MetalLB Version

0.14.5

Deployment method

Charts

Main CNI

calico

Kubernetes Version

No response

Cluster Distribution

No response

Describe the bug

Related issue: #1339

here is the service:

apiVersion: v1
kind: Service
metadata:
name: nginx-ingress
namespace: nginx-ingress
spec:
allocateLoadBalancerNodePorts: true
clusterIP: 10.0.0.79
clusterIPs:

• 10.0.0.79
• fd00::d899
  externalTrafficPolicy: Local
  healthCheckNodePort: 30272
  internalTrafficPolicy: Cluster
  ipFamilies:
• IPv4
• IPv6
  ipFamilyPolicy: PreferDualStack
  ports:
• name: http
  nodePort: 30122
  port: 80
  protocol: TCP
  targetPort: http
• name: https
  nodePort: 32678
  port: 443
  protocol: TCP
  targetPort: https
  selector:
  app.kubernetes.io/instance: nginx-ingress
  app.kubernetes.io/name: nginx-ingress
  sessionAffinity: None
  type: LoadBalancer
  status:
  loadBalancer: {}
  Because the service is dual stack, but metallb has no IPv6 configured in any pool, the controller fails to allocate an IPv6 address and stops there.

To Reproduce

Steps:

1. Create IPaddresspool pool1 only for IPv4 family.
2. In case service created with addresspool annotation as pool1. and ipFamilyPolicy is 'PreferDualStack'.
3. Expected is the service get IP address assigned from ipv4 family only. But actually service stuck in pending state for LBIP.
4. The same behavior can be seen without addresspool annotation if all pools are only IPv4 family.

Expected Behavior

As mentioned above

Additional Context

If there is only single IP family is available in metallb ipaddresspool (no IPv6 address defined) :

1. In case of preferDualstack metallb controller should allocate that IP
2. In case of RequireDualStack it should not allocate any IP and status will be in pending state.

Currently with respect to metallb controller there is no difference between RequireDualStack and preferDualstack and both are treated as mentioned above as 'RequireDualStack'.

I've read and agree with the following

• I've checked all open and closed issues and my request is not there.
• I've checked all open and closed pull requests and my request is not there.

I've read and agree with the following

• I've checked all open and closed issues and my issue is not there.
• This bug is reproducible when deploying MetalLB from the main branch
• I have read the troubleshooting guide and I am still not able to make it work
• I checked the logs and MetalLB is not discarding the configuration as not valid
• I enabled the debug logs, collected the information required from the cluster using the collect script and will attach them to the issue
• I will provide the definition of my service and the related endpoint slices and attach them to this issue