The Customer Access Network (CAN) and Customer Management Network (CMN) needs to be connected to both spines in a dual-spine configuration so that each spine can access the outside network. However, the NCNs should only have one default gateway. Therefore, the multi-active gateway protocol (MAGP) on the Mellanox spines can be used to create a virtual router gateway IP address that can direct to either of the spines, depending on the state of the spines. The Virtual Switching Extension (VSX) for Aruba spines serve the same purpose.
The following is an example of the point-to-point configuration on the spine switches. The IP address should be replaced with the IP address chosen by the customer that matches the switch configuration.
In CSM 1.2 the Customer VRF is introduced, this requires the interface that is connected to the site to be configured to use this VRF.
Mellanox:
interface ethernet 1/11 speed auto force
interface ethernet 1/11 description to-can
interface ethernet 1/11 no switchport force
interface ethernet 1/11 vrf forwarding Customer
interface ethernet 1/11 ip address 10.101.15.150/30 primary
Aruba:
interface 1/1/36
no shutdown
vrf attach Customer
description to-can
ip address 10.101.15.150/30
exit
There must then be two routes on the customer’s switch directing traffic for the customer_access_network
and customer_management_network
subnet to the endpoint on the spine switch. The following is an example
of the route configuration on the customer switch.
ip route 10.103.9.0/25 10.101.15.150
ip route 10.103.9.0/25 10.101.15.152
The next hop IP address 10.101.15.150
would be the interface IP address on the Spine switch.
There must be a default route on each spine switch that will direct traffic that does not match other routes to the endpoint on the customer switch. The following examples are for the route configuration on sw-spine-001
.
Mellanox:
ip route vrf Customer 0.0.0.0/0 10.101.15.149
Aruba:
ip route 0.0.0.0/0 10.101.15.149 vrf Customer
The connection between the distribution/site switch and the spines require two separate uplinks from the spine switch to the distribution switch. Two static routes need to be created on the distribution switch to route the CAN subnet to each of the spine switches. These routes will have equal cost (ECMP) to split the load across the two spines and provide redundancy if one of the spines should go down.
Example
interface 1/1/41
no shutdown
description WASP spine-001 1/11
ip address 10.101.15.149/30
interface 1/1/42
no shutdown
description WASP spine-002 1/11
ip address 10.101.15.153/30
ip route 10.102.5/26 10.101.15.150
ip route 10.102.5/26 10.101.15.154