I85PREP
⌘ K
Aruba Data Center Professional study materials
Aruba Data Center Professional study resources

Aruba Data Center Professional practice questions

One question at a time. Build your Aruba Data Center Professional understanding.

First 50 of 199 questions. Each one lists the choices, the correct answer, and an explanation. The checker at the top of this page saves progress in this browser when it is available.

Continue in the practice tool above to load more questions, 50 at a time.

Question 1Data Center Network Architecture and Design

In a two-tier spine-leaf data center fabric, what is the primary function of the spine layer?

  1. Aggregating access VLANs and running Rapid Spanning Tree
  2. Connecting the fabric to the WAN edge and firewalls
  3. Hosting the default gateways for all server VLANs
  4. Providing equal-cost paths between all leaves Correct answer.

Explanation

Every spine connects to every leaf, so any leaf reaches any other leaf in exactly two hops regardless of which spine is used. This uniform hop count and equal path cost is what allows the fabric to scale bandwidth as leaves are added. Gateways, WAN edge, and VLAN aggregation are functions placed at the leaf or border, not at the spine core.

Question 2Data Center Network Architecture and Design

How does a leaf-spine fabric spread traffic across multiple uplinks from a single leaf?

  1. By assigning each server a different default gateway
  2. By hashing each flow across the equal-cost uplinks Correct answer.
  3. By using Rapid Spanning Tree to block redundant links
  4. By dedicating one uplink per VLAN

Explanation

All uplinks from a leaf to the spines are kept forwarding because the underlay is routed, so the switch needs a way to pick among equal-cost next hops. ECMP hashes each flow onto one of those paths, keeping a single flow in order while distributing many flows. Spanning tree blocking and per-VLAN uplink assignment are Layer 2 workarounds this design removes.

Question 3Data Center Network Architecture and Design

Which statement describes the overhead that VXLAN encapsulation adds to a frame?

  1. About 50 bytes total across outer Ethernet, IP, UDP, and VXLAN headers Correct answer.
  2. Exactly 8 bytes, because VXLAN only inserts a shim header between the original frame and the wire
  3. About 36 bytes, because VXLAN replaces the original Ethernet header entirely
  4. About 4 bytes, because only an extra 802.1Q tag is needed per frame

Explanation

VXLAN adds an 8-byte VXLAN header inside an 8-byte UDP header, a 20-byte outer IP header, and a 14-byte outer Ethernet header, for 50 bytes total. The original inner frame is preserved and carried intact. This is why underlay MTU must be raised when the inner payload is already at 1500 or 9000 bytes.

Question 4Data Center Network Architecture and Design

What is the purpose of the VSX keepalive link between two VSX peers?

  1. To check peer liveness when the ISL is down Correct answer.
  2. To provide a backup path for the management VLAN
  3. To carry all user data traffic if the ISL fails
  4. To synchronize MAC address tables between the peers

Explanation

The keepalive link is a separate path used only for peer liveness signalling, so each switch can tell whether the other is still alive even if the ISL breaks. Synchronization of MAC and ARP state happens over the ISL, not the keepalive. The keepalive carries no user traffic.

Question 5Data Center Network Architecture and Design

Why do large leaf-spine fabrics commonly avoid Layer 2 spanning tree?

  1. Because spanning tree blocks all leaf uplinks by design
  2. Because VXLAN disables spanning tree on the physical switches
  3. Because 802.1w cannot run on 25G or 100G links
  4. Because L3 routing to every leaf removes Layer 2 loops Correct answer.

Explanation

When each leaf routes to every other leaf over a routed underlay, there is no Layer 2 loop to block, so all uplinks can forward and ECMP can be used. Spanning tree would otherwise leave half the links idle. VXLAN overlays ride on top of that routed underlay rather than replacing the need for it.

Question 6Data Center Network Architecture and Design

Which traffic pattern is described as east-west in a data center?

  1. Traffic entering from the internet and leaving to clients
  2. Traffic between servers inside the same data center Correct answer.
  3. Traffic from a backup appliance to offsite tape
  4. Traffic between the data center and a branch office

Explanation

East-west traffic stays inside the data center, typically between servers, tiers, or storage nodes in a distributed application. North-south traffic crosses the data center boundary to clients or remote sites. Modern designs size the fabric primarily for east-west volume.

Question 7Data Center Network Architecture and Design

A leaf has 48 x 25G server-facing ports and 4 x 100G uplinks to the spines. What is its oversubscription ratio?

  1. 2:1
  2. 4:1
  3. 1:1
  4. 3:1 Correct answer.

Explanation

Server-facing capacity is 48 x 25G = 1200G and uplink capacity is 4 x 100G = 400G. Dividing capacity toward the servers by capacity toward the core gives 1200/400 = 3:1. The ratio expresses how much server bandwidth is statistically shared across each unit of uplink bandwidth.

Question 8Data Center Network Architecture and Design

When is a direct attach copper cable normally preferred over optical fiber?

  1. When the run stays within a rack or between adjacent racks Correct answer.
  2. When the link must pass between separate buildings
  3. When the ports need to support long-wave optics
  4. When connecting racks more than 300 meters apart

Explanation

DAC assemblies are passive or active copper twinax with limited reach, so they fit short runs such as within a rack or to the neighboring rack. They cost less and draw less power than optics at those distances. Longer runs require fiber because copper attenuation and equalization limits are exceeded.

Question 9Data Center Network Architecture and Design

Which BGP EVPN route type advertises a host MAC address, optionally together with its IP address?

  1. Type 4
  2. Type 5
  3. Type 3
  4. Type 2 Correct answer.

Explanation

The Type 2 MAC/IP advertisement route carries a tenant MAC, and when the host's IP is known it carries that too as an optional attribute. This lets remote VTEPs populate both bridging and ARP/ND suppression tables. Type 3 handles BUM membership and Type 5 handles IP prefixes.

Question 10Data Center Network Architecture and Design

All leaf switches in a VXLAN fabric use the same IP and MAC address for a tenant gateway. What does this design achieve?

  1. It removes the need for a routing protocol in the underlay
  2. It allows the gateway MAC to change per host
  3. It lets a host route through its local leaf without tromboning Correct answer.
  4. It forces all inter-subnet traffic through a central firewall

Explanation

With a distributed anycast gateway, every leaf owns the same gateway IP and MAC, so a host's default gateway is always local. The packet is routed once at the ingress leaf and then carried over VXLAN to the destination leaf, avoiding a hairpin through a central gateway. Underlay routing is still required to reach the remote VTEP.

Question 11Data Center Network Architecture and Design

What is a VTEP in a VXLAN overlay?

  1. A physical cable that bonds two leaf switches together
  2. A routing policy that filters tenant prefixes
  3. A tunnel endpoint that encapsulates and decapsulates frames Correct answer.
  4. A table that maps VLAN IDs to spanning tree instances

Explanation

A VXLAN tunnel endpoint sits at the edge of the overlay and wraps a tenant frame in the VXLAN/UDP/IP outer headers, then strips them at the far end. Its IP address is the loopback reachable through the routed underlay. Each VTEP maintains a mapping of remote VTEPs per VNI.

Question 12Data Center Network Architecture and Design

Why is BGP EVPN preferred over flood-and-learn VXLAN for MAC learning?

  1. It distributes MAC and IP reachability, cutting unknown-unicast flooding Correct answer.
  2. It compresses the VXLAN header to 4 bytes
  3. It replaces ECMP with a single active uplink
  4. It removes the need for an underlay routing protocol, because the overlay itself carries all physical link state and reachability information

Explanation

EVPN advertises MAC and IP reachability in the control plane, so a leaf knows which remote VTEP owns a host before sending traffic. Unknown unicast and ARP flooding are therefore largely eliminated, and MAC aging problems from silent hosts disappear. The data plane encapsulation is unchanged from flood-and-learn VXLAN.

Question 13Data Center Network Architecture and Design

In a VSX pair, the ISL fails but the keepalive link stays up. What does the secondary peer do?

  1. Shuts down its VSX LAG member ports to prevent a split-brain Correct answer.
  2. Keeps all VSX LAG member ports forwarding
  3. Reboots automatically to re-form the pair
  4. Takes over the primary role and the primary shuts down

Explanation

If both peers kept forwarding while unable to synchronize over the ISL, dual-homed devices could see duplicate frames and the pair could act as two independent switches. The secondary detects the ISL loss through the still-healthy keepalive and disables its multichassis LAG members. The primary continues forwarding, so connectivity survives through the remaining links.

Question 14Data Center Network Architecture and Design

What does a VRF provide on a data center switch?

  1. A separate Layer 3 routing table for each tenant or function Correct answer.
  2. An encrypted tunnel between leaf switches
  3. A separate Layer 2 broadcast domain per rack
  4. A hardware queue reserved for lossless traffic

Explanation

A VRF is a distinct routing table instance, so overlapping address space from different tenants can coexist on the same physical switch. Each VRF is typically mapped to its own Layer 3 VNI in an EVPN fabric. VRFs partition Layer 3 reachability, not Layer 2 broadcast domains.

Question 15Data Center Network Architecture and Design

The underlay must carry VXLAN-encapsulated frames whose inner payload is 9000 bytes. Which underlay MTU is required?

  1. 9014 bytes
  2. 9050 bytes Correct answer.
  3. 9000 bytes
  4. 9100 bytes

Explanation

VXLAN adds 50 bytes of outer headers to the original frame, so the underlay must be able to carry 9000 + 50 = 9050 bytes. Setting the underlay to the inner payload size would cause the encapsulated frames to be dropped or fragmented. Most designs round up to 9216 on all fabric links.

Question 16Data Center Network Architecture and Design

Which fields are typically used to hash flows across equal-cost uplinks while keeping a flow on one path?

  1. Only the VLAN ID assigned to the server port
  2. Only the source IP address
  3. Only the destination MAC address
  4. Source/destination IP, protocol, and L4 ports Correct answer.

Explanation

Hashing on the five-tuple spreads many concurrent flows across all equal-cost paths while keeping packets of a single flow in order. Destination MAC alone would collapse traffic toward the router MAC onto one link. The hash must include enough entropy to avoid polarization.

Question 17Data Center Network Architecture and Design

Which mechanism provides lossless behaviour for RoCEv2 traffic on a converged Ethernet fabric?

  1. Spanning tree on the storage VLAN
  2. Static ARP entries for storage nodes
  3. LACP fast timers on server bonds
  4. Priority Flow Control on a dedicated priority Correct answer.

Explanation

PFC pauses only the priority carrying storage traffic when buffers fill, preventing drops without pausing the whole link. It is normally combined with ECN marking and a lossless QoS pool. The other options do not prevent buffer overflow on congested ports.

Question 18Data Center Network Architecture and Design

What role does Aruba Fabric Composer play in an Aruba CX data center fabric?

  1. It replaces BGP EVPN with a proprietary control plane
  2. It provides only SNMP polling of interface counters
  3. It is a physical line card for 400G uplinks
  4. It orchestrates and automates fabric-wide configuration and policy Correct answer.

Explanation

Fabric Composer is a software orchestration layer that builds and maintains fabric-wide intent such as EVPN-VXLAN configuration, VSX pairing, and policy. It works with the standard protocols on the switches rather than replacing them. It is not a hardware component or a monitoring-only tool.

Question 19Data Center Network Architecture and Design

A server is dual-homed with a single LACP bond to two VSX peers. How is that bond presented to the server?

  1. A single logical LAG with both links active Correct answer.
  2. Two separate bonds, one per switch, with active/standby
  3. A bond that requires the server to run a routing protocol
  4. A bond that depends on spanning tree to forward

Explanation

A multichassis LAG lets the two VSX peers present one logical LAG to the server, so both links forward simultaneously. The peers coordinate LACP state over the ISL. This removes the need for active/standby bonding or spanning tree on the server links.

Question 20Data Center Network Architecture and Design

In an Aruba EVPN-VXLAN data center fabric, which routing protocol is commonly used for the routed underlay between leaf and spine?

  1. eBGP between leaf and spine Correct answer.
  2. RIPng
  3. OSPFv2 only
  4. IS-IS level 1 only

Explanation

Aruba EVPN-VXLAN reference designs commonly use eBGP for the underlay because it scales cleanly with per-leaf autonomous systems and provides straightforward policy control. BGP is then also used for the EVPN address family in the overlay. RIPng and single-area IS-IS are not typical underlay choices here.

Question 21Data Center Network Architecture and Design

A leaf already holds a Type 2 MAC/IP route for a host in a remote subnet. What does ARP suppression do when a local host ARPs for that remote host?

  1. It converts ARP requests into ICMP redirects
  2. It ages the host route out faster than the MAC entry
  3. It answers ARP locally from the EVPN table rather than flooding Correct answer.
  4. It blocks all ARP requests from leaving the host VLAN

Explanation

Because the leaf knows the remote host's MAC and IP from the EVPN route, it can proxy the ARP reply itself. The request never floods across the VXLAN overlay, which reduces BUM traffic and ARP storms. Only unresolved addresses are flooded.

Question 22Data Center Network Architecture and Design

Which EVPN route type carries an IP prefix so that traffic can be routed between VRFs across the fabric?

  1. Type 3
  2. Type 1
  3. Type 8
  4. Type 5 Correct answer.

Explanation

The Type 5 IP prefix route advertises an IP prefix, optionally with a gateway and VNI, which is what symmetric or asymmetric inter-VRF routing needs. It complements Type 2 host routes by summarizing subnets. Type 1 and Type 3 serve multihoming and BUM membership respectively.

Question 23Data Center Network Architecture and Design

A fabric must detect a failed leaf-spine link and reconverge in well under one second. Which approach best fits?

  1. Disable ECMP so one path remains
  2. Rely on LACP slow timers
  3. Lower the OSPF hello interval to 30 seconds
  4. Run BFD on the underlay routing adjacencies Correct answer.

Explanation

BFD provides sub-second failure detection independent of the routing protocol's own timers, so the adjacency is torn down almost immediately. The routing protocol then withdraws and reconverges over the surviving equal-cost paths. Tuning hello timers is slower and riskier than adding BFD.

Question 24Data Center Network Architecture and Design

How does symmetric IRB handle inter-VRF traffic between leaf switches in an EVPN-VXLAN fabric?

  1. A shared border gateway always performs all inter-VRF routing
  2. The ingress leaf rewrites the destination MAC and floods the frame on the tenant VLAN
  3. The spine performs the VRF lookup on every packet
  4. The ingress leaf routes into an L3 VNI and the egress leaf routes out Correct answer.

Explanation

In symmetric IRB the ingress leaf performs the first lookup and sends the packet over the Layer 3 VNI with both source and destination VNIs carried in the encapsulation. The egress leaf performs the second lookup into the destination VRF. This keeps routing distributed and avoids a central hairpin.

Question 25Data Center Network Architecture and Design

In an EVPN-VXLAN fabric using ingress replication, how is BUM traffic delivered?

  1. The underlay builds PIM-SM multicast trees for each VNI
  2. BUM traffic is dropped because EVPN suppresses all flooding
  3. The ingress VTEP unicasts a copy to each VTEP in the VNI's IMET list Correct answer.
  4. The spine replicates the frame to all leaves in the fabric regardless of membership

Explanation

Each VTEP advertises a Type 3 IMET route when it has a VNI with interested ports, and the ingress VTEP makes one unicast copy per remote VTEP in that list. This avoids multicast routing in the underlay and scopes replication to interested leaves only. It is the default mode in most EVPN fabrics.

Question 26Data Center Network Architecture and Design

In a two-tier spine-leaf fabric where every leaf connects to every spine, what is the main design benefit?

  1. Layer 2 domains are extended across the entire fabric
  2. Broadcast traffic is flooded to every spine for faster learning
  3. Consistent hop count and equal-cost paths between hosts Correct answer.
  4. Spine-to-spine links provide redundant backup paths

Explanation

Because every leaf is exactly one hop from every other leaf via a spine, all host-to-host paths have the same hop count and are equal cost, so ECMP can use every uplink and latency is predictable. This is the defining property of a full-mesh spine-leaf topology used in modern data centers.

Question 27Data Center Network Architecture and Design

In a VXLAN overlay, what does mapping a VLAN to a VNI on a VTEP accomplish?

  1. It binds the segment to a VXLAN network identifier Correct answer.
  2. It encrypts each frame before it crosses the underlay
  3. It assigns the VLAN an IP subnet automatically
  4. It converts the VLAN into a routed VRF instance

Explanation

The VLAN-to-VNI mapping defines which VXLAN network identifier carries that Layer 2 segment, and performs the tag translation in each direction at the VTEP. Frames from that VLAN are encapsulated with that VNI; frames arriving with that VNI are placed back into the local VLAN.

Question 28Data Center Network Architecture and Design

What is the purpose of the VSX keepalive link between two peer switches?

  1. It lets peers distinguish a dead peer from a failed ISL Correct answer.
  2. It synchronises MAC and ARP tables between the peers
  3. It carries user data traffic when the ISL is congested
  4. It load-balances broadcast traffic between the two peers

Explanation

The keepalive is a lightweight control path used for heartbeats, so each peer can tell whether the other switch is alive or whether only the inter-switch link has failed. That distinction drives the correct split-brain response, such as a peer taking its VSX links down.

Question 29Data Center Network Architecture and Design

Which benefit comes from terminating Layer 2 at the leaf and routing between subnets in the fabric?

  1. Broadcast domains become larger and simpler to manage
  2. Spanning tree is used to block redundant leaf-spine links
  3. MAC tables are synchronised across all leaf switches
  4. All uplinks stay active and forward traffic at once Correct answer.

Explanation

When each subnet ends on its leaf and traffic between subnets is routed over the spine layer, every leaf-spine link can be used simultaneously by ECMP. Spanning tree is no longer needed to break loops in the fabric, so no capacity is left idle.

Question 30Data Center Network Architecture and Design

What is the role of a border leaf in an EVPN-VXLAN data center fabric?

  1. It connects the fabric to external networks or other sites Correct answer.
  2. It is the only leaf permitted to learn remote MAC addresses
  3. It stores the configuration for all other leaf switches
  4. It acts as the spanning tree root for every VLAN in the fabric

Explanation

The border leaf provides the exit and entry point between the overlay fabric and external domains such as the WAN, the core, a legacy data center, or another EVPN site. It peers with external routers and often handles Layer 3 handoff, route leaking, and inter-site VXLAN gateways.

Question 31Data Center Network Architecture and Design

A fabric is designed to use multicast replication in the underlay for VXLAN BUM traffic. What does an ingress VTEP do with such a frame?

  1. It floods the frame to all spines, which forward it to every leaf
  2. It sends one unicast copy to every remote VTEP listed in its configuration
  3. It sends a single copy to a multicast group joined by interested VTEPs Correct answer.
  4. It drops BUM traffic and relies on ARP suppression instead

Explanation

With multicast replication the underlay builds a shared distribution tree, and each VTEP joins the group associated with the VNI. The ingress VTEP sends one encapsulated copy to the group address and the multicast routing delivers it only to VTEPs that joined.

Question 32Data Center Network Architecture and Design

Two VRFs on the same leaf switch must exchange a small number of routes. What is the standard way to achieve this?

  1. Merge the two VRFs into a single routing table
  2. Disable VRF separation for those prefixes on the uplink
  3. Leak the routes through route targets between the VRFs Correct answer.
  4. Assign both VRFs the same route distinguisher

Explanation

VRF separation is maintained by importing and exporting routes with matching route targets, optionally through a shared transit VRF. Only the intended prefixes cross over, so the rest of each VRF remains isolated.

Question 33Data Center Network Architecture and Design

What is the main risk of relying on PFC to make a converged Ethernet fabric lossless for storage traffic?

  1. PFC resets the link whenever a pause frame is sent
  2. PFC disables ECN marking on the congested queue
  3. Pause can spread congestion and stall other traffic Correct answer.
  4. PFC limits the fabric to a single path per flow

Explanation

PFC works hop by hop, so a paused queue upstream can pause its upstream neighbour, propagating back-pressure along the path. If the cycle is not broken, this can cause head-of-line blocking and in the worst case a pause deadlock that affects unrelated traffic sharing the same buffers.

Question 34Data Center Network Architecture and Design

Why is a loopback interface configured on each VTEP in a VXLAN fabric?

  1. To terminate server VLANs carrying untagged traffic
  2. To provide a physical path for encapsulated frames
  3. To act as the gateway MAC for the anycast service
  4. To give a stable tunnel source and router ID Correct answer.

Explanation

The loopback address is always up regardless of which physical uplink is used, so it makes a stable source address for VXLAN tunnels and a stable BGP router ID. That keeps tunnels from flapping when individual links change state.

Question 35Data Center Network Architecture and Design

A VTEP must signal to the rest of the fabric that it wants to receive BUM traffic for a VNI. Which EVPN route type does it advertise?

  1. Type 3, the inclusive multicast Ethernet tag route Correct answer.
  2. Type 2, the MAC/IP advertisement route for each host
  3. Type 4, the Ethernet segment route used by multihoming
  4. Type 5, the IP prefix route for summarized subnets

Explanation

The Type 3 inclusive multicast Ethernet tag route identifies the VTEP as a member of the VNI's broadcast domain, so other VTEPs include it when replicating BUM traffic. It also carries the multicast flags used when the underlay provides multicast replication.

Question 36Data Center Network Architecture and Design

Why do leaf-spine fabrics normally hash each flow to a single uplink instead of balancing per packet?

  1. Reordering makes TCP retransmit and slows the flow Correct answer.
  2. Per-packet balancing cannot use equal-cost paths
  3. Per-packet balancing needs a routing protocol change
  4. Per-packet balancing disables the overlay tunnel

Explanation

Spreading one flow across parallel links with different delays delivers packets out of order, which TCP interprets as loss and answers with duplicate ACKs and retransmissions. Hashing on the flow keeps every packet of a conversation on one path, preserving order while still using all links across many flows.

Question 37Aruba CX Switching Platforms and AOS-CX Fundamentals

Which description best characterizes the software foundation of AOS-CX?

  1. A hypervisor that runs a separate virtual machine for each line card
  2. An embedded Linux environment with a centralized configuration database Correct answer.
  3. A Java runtime that loads forwarding behavior from an external server
  4. A proprietary real-time operating system with no filesystem or scripting support

Explanation

AOS-CX is built on an embedded Linux base and keeps the whole configuration in a structured, schema-driven database that the CLI, REST API and other agents all read and write. This shared data model is what enables programmatic access and features such as NAE and checkpoints.

Question 38Aruba CX Switching Platforms and AOS-CX Fundamentals

What does Virtual Switching Framework (VSF) provide on supported Aruba CX switches?

  1. Multiple physical switches acting as one logical switch with a single management IP Correct answer.
  2. A licensing mode that lets access points terminate directly on the switch
  3. Two modular chassis that share a common data plane while keeping independent control planes
  4. A Layer 3 overlay that stretches VLANs between separate data centers

Explanation

VSF merges the member switches into a single logical device with one control plane, one management IP address and a single configuration file, so the stack is managed as a unit. Conductor and standby roles are elected automatically among the members.

Question 39Aruba CX Switching Platforms and AOS-CX Fundamentals

A VSX pair requires which two components to be in place?

  1. A stacking cable plus a shared console server connection
  2. An interswitch link (ISL) plus a separate IP keepalive path Correct answer.
  3. A VSF member ID plus a conductor election priority
  4. Two out-of-band management ports configured in the same VLAN

Explanation

Every VSX pair needs an ISL that carries control-plane synchronization and, in most designs, data traffic, plus an independent IP keepalive path over which each peer confirms the other is alive. The keepalive is what lets the pair detect a split when the ISL fails.

Question 40Aruba CX Switching Platforms and AOS-CX Fundamentals

You are at the privileged EXEC prompt (switch#) on an AOS-CX switch. Which mode does the command 'configure terminal' enter?

  1. Global configuration mode, indicated by a (config)# prompt Correct answer.
  2. Interface configuration mode for the management port
  3. The boot menu used to select the system image
  4. A read-only monitoring mode with no configuration rights

Explanation

AOS-CX uses a hierarchical CLI: operator (>) and privileged EXEC (#) modes are for show and maintenance commands, and 'configure terminal' moves into global configuration, from which interface, routing and system settings are configured.

Question 41Aruba CX Switching Platforms and AOS-CX Fundamentals

What is the purpose of Zero Touch Provisioning (ZTP) on an AOS-CX switch?

  1. A backup mechanism that copies the running configuration to an attached USB drive
  2. A protocol that synchronizes MAC address tables between VSX peers
  3. A tool that discovers neighbors and draws a topology map of the network
  4. A factory-default switch obtains its image and configuration automatically at boot Correct answer.

Explanation

ZTP lets an unconfigured switch contact a provisioning service at first boot and pull down the correct firmware and full configuration, so a technician only has to cable and power the device. DHCP options typically point the switch to the provisioning server.

Question 42Aruba CX Switching Platforms and AOS-CX Fundamentals

You have finished changing the running configuration and want the changes to survive a reboot. Which command accomplishes this on AOS-CX?

  1. reload --force
  2. copy running-config startup-config Correct answer.
  3. show running-config
  4. copy startup-config running-config

Explanation

AOS-CX keeps unsaved changes only in the running configuration, which is held in RAM and lost on reload. Copying the running configuration to the startup configuration writes it to flash so it is applied at the next boot.

Question 43Aruba CX Switching Platforms and AOS-CX Fundamentals

Which Aruba CX model is a modular chassis switch rather than a fixed-port switch?

  1. Aruba CX 6200F
  2. Aruba CX 8400 Correct answer.
  3. Aruba CX 6300M
  4. Aruba CX 6100

Explanation

The Aruba CX 8400 is a modular chassis with supervisor and line-card slots, positioned for large campus aggregation and core roles. The 6100, 6200 and 6300 families are fixed-port switches that stack or pair using VSF or VSX.

Question 44Aruba CX Switching Platforms and AOS-CX Fundamentals

Which statement about the dedicated management (OOBM) port on an Aruba CX switch is correct?

  1. It provides an out-of-band management path separate from the data ports Correct answer.
  2. It is the stacking port used to build a VSF fabric between members
  3. It carries all user data traffic whenever no other uplink has been configured on the switch
  4. It must be placed in VLAN 1 before the switch will complete its boot

Explanation

The management port is a separate Ethernet interface with its own default gateway, used to reach the switch for CLI, REST and NAE access even when the data plane is misconfigured or down. It is not part of the forwarding path for user traffic.

Question 45Aruba CX Switching Platforms and AOS-CX Fundamentals

Which AOS-CX feature allows small Python agents to monitor conditions on the switch and raise alerts or run actions automatically?

  1. The REST API
  2. Unidirectional Link Detection
  3. The Network Analytics Engine (NAE) Correct answer.
  4. Zero Touch Provisioning

Explanation

NAE agents are Python scripts that run on the switch, sample time-series data from interfaces and system resources, and take action or generate alerts when a rule condition is met. Unlike a one-off script, an agent runs continuously and stores its collected data in the switch's time-series database.

Question 46Aruba CX Switching Platforms and AOS-CX Fundamentals

How is the AOS-CX REST API accessed?

  1. Through a Telnet session on TCP port 23
  2. Through TFTP transfers of the running configuration file
  3. Through SNMP SET operations against the MIB tree
  4. Through HTTPS requests to versioned /rest/ URIs using authenticated credentials Correct answer.

Explanation

The REST API is an HTTPS interface: a client authenticates, then issues GET, POST, PUT and DELETE requests to versioned paths such as /rest/v10.04/system. The same schema-driven configuration database backs the CLI, so everything the CLI can configure is reachable through the API.

Question 47Aruba CX Switching Platforms and AOS-CX Fundamentals

You want to capture the current running configuration so that you can compare against it or restore it later. Which AOS-CX feature should you use?

  1. An upload of the running configuration to a TFTP server
  2. A VLAN snapshot taken with the snapshot command
  3. A checkpoint created with the checkpoint command Correct answer.
  4. The archive log configured with an external syslog server

Explanation

A checkpoint stores a copy of the running configuration in flash that can later be compared with 'show checkpoint diff' or restored with 'checkpoint rollback'. It is the built-in mechanism for configuration versioning on AOS-CX.

Question 48Aruba CX Switching Platforms and AOS-CX Fundamentals

What does the VSX active-gateway feature provide to hosts attached to a VSX pair?

  1. Load sharing of the keepalive path across all member interfaces of the pair
  2. A single control plane that removes the need for an ISL between the two peers
  3. Automatic election of a conductor switch to run the VSF stack fabric
  4. A shared virtual MAC and IP for the default gateway so both peers forward Correct answer.

Explanation

With active-gateway, both VSX peers present one virtual IP and one virtual MAC to hosts and both actively route traffic for that gateway, so hosts keep a single default-gateway address while both switches forward. This avoids the idle standby behavior of a classic VRRP master/backup pair.

Question 49Aruba CX Switching Platforms and AOS-CX Fundamentals

Which statement describes a VSX LAG?

  1. A LAG whose member links are split across both VSX peers Correct answer.
  2. A LAG reserved for carrying VSX keepalive messages between peers
  3. A LAG whose member links all terminate on a single physical switch
  4. A LAG created automatically between the two ends of the ISL

Explanation

A VSX LAG (multichassis LAG) has member links on both peers but presents a single LAG to the downstream device, which can use standard 802.3ad hashing. If one peer is lost, the remaining links keep forwarding, so the attached device keeps its LAG up.

Question 50Aruba CX Switching Platforms and AOS-CX Fundamentals

What does configuring a VRF on an AOS-CX switch give you?

  1. A standby supervisor that takes over if the active one fails
  2. A dedicated hardware queue reserved for voice traffic
  3. A separate VLAN database that is bound to a single port
  4. An independent routing table with its own interfaces and routes Correct answer.

Explanation

A VRF is a separate Layer 3 routing instance: interfaces, static routes and routing protocol adjacencies are all bound to it, so overlapping address space can be reused in different VRFs on the same physical switch.