Uploaded image for project: 'OpenShift Bugs'
  1. OpenShift Bugs
  2. OCPBUGS-33349

Customers on 4.12 need the "one-shot time sync on boot" fix from 4.14

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Unresolved
    • Icon: Major Major
    • 4.13.z
    • 4.13, 4.12
    • GitOps ZTP
    • None
    • Important
    • No
    • False
    • Hide

      None

      Show
      None
    • Hide
      Following a node reboot, sometimes the server RTC time is not correct.
      This can occur when the server has a depleted RTC battery.
      To correct this issue, the {ztp} {sno} DU profile includes a sync-time-once `MachineConfig` CR that is applied during cluster installation.
      The CR configures a oneshot job that runs when the server boots and uses the the `chronyd` service to set the server RTC time using NTP.
      Show
      Following a node reboot, sometimes the server RTC time is not correct. This can occur when the server has a depleted RTC battery. To correct this issue, the {ztp} {sno} DU profile includes a sync-time-once `MachineConfig` CR that is applied during cluster installation. The CR configures a oneshot job that runs when the server boots and uses the the `chronyd` service to set the server RTC time using NTP.
    • Bug Fix
    • Done

      Description of problem:

      https://access.redhat.com/support/cases/#/case/03795020

      RT clock on a customer's system provided incorrect time on boot; The fix in https://github.com/openshift-kni/cnf-features-deploy/pull/1438/commits (which was introduced in 4.14) would have mitigated this issue properly.

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

      4.12

      How reproducible:

      Infrequent

      Steps to Reproduce:

          1. Multiple reboots occur
    2. Kernel reads a wrong date from the RTC and sets the system time to 30 years in the future
    3.

      Actual results:

      System gets the wrong time and fails to come up fully due to certificate errors

      Expected results:

      The time is set properly and the system recovers fully.

      Additional info:

              jramsay1@redhat.com Jim Ramsay
              jramsay1@redhat.com Jim Ramsay
              Yang Liu Yang Liu
              Aidan Reilly Aidan Reilly
              Votes:
              0 Vote for this issue
              Watchers:
              6 Start watching this issue

                Created:
                Updated: