Unable to change to a static IP address on a single-node k3s cluster.

[liushan9981]

What happened:
Created an NNCP resource to configure a static IP address, routing, and DNS. After a while, the NNCP status shows as Degraded.
Error message after running "kubectl describe nncp ip1":

Warning  ReconcileFailed  10m   <myhostname>.nmstate-handler  error reconciling NodeNetworkConfigurationPolicy on node <myhostname> at 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:

What you expected to happen:
Changing to a static IP should be successful, including the k3s services, and all pods should function normally.

How to reproduce it (as minimally and precisely as possible):

• Install k3s on a single machine as a single-node k3s cluster
• Using the manifest to create a nncp
• Watch the status of the nncp we created above.

nncp manifest:

apiVersion: nmstate.io/v1
kind: NodeNetworkConfigurationPolicy
metadata:
  name: ip1
spec:
  nodeSelector:
    kubernetes.io/hostname: k3s-server
  desiredState:
    interfaces:
    - name: eth0
      type: ethernet
      state: up
      ipv4:
        enabled: true
        dhcp: false
        address:
        - ip: 10.1.1.55
          prefix-length: 24
    routes:
      config:
        - destination: 0.0.0.0/0
          next-hop-address: 10.1.1.1
          next-hop-interface: eth0
          metric: 100
          source: 10.1.1.55
    dns-resolver:
      config:
        search:
        - .
        server:
        - 8.8.4.4

Anything else we need to know?:

Environment:

• nmstate-operator version: v0.83.0
• OS: SUSE micro SLE 5.5
• k3s version: v1.31.1+k3s1

[qinqon]

Can you dump the whole error ?

[liushan9981]

The log file is too large, please see the attachment.

• The file nmstate-handler-qmgxs.log contains the logs from the pod nmstate-handler-qmgxs.
• The file nncp-describe.txt contains the output of the command kubectl describe nncp <obj_name>.

nncp-describe.txt
nmstate-handler-qmgxs.log

[qinqon]

So api-server probe is failing at

"failed to creating new custom client","error":"Get \"https://10.195.0.1:443/api?timeout=32s\": dial tcp 10.195.0.1:443: connect: no route to host"

And the routing table looks like

routes:                                                                         
  running:                                                                      
  - destination: 0.0.0.0/0                                                      
    next-hop-interface: eth0                                                    
    next-hop-address: 10.1.1.1                                                  
    metric: 100                                                                 
    table-id: 254                                                               
    source: 10.1.1.55                                                           
  config:                                                                       
  - destination: 0.0.0.0/0                                                      
    next-hop-interface: eth0                                                    
    next-hop-address: 10.1.1.1                                                  
    metric: 100                                                                 
    table-id: 254                                                               
    source: 10.1.1.55                   

[qinqon]

I think is not able to access 10.1.1.1, like the node is not receiving the ARP to resolve the mac of tha address.