Supported options on an NSP cluster

Description

This appendix describes the supported options for network interfaces on an NSP cluster.

For all scenarios:

  • Each NSP cluster node that is an ingress gateway must have a network interface address that belongs to the same subnet as each VIP configured in the cluster.

  • Each specified VIP must be different from other VIPs or IP addresses on the cluster nodes.

  • trapForwarder VIP is required only when MDM is deployed.

  • flowForwarder VIP is required only when Flow Collector is deployed.

  • The ingressController mediation VIP is required only when there are a dedicated mediation network and network elements that require access to applications through the ingress controller. An example of a network element is a zero-touch provisioning (ZTP) node that needs access the file-server application.

The examples of nsp-config.yml and k8s-deployer.yml use specific IP addresses to identify the subnets to which they belong.

nsp-config.yml has the following IP addresses in the ingressApplications section:

ingressApplications:

   ingressController:

     clientAddresses:

       virtualIp: "client_IP"

     internalAddresses:

       virtualIp: "internal_IP"

     mediationAddresses:

       virtualIp: "mediation_IP"

   trapForwarder:

     mediationAddresses:

       virtualIpV4: "trapV4_mediation_IP"

   flowForwarder:

     mediationAddresses:

       virtualIpV4: "flowV4_mediation_IP"

k8s-deployer.yml has the following IP addresses in the loadBalancerExternalIps section:

loadBalancerExternalIps:

  - client_IP

  - internal_IP

  - trapV4_mediation_IP

  - flowV4_mediation_IP

where

client_IP is the address for external client access

internal_IP is the address for internal communication

mediation_IP is the address for network mediation

trapV4_mediation_IP is the address for IPv4 network mediation

flowV4_mediation_IP is the address for IPv4 flow collection