Uploaded image for project: 'OpenShift Bugs'
  1. OpenShift Bugs
  2. OCPBUGS-46372

Static pod node controller can overwrite nodestatuses with stale information

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Unresolved
    • Icon: Critical Critical
    • None
    • 4.18.0, 4.19.0
    • kube-apiserver
    • None
    • Rejected
    • False
    • Hide

      None

      Show
      None

      Description of problem:

      The node controller reads the current nodestatuses from a lister, then makes builds an apply configuration that includes all fields of each nodestatus entry based on the value observed from the lister. If another controller (especially the installer controller) has written to a nodestatus, but that write was not reflected in the lister before the node controller observes it, the node controller's apply request will effectively revert to the last status its lister observed.

Before 4.18, these control loops were update-based and would receive 409 responses instead of successfully stomping unobserved writes.

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

      How reproducible:

      Steps to Reproduce:

          1.
    2.
    3.

      Actual results:

      Expected results:

      Additional info:

              bluddy Ben Luddy
              bluddy Ben Luddy
              Ke Wang Ke Wang
              Votes:
              0 Vote for this issue
              Watchers:
              6 Start watching this issue

                Created:
                Updated: