NNCE Enters "Failing" State After Node Reboot due to API Server Probe Timeout

    4.16.54

When a node with an active NodeNetworkConfigurationPolicy (NNCP) is rebooted, the nmstate-handler attempts to reconcile the state immediately upon startup. The reconciliation fails with a "failed running probe 'api-server'" error, causing the NodeNetworkConfigurationEnactment (NNCE) to enter a Failing state.Restarting the nmstate-handler daemonset (oc rollout restart ds/nmstate-handler) resolves the issue immediately.

1. Apply a valid NodeNetworkConfigurationPolicy for static routes)
   apiVersion: nmstate.io/v1

kind: NodeNetworkConfigurationPolicy
metadata:
  labels:
    app.kubernetes.io/instance: openshift-nmstate-config
  name: static-route-hnas-workers
spec:
  desiredState:
    routes:
      config:
      - destination: x.x.x.x/32
        next-hop-address: x.x.x.x
        next-hop-interface: br-ex
        table-id: 254
  nodeSelector:
    node-role.kubernetes.io/worker: ""


2. Verify the NNCE status is Available and SuccessfullyConfigured.

oc get nnce
Thu Jan 15 04:42:47 PM CET 2026
NAME                                                         STATUS      STATUS AGE   REASON
1-workers   Available   2d5h         SuccessfullyConfigured
2-workers   Available   2d5h         SuccessfullyConfigured
3-workers   Available   2d5h         SuccessfullyConfigured
4-workers   Available   2d5h         SuccessfullyConfigured

3. Reboot the worker node (systemctl reboot).

4. Wait for the node to return to Ready state. 

5. Check the NNCE status for that node.

NAME                                                         STATUS      STATUS AGE   REASON
1-workers   Available   2d5h         SuccessfullyConfigured
2-workers   Available   2d5h         SuccessfullyConfigured
3-workers   Available   2d5h         SuccessfullyConfigured
4-workers   Failing     55s          FailedToConfigure

   message: |-
  error reconciling NodeNetworkConfigurationPolicy on node <node-name> desired state apply: "",
  rolling back desired state configuration: failed runnig probes after network changes: failed runnig probe 'api-server' with after network reconfiguration -> currentState: hostname:
  running: node-name

Resolve:

Need to restart the handler pods. 

 The NNCE goes into Failing state after reboot. 

   The NNCE after node reboot automatically reconcile to available state.