HPE2-W09: Aruba Data Center Network Specialist

Automated provisioning of LAGs between AHV and VSX is an operation supported when Aruba Fabric Composer (AFC) is integrated with Nutanix. AFC is a tool that provides automation and orchestration for managing data center networks composed of ArubaOS-CX switches. AFC can integrate with various data center software such as VMware vSphere, Nutanix AHV, Microsoft Hyper-V, etc. AFC can discover, monitor, and configure Nutanix AHV clusters and hosts using REST APIs. AFC can also automate the provisioning of LAGs between AHV and VSX by creating VSX LAGs or MC-LAGs on the ArubaOS-CX switches and configuring the corresponding LAGs on the AHV hosts1.

Different virtual MAC must be used for IPv4 and IPv6 Active Gateway is not a true statement about EVPN and IPv6 for configuring an EVPN solution for a dual-stack IPv4 and IPv6 protocol in the overlay networks. Active Gateway is a feature that provides first-hop redundancy for hosts connected to VTEPs using anycast gateway addresses. Active Gateway can use the same virtual MAC address for both IPv4 and IPv6 protocols on the same VNI2.

Implement Virtual Switching Extension (VSX) on pairs of ERPS switches at the same site. Then combine multiple links between two data centers into VSX LAGs (MC-LAGs) is not part of a valid strategy for load sharing traffic across the links in an Ethernet Ring Protection Switching (ERPS) ring. ERPS is a feature that provides loop prevention and fast convergence for Layer 2 networks that use ring topologies. VSX is a feature that provides active-active forwarding and redundancy for ArubaOSCX switches. VSX LAGs or MC-LAGs are LAGs that span across two VSX nodes and provide load balancing and resiliency. However, VSX LAGs or MC-LAGs are not supported by ERPS because they can create loops in the ring topology. A better way to load share traffic across the links in an ERPS ring would be to use link aggregation groups (LAGs) between two nodes in a ring as long as they are not multi-chassis LAGs (MC-LAGs)1.

To help the switch to look inside tunneled traffic and apply different quality of service (QoS) settings to different types of traffic is not a use case for implementing Enhanced Transmission Selection (ETS) on an ArubaOS-CX switch. ETS is a feature that provides bandwidth allocation and priority assignment for different traffic classes based on IEEE 802.1Qaz standard. ETS does not help the switch to look inside tunneled traffic, but rather relies on the priority values in the outer header of the tunneled traffic to apply QoS settings. A better way to help the switch to look inside tunneled traffic and apply different QoS settings to different types of traffic would be to use deep packet inspection (DPI) or application visibility and control (AVC) features.

Place all three VRFs in the same OSPF process on Switch-1 is not a valid strategy for meeting this goal of enabling devices in VRF B and VRF C to reach shared resources in VRF A. This strategy would not work because OSPF does not support multiple VRFs in the same process on ArubaOS-CX switches. Each VRF must have its own OSPF process with a unique process ID1.

On Switch-1, set 192.168.1.3 as a peer IP address in the VNI 5020 context is part of the process for setting up VXLAN to meet the requirements of enabling servers to be part of the same VXLANs and VNIs as shown in the exhibit. Switch-1, Switch-2, and Switch-3 are ArubaOS-CX switches that use VXLAN to provide Layer 2 extension over Layer 3 networks without EVPN. VXLAN is a feature that uses UDP encapsulation to tunnel Layer 2 frames over Layer 3 networks using VNIs. To set up VXLAN without EVPN on Switch-1, you need to do the following steps:

Configure loopback interfaces with IP addresses on each switch

Configure VLAN interfaces with IP addresses on each switch

Configure VXLAN interfaces with VNIs on each switch

Configure peer IP addresses for each VNI on each switch

Configure static routes or dynamic routing protocols to enable reachability between loopback interfaces On Switch-1, setting 192.168.1.3 as a peer IP address in the VNI 5020 context means that Switch-1 can send and receive VXLAN traffic for VNI 5020 to and from Switch-2, which has the loopback interface with IP address 192.168.1.3.

The ISL and keepalive goes down, and after a few seconds, the keepalive link restores. Switch-1 and Switch-2 remains up. The Split-recovery mode is enabled. In this case the secondary switch shutdowns SVIs when keepalive is restored is a correct description of how the Virtual Switching Extension (VSX) fabric reacts to various component failure scenarios. VSX is a feature that provides active-active forwarding and redundancy for ArubaOS-CX switches. The ISL is the inter-switch link that connects two VSX nodes and carries data traffic. The keepalive link is a separate link that carries control traffic between two VSX nodes. The split-recovery mode is a feature that prevents split-brain scenarios when both VSX nodes lose connectivity with each other but remain up. When the ISL and keepalive goes down, both VSX nodes continue to forward traffic independently. When the keepalive link restores, the secondary switch detects that it has lost synchronization with the primary switch and shuts down its SVIs to prevent traffic loops1.

Work with the server admins to assign a consistent VLAN for VMs 1 and 4. Assign interface 1/1/2 on Switch-1 to the same VLAN is part of a valid configuration to meet the requirements for providing VXLAN services for several VMs and servers using ArubaOS-CX switches. VMs 1 and 4 belong to the same VXLAN segment (VNI 5010), so they should be assigned to the same VLAN on their respective hypervisors. Interface 1/1/2 on Switch-1 should also be assigned to the same VLAN as VMs 1 and 4, so that Switch-1 can act as a VTEP for them1.