We have a kubernetes cluster which we deploy using aks-engine and we recently moved from VM to VMSS as nodes. But, when we deploy a service using helm, the backend pools of the load balancer that gets created doesn't get associated to the Virtual Machine Scale Set that's created as part of the deployment. We expect it to be associated to it by default. Right now, we need to manually perform this step. Can we have a way to automatically do this ?

Here's the kubernetes.json file

    {  
   "apiVersion":"vlabs",
   "properties":{  
      "orchestratorProfile":{  
         "orchestratorType":"Kubernetes",
         "orchestratorRelease":"1.12",
         "kubernetesConfig":{  
            "networkPolicy":"azure",
            "useManagedIdentity":true
         }
      },
      "masterProfile":{  
         "count":1,
         "dnsPrefix":"testcluster",
         "vmSize":"Standard_D2_v2"
      },
      "agentPoolProfiles":[  
         {  
            "name":"windowspool2",
            "count":2,
            "vmSize":"Standard_D2_v2",
            "availabilityProfile":"VirtualMachineScaleSets",
            "osType":"Windows",
            "osDiskSizeGB":127
         }
      ],
      "windowsProfile":{  
         "adminUsername":"azureuser",
         "adminPassword":"test1234",
         "windowsPublisher":"MicrosoftWindowsServer",
         "windowsOffer":"WindowsServerSemiAnnual",
         "windowsSku":"Datacenter-Core-1803-with-Containers-smalldisk"
      },
      "linuxProfile":{  
         "adminUsername":"azureuser",
         "ssh":{  
            "publicKeys":[  
               {  
                  "keyData":<ssh-key>
               }
            ]
         }
      },
      "servicePrincipalProfile":{  
         "clientId":         <SP-client ID> ,
         "secret":<SP-client passowrd>
      }
   }
}