HPE7-A08
Free HPE7-A08 practice questions
10 questions from our HPE7-A08 practice tests, each with the correct answer and a written explanation. Read them with the answers shown, or switch to quiz mode and answer first.
Question 1 of 10, Create the implementation plan for the wired network solution
An engineer is designing a new campus with a CX 6400 VSX pair as the core and CX 6300 VSF stacks at the access layer. The customer wants the smallest possible spanning-tree domain and fast convergence when an access uplink fails. Where should the Layer 2/Layer 3 boundary be placed, and how should the access uplinks be built?
- ARoute at the core SVIs and connect each stack through a single uplink cable with a second cable in standby
- BRoute at the core SVIs with active-gateway and use one VSX LAG per stack, spread across members Correct answer
- CRoute at each access stack and run OSPF over individual point-to-point uplinks to each core switch
- DRoute at the core SVIs with VRRP and connect each stack through two separate LAGs blocked by MSTP
Answer and explanation: With routing at the VSX core using active-gateway and a VSX LAG spread across stack members, every uplink forwards, spanning tree has no blocked loops to converge, and member or link failures only remove a LAG member. Routing at the access layer is a valid design but moves the Layer 3 boundary away from the VSX core the customer chose and requires many more routed links and subnets. Two LAGs blocked by MSTP waste bandwidth and depend on spanning-tree convergence. A single active uplink is a single point of failure until a standby link is brought up.
Question 2 of 10, Create the implementation plan for the wired network solution
A university needs to extend the same Layer 2 segments for IoT devices across three buildings while routing all other traffic locally, keep each building in its own failure domain, and apply group-based policy between user roles. The customer manages the campus from HPE Aruba Networking Central. Which approach BEST fits?
- AA VSF stack that spans the three buildings so the IoT VLAN exists on one logical switch
- BA single MSTP region spanning all three buildings with the IoT VLAN trunked everywhere
- CA Central NetConductor EVPN-VXLAN fabric with role-based policy carried in the VXLAN header Correct answer
- DStatic VXLAN tunnels between building cores with ACLs applied on every access port
Answer and explanation: A NetConductor fabric uses EVPN-VXLAN to stretch only the needed segments over a routed underlay, so each building stays its own Layer 3 failure domain, and it carries group policy identifiers in the VXLAN header for role-to-role policy. A campus-wide MSTP region makes the whole campus one Layer 2 failure domain. Static VXLAN lacks a control plane and would need per-port ACLs for policy, which does not scale. A VSF stack across buildings is not a supported campus design and creates one large failure domain.
Question 3 of 10, Install and configure the wired network solution
In an EVPN-VXLAN fabric, leaf switches peer with two spines using iBGP. The spines do not host VTEPs. What must be configured on the spines for leaf routes to reach other leaves?
- AThe spines must run eBGP to each leaf with the same AS on all leaves and no other changes
- BThe spines must redistribute EVPN routes into OSPF so leaves learn them
- CThe spines must act as route reflectors with the leaves as clients in L2VPN EVPN Correct answer
- DThe spines must use next-hop-self toward the leaves for every EVPN route
Answer and explanation: iBGP does not re-advertise routes learned from one iBGP peer to another, so spines must be route reflectors with the leaves configured as route-reflector clients for the L2VPN EVPN family. EVPN routes are not redistributed into OSPF. Setting next-hop-self on spines would make the spines the VXLAN next hop, which breaks the tunnel endpoints. Using eBGP with the same AS on all leaves would cause routes to be rejected by AS-path loop prevention unless extra settings are used, and it is a different design from the stated iBGP fabric.
Question 4 of 10, Install and configure the wired network solution
A CX aggregation switch advertises 32 user subnets from area 10 into area 0. The core team wants a single summary 10.10.0.0/16 in the backbone and no change to routing inside area 10. Where and how should this be configured?
- AOn the core, with a distribute-list that filters the /24 routes inbound
- BOn the ABR, with an area 10 range 10.10.0.0/16 under router ospf Correct answer
- COn each access-layer router, with a summary-address for its own subnet
- DOn the ABR, with a static route to null for 10.10.0.0/16 redistributed as external
Answer and explanation: Inter-area summarization is done on the area border router with an area range, which replaces the individual type 3 LSAs with one summary in area 0 while keeping area 10 routing unchanged. Summary statements on internal routers do not summarize inter-area routes. Redistributing a null route creates an external route and does not suppress the specific type 3 LSAs. Filtering on the core hides routes instead of summarizing and does not reduce LSAs in the backbone.
Question 5 of 10, Install and configure the wired network solution
An engineer configures an uplink trunk on a CX 6300 that must carry VLANs 10, 20, and 30 and send untagged traffic in VLAN 99. Which configuration is correct?
- Avlan trunk native 99 and vlan trunk allowed 10,20,30,99 Correct answer
- Bvlan trunk native 10 and vlan trunk allowed 20,30,99
- Cvlan trunk allowed all and vlan trunk native tag 99
- Dvlan access 99 and vlan trunk allowed 10,20,30
Answer and explanation: The trunk must send untagged traffic in VLAN 99, so the native VLAN is 99, and the allowed list must include the three tagged VLANs plus the native VLAN. Setting native VLAN 10 sends VLAN 10 untagged and leaves VLAN 99 tagged, which reverses the requirement. vlan access is for access ports and cannot be combined with a trunk allowed list. Tagging the native VLAN and allowing every VLAN both contradict the requirement.
Question 6 of 10, Troubleshoot the wired network solution. Choose 2.
All 802.1X clients on a CX switch fail authentication. ClearPass Access Tracker shows no requests from that switch. Other switches work. Which causes are MOST likely? (Choose two.)
- AThe RADIUS server is configured on the switch in a VRF that has no route to ClearPass Correct answer
- BThe switch's local role names do not match the enforcement profiles returned by ClearPass
- CThe clients' supplicants are configured for PEAP rather than EAP-TLS on every laptop
- DThe port-access client limit on the switch is lower than the number of users
- EThe switch is not added to ClearPass as a network device with a matching shared key Correct answer
Answer and explanation: If ClearPass logs nothing from the switch, the requests are not arriving or are being dropped silently: an unknown network device or wrong key causes ClearPass to discard requests, and a RADIUS configuration in the wrong VRF prevents them from being sent at all. EAP method or role mismatches would appear as failed or wrongly enforced requests in Access Tracker. A client limit affects only the clients beyond the limit.
Question 7 of 10, Troubleshoot the wired network solution
A BGP session between a CX border switch and a peer stays in Active state. The peer is configured to use its loopback as the source, and the CX switch points to that loopback address. Ping from the CX switch to the loopback succeeds. What is the MOST likely missing setting on the CX switch?
- AA route-map permitting all prefixes inbound from the BGP neighbor
- BA higher local preference for routes received from that BGP neighbor
- Cupdate-source loopback on the CX neighbor, plus ebgp-multihop if it is eBGP Correct answer
- DThe neighbor's address advertised into BGP with a network statement
Answer and explanation: When peering between loopbacks, both sides must source the session from their loopback, and eBGP also needs multihop because the peer is not directly connected; otherwise the TCP session is rejected and the state cycles between Active and Connect. Local preference and inbound route-maps apply after the session is established. Advertising the neighbor's address is not required for the session to come up.
Question 8 of 10, Troubleshoot the wired network solution
After a contractor connects a small switch to a wall jack, the access stack logs MAC address moves between ports 1/1/10 and 1/1/24 and CPU usage spikes. Spanning tree is enabled. What is the MOST likely cause and fix?
- AAn LACP hash imbalance on the uplink; change the LAG hash to l4-src-dst
- BAn asymmetric route from the VSX core; enable active forwarding on all core SVIs
- CA duplicate IP address on the network; enable ARP inspection on all VLANs immediately
- DA loop through the contractor switch; enable BPDU guard or loop protect on edge ports Correct answer
Answer and explanation: MAC flapping between two access ports combined with high CPU is a classic Layer 2 loop, likely through a device that does not pass BPDUs; BPDU guard or loop protect on edge ports detects and disables the looped port. Duplicate IPs cause ARP conflicts, not MAC flaps between edge ports. Asymmetric routing at the core does not create MAC moves on access ports. LAG hashing does not cause MAC flapping.
Question 9 of 10, Manage, maintain, optimize, and monitor the wired network solution
The NAE dashboard on a core switch shows the routing table size growing by thousands of routes over a few minutes after a WAN change. What should the engineer do FIRST?
- AReboot the core to clear the routing table and start over
- BRaise the hardware route table limit with a system capacity command
- CDisable NAE agents to reduce the CPU impact of the alert
- DCheck the BGP peers for a leak of full or unexpected routes Correct answer
Answer and explanation: A sudden route surge after a WAN change usually means a peer is leaking routes or a filter was removed; checking the BGP peers and inbound policies identifies it, and a maximum-prefix limit can protect the switch. Capacity limits are hardware-defined. Disabling NAE hides the alert. Rebooting relearns the same leaked routes.
Question 10 of 10, Manage, maintain, optimize, and monitor the wired network solution. Choose 2.
A team wants to detect unauthorized configuration changes on CX switches within minutes and know who made them. Which combination is MOST effective? (Choose two.)
- ADisabling checkpoints so changes are visible in startup-config
- BDaily reboots that reload the approved startup configuration
- CSNMPv1 polling of interface counters every minute
- DSyslog of configuration change events to a central collector Correct answer
- ETACACS+ command accounting to record each user's commands Correct answer
Answer and explanation: TACACS+ command accounting records which administrator ran which commands, and forwarding configuration change events to a syslog collector provides near-real-time alerts. Disabling checkpoints removes history. Reboots disrupt service and do not show who made changes. Interface counter polling does not track configuration.