Uploaded image for project: 'OpenShift Bugs'
  1. OpenShift Bugs
  2. OCPBUGS-13075

ice-gnss process used by PTP GM config is taking more than half CPU, sometimes almost a full CPU

XMLWordPrintable

    • Important
    • No
    • CNF RAN Sprint 235, CNF RAN Sprint 236, CNF RAN Sprint 237, CNF RAN Sprint 238, CNF RAN Sprint 239, CNF RAN Sprint 240
    • 6
    • False
    • Hide

      None

      Show
      None
    • Hide
      7/31: needs QE to verify in 4.13
      7.24: test with latest rhel build using v10 patch
      7/19: need to verify this for RHEL - this should be in 9.2.
      Rel Note for Telco: Not Required (4.13) - this requires an upstream fix and is not a GA feature. Lab use only.
      Show
      7/31: needs QE to verify in 4.13 7.24: test with latest rhel build using v10 patch 7/19: need to verify this for RHEL - this should be in 9.2. Rel Note for Telco: Not Required (4.13) - this requires an upstream fix and is not a GA feature. Lab use only.

      Description of problem:

      The ice-gnss process used by PTP GM config is taking almost a full CPU.

   2195 root     -11   0       0      0      0 R  88.4   0.0   1177:12 ice-gnss-0000:c                                                                                                                                                                                                    

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

      4.13.0-rc.6 (and rc.5)
ptp-operator.v4.13.0-202304190216

      How reproducible:

      Always - observed on two different servers with GM config

      Steps to Reproduce:

      1. Install SNO DU node and configure grandmaster via PTP operator using WPC card
2. Let system run in idle
3.

      Actual results:

      The ice-gnss process used by PTP GM config is taking almost a full CPU. And cluster can become unreachable sometimes.

   2195 root     -11   0       0      0      0 R  88.4   0.0   1177:12 ice-gnss-0000:c  

      Expected results:

      It should use much less cpu than that.

      Additional info:

      Some info extracted by Brent from one of the affected servers (cnfde4): 

27300.874763] ice 0000:3b:00.0: Direct firmware load for 20230503165920_E810_XXVDA4_FH_O_SEC_FW_1p7p2p4_NVM_4p20_PLDMoMCTP_0.26_80017789.bin failed with error -2
[27341.363952] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[27341.364045] IPv6: ADDRCONF(NETDEV_CHANGE): 55ca212dcd54b05: link becomes ready
[27341.425438] device 55ca212dcd54b05 entered promiscuous mode
[27351.565197] device 55ca212dcd54b05 left promiscuous mode
[27571.841323] ice 0000:af:00.0: Direct firmware load for 20230503170351_E810_XXVDA4_FH_O_SEC_FW_1p7p2p4_NVM_4p20_PLDMoMCTP_0.26_80017789.bin failed with error -2
[27643.130452] ice 0000:af:00.0: Direct firmware load for 20230503170502_E810_XXVDA4_FH_O_SEC_FW_1p7p2p4_NVM_4p20_PLDMoMCTP_0.26_80017789.bin failed with error -2
[28054.673451] ice 0000:3b:00.0: Direct firmware load for 20230503171154_E810_XXVDA4_FH_O_SEC_FW_1p7p2p4_NVM_4p20_PLDMoMCTP_0.26_80017789.bin failed with error -2

              aputtur@redhat.com Aneesh Puttur
              rhn-support-yliu1 Yang Liu
              Hen Shay Hassid Hen Shay Hassid
              Votes:
              0 Vote for this issue
              Watchers:
              10 Start watching this issue

                Created:
                Updated:
                Resolved: