Description of problem:

The nfs ganesha service is deployed using cephadm via the spec:

---
service_type: nfs
service_id: standalone_nfs
service_name: nfs.standalone_nfs
placement:
  hosts:
    - oc0-controller-0
    - oc0-controller-1
    - oc0-controller-2
spec:
  namespace: ganesha
  pool: manila_data

where the port and a few other parameters can be specified.
However, when the spec section described above is processed, this component is
started on  *::<port>, instead of using the cluster network or the ip address on a specific network defined by the operator.
This is a gap compared to ceph-ansible, which is able to provide this kind of feature via [1]


[1] https://github.com/ceph/ceph-ansible/blob/master/roles/ceph-nfs/templates/ganesha.conf.j2

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


How reproducible:


Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:


Additional info: