 openshift_ptp_clock_state{iface="ens7fx",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="ptp4l"} 1
      
      
      Following ptp events and metrics tests:
      
      [kni@registry.ran-vcl01 ~]$ oc -n openshift-ptp exec $(oc get pods -n openshift-ptp|awk '/linuxptp-daemon.*Running/ {print $1}') -c linuxptp-daemon-container -- curl -s 127.0.0.1:9091/metrics # HELP cne_api_events_published Metric to get number of events published by the rest api [...] # HELP openshift_ptp_clock_class  # TYPE openshift_ptp_clock_class gauge openshift_ptp_clock_class{node="cnfde4.ptp.lab.eng.bos.redhat.com",process="ptp4l"} 248 # HELP openshift_ptp_clock_state 0 = FREERUN, 1 = LOCKED, 2 = HOLDOVER # TYPE openshift_ptp_clock_state gauge openshift_ptp_clock_state{iface="CLOCK_REALTIME",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="phc2sys"} 0 openshift_ptp_clock_state{iface="ens3f0",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="GM"} 1 openshift_ptp_clock_state{iface="ens3f0",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="gnss"} 1 openshift_ptp_clock_state{iface="ens3fx",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="ts2phc"} 0 openshift_ptp_clock_state{iface="ens5fx",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="ptp4l"} 0 openshift_ptp_clock_state{iface="ens7fx",node="cnfde4.ptp.lab.eng.bos.redhat.com",process="ptp4l"} 0

      Expected results:

      ptp clock is in sync.

      Additional info:

       

            aputtur@redhat.com Aneesh Puttur
            openshift-crt-jira-prow OpenShift Prow Bot
            Bonnie Block Bonnie Block
            Votes:
            0 Vote for this issue
            Watchers:
            5 Start watching this issue

              Created:
              Updated:
              Resolved: