Uploaded image for project: 'OpenShift Bugs'
  1. OpenShift Bugs
  2. OCPBUGS-62262

deleted nodes still present in egressfirewalls managed fields

    • Quality / Stability / Reliability
    • False
    • Hide

      None

      Show
      None
    • None
    • None
    • None
    • None
    • CORENET Sprint 278, CORENET Sprint 279, CORENET Sprint 280
    • 3
    • Done
    • Bug Fix
    • Hide
      Before this update, the `AdminNetworkPolicy`, `AdminPolicyBasedRouteListers`, `EgressFirewall`, `EgressQoS` and `NetworkQoS` objects kept the `managedFields` status entries for nodes that had been deleted. As a consequence, a buildup of stale data occurred in etcd for large clusters with frequent node churn. With this release, the cleanup logic is fixed for all of these resource types. As a result, stale data buildup does not occur. (link:https://issues.redhat.com/browse/OCPBUGS-62262[OCPBUGS-62262])
      Show
      Before this update, the `AdminNetworkPolicy`, `AdminPolicyBasedRouteListers`, `EgressFirewall`, `EgressQoS` and `NetworkQoS` objects kept the `managedFields` status entries for nodes that had been deleted. As a consequence, a buildup of stale data occurred in etcd for large clusters with frequent node churn. With this release, the cleanup logic is fixed for all of these resource types. As a result, stale data buildup does not occur. (link: https://issues.redhat.com/browse/OCPBUGS-62262 [ OCPBUGS-62262 ])
    • None
    • None
    • None
    • None

      Description of problem:

      EgressFirewalls still retain information about each deleted machine. On large clusters this build up overtime and end up filling etcd / breaking the api server etc.

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

      reproducing with 4.19.10

      How reproducible:

      trivial:

      Steps to Reproduce:

      1. create an EgressFirewall

      2. delete some machines

      3. oc get egressfirewall default -o yaml --show-managed-fields 

      Actual results:

      managedFields still contains fields

      - apiVersion: k8s.ovn.org/v1
        fieldsType: FieldsV1
        fieldsV1:
          f:status: {}
        manager: worker-7j7fv
        operation: Apply
        subresource: status
        time: "2025-09-25T21:57:58Z" 

      Expected results:

      no leaked managed field

      Additional info:

       

      Please fill in the following template while reporting a bug and provide as much relevant information as possible. Doing so will give us the best chance to find a prompt resolution.

      Affected Platforms:

      OpenShift 4.16, 4.19

      customer issue

       

              rravaiol@redhat.com Riccardo Ravaioli
              frigault Francois Rigault
              None
              None
              Huiran Wang Huiran Wang
              None
              Votes:
              2 Vote for this issue
              Watchers:
              14 Start watching this issue

                Created:
                Updated:
                Resolved: