Egress + service without selector

[gulecroc]

Describe the bug
I would like to set the egress IP to connect to a service outside the k8s cluster.

For this purpose, I create an endpointslice :

apiVersion: discovery.k8s.io/v1
kind: EndpointSlice
metadata:
  name: my-service-1
  namespace: my-namespace
  labels:
    kubernetes.io/service-name: my-service
addressType: IPv4
ports:
  - name: ldap
    port: 1234
    protocol: TCP
endpoints:
  - addresses:
      - "1.2.3.4"

And a service without selector : https://kubernetes.io/docs/concepts/services-networking/service/#services-without-selectors

With kube-vip egress requirements : https://kube-vip.io/docs/usage/egress/#using-kube-vip-egress

apiVersion: v1
kind: Service
metadata:
  annotations:
    kube-vip.io/egress: "true"
  name: my-service
  namespace: my-namespace
spec:
  externalTrafficPolicy: Local
  ports:
    - name: ldap
      port: 1234
      protocol: TCP
      targetPort: 1234
  type: LoadBalancer
  loadBalancerIP: 5.6.7.8

But the external-ip stay in pending state :

> kubectl -n my-namespace get svc my-service
NAME                                TYPE           CLUSTER-IP      EXTERNAL-IP   PORT(S)            AGE
my-service                  LoadBalancer   10.43.185.136   <pending>     1234:31787/TCP      10s

If I add a selector to my service, or If I remove externalTrafficPolicy: Local, the external IP is correctly set. But I don't need a selector for egress purpose, and I need externalTrafficPolicy: Local from kube-vip documentation.

Environment (please complete the following information):

• OS/Distro: Ubuntu 20.04
• Kubernetes Version: 1.26.15
• Kube-vip Version: 0.8.0

Kube-vip.yaml:

> kubectl -n kube-vip get ds kube-vip-load-balancer -o yaml
apiVersion: apps/v1
kind: DaemonSet
metadata:
  annotations:
    deprecated.daemonset.template.generation: "9"
    meta.helm.sh/release-name: kube-vip-load-balancer
    meta.helm.sh/release-namespace: kube-vip
  creationTimestamp: "2024-04-24T12:41:02Z"
  generation: 9
  labels:
    app.kubernetes.io/managed-by: Helm
  name: kube-vip-load-balancer
  namespace: kube-vip
  resourceVersion: "783636164"
  uid: 395ecfca-20d2-446a-a628-c35698bbdb22
spec:
  revisionHistoryLimit: 10
  selector:
    matchLabels:
      app.kubernetes.io/instance: kube-vip-load-balancer
      app.kubernetes.io/name: kube-vip-load-balancer
  template:
    metadata:
      creationTimestamp: null
      labels:
        app.kubernetes.io/instance: kube-vip-load-balancer
        app.kubernetes.io/name: kube-vip-load-balancer
    spec:
      affinity:
        nodeAffinity:
          requiredDuringSchedulingIgnoredDuringExecution:
            nodeSelectorTerms:
            - matchExpressions:
              - key: kube-vip-load-balancer
                operator: In
                values:
                - default
      containers:
      - args:
        - manager
        env:
        - name: cp_enable
          value: "false"
        - name: lb_enable
          value: "false"
        - name: lb_port
          value: "6443"
        - name: prometheus_server
          value: :2113
        - name: svc_election
          value: "true"
        - name: svc_enable
          value: "true"
        - name: vip_arp
          value: "true"
        - name: vip_cidr
          value: "32"
        - name: vip_interface
          value: ens160
        - name: vip_leaderelection
          value: "false"
        - name: vip_loglevel
          value: "5"
        image: ghcr.io/kube-vip/kube-vip-iptables:v0.8.0
        imagePullPolicy: IfNotPresent
        name: kube-vip
        resources: {}
        securityContext:
          capabilities:
            add:
            - NET_ADMIN
            - NET_RAW
        terminationMessagePath: /dev/termination-log
        terminationMessagePolicy: File
      dnsPolicy: ClusterFirst
      hostNetwork: true
      restartPolicy: Always
      schedulerName: default-scheduler
      securityContext: {}
      serviceAccount: kube-vip-load-balancer
      serviceAccountName: kube-vip-load-balancer
      terminationGracePeriodSeconds: 30
      tolerations:
      - effect: NoSchedule
        key: node-role.kubernetes.io/control-plane
        operator: Exists
  updateStrategy:
    rollingUpdate:
      maxSurge: 0
      maxUnavailable: 1
    type: RollingUpdate
status:
  currentNumberScheduled: 2
  desiredNumberScheduled: 2
  numberAvailable: 2
  numberMisscheduled: 0
  numberReady: 2
  observedGeneration: 9
  updatedNumberScheduled: 2

Additional context
Add any other context about the problem here.