Huawei H12-891 HCIE-Datacom Practice Questions with Explanations

Free Huawei H12-891 HCIE-Datacom practice questions. 122 of them, each with the correct answer, a full explanation, and the reason every other option is wrong.

H12-891 practice questions

Question 1

For telemetry data push, data can be transferred between devices and collectors using the TLS protocol. TLS must be configured for data encryption if data is pushed based on gRPC.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

Telemetry data pushed over gRPC can be encrypted with TLS, and TLS has to be configured for that encryption to happen, so the statement is true.

Why the other options are wrong:

  • B. The statement is true, so False is wrong.

Question 2

SSH must be configured before you can enable NETCONF on Huawei network devices.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

NETCONF runs over SSH on Huawei devices, so SSH has to be configured before NETCONF can be enabled.

Why the other options are wrong:

  • B. The statement is true, so False is wrong.

Question 3

SR is a technology that forwards data packets based on source IP addresses.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

Segment Routing is a source routing technique: the ingress node encodes an ordered list of segments in the packet. It does not forward by source IP address.

Why the other options are wrong:

  • A. SR forwards using a list of segments, not the source IP address, so True is wrong.

Question 4

Segment Routing (SR) is a technical architecture designed to forward data packets on a network based on the source routing paradigm. SR-MPLS, as its name suggests, is SR based on MPLS label forwarding.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

Segment Routing is built on the source routing paradigm, and SR-MPLS is the variant that carries segments as MPLS labels, so both parts of the statement are correct.

Why the other options are wrong:

  • B. Both parts of the statement are correct, so False is wrong.

Question 5

In 6PE and 6VPE networking, no VPN instance needs to be created on PEs.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

6PE carries IPv6 routes in the public IPv6 routing table, so it needs no VPN instance, but 6VPE isolates each IPv6 VPN in its own VPN instance on the PEs. The statement is false because it also covers 6VPE.

Why the other options are wrong:

  • A. 6VPE requires a VPN instance with the IPv6 address family on each PE, so True is wrong.

Question 6

The display current-configuration command displays the running configurations in memory.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

The display current-configuration command shows the configuration that is currently running in the device memory, which can differ from the saved configuration file used at the next startup.

Why the other options are wrong:

  • B. The statement is true, so False is wrong.

Question 7

In the firewall hot standby scenario, heartbeat interfaces can be connected directly or through an intermediate device such as a switch or router.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

The heartbeat interfaces of two firewalls in hot standby can be connected directly, or through an intermediate device such as a switch or router, as long as the heartbeat packets can reach the peer.

Why the other options are wrong:

  • B. The statement is true, so False is wrong.

Question 8

In DU label advertisement mode, if the liberal label retention mode is used, the device reserves labels received from all the LDP peers regardless of whether the LDP peer is the optimal next hop for reaching the destination network.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

In liberal label retention mode, an LSR keeps every label mapping it receives from its LDP peers, whether or not the peer is the optimal next hop to the destination. This speeds up convergence when the route changes.

Why the other options are wrong:

  • B. The statement is true, so False is wrong.

Question 9

In Huawei SD-WAN Solution, after a CPE goes online, it automatically selects an RR without the help of iMaster NCE-WAN.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

A CPE does not pick its RR by itself. iMaster NCE-WAN assigns the RR to the CPE and delivers the RR information once the CPE is online, and the CPE then sets up its control channel with that RR.

Why the other options are wrong:

  • A. The RR is assigned by iMaster NCE-WAN, not selected independently by the CPE, so True is wrong.

Question 10

On a CloudCampus virtualized campus network, virtual networks are defined based on services. Each virtual network corresponds to a service, and isolation needs to be manually configured between virtual networks.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

Each virtual network on a CloudCampus virtualized campus maps to its own VPN instance, so virtual networks are isolated from one another by default. No manual isolation configuration is needed.

Why the other options are wrong:

  • A. Virtual networks are isolated by default, so manual isolation is not required and True is wrong.

Question 11

On a virtualized campus network, all devices in a VXLAN domain must support VXLAN.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

Only the devices that act as VXLAN tunnel endpoints need to support VXLAN. Devices between them forward the encapsulated packets as ordinary IP packets on the underlay network.

Why the other options are wrong:

  • A. Intermediate devices only need IP forwarding, so not every device must support VXLAN and True is wrong.

Question 12

VXLAN identifies tenants using VNIs, which are 24 bits long. A tenant can have one or more VNIs, and VXLAN supports a maximum of 12 million tenants.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

A 24-bit VNI gives about 16 million values (2 to the power of 24 is 16,777,216), so VXLAN supports up to about 16 million tenants, not 12 million.

Why the other options are wrong:

  • A. A 24-bit VNI supports about 16 million tenants, not 12 million, so True is wrong.

Question 13

NETCONF provides a set of mechanisms for managing network devices.

What layers can NETCONF be conceptually partitioned into?

  • A. Secure transport layer, messages layer, operations layer, and content layer
  • B. Secure transport layer, messages layer, configuration layer, and content layer
  • C. Encrypted transport layer, messages layer, configuration layer, and content layer
  • D. Encrypted transport layer, messages layer, operations layer, and content layer
Show answer and explanation ▾

Correct answer: A

NETCONF is divided into four layers: the secure transport layer, the messages layer (RPC framing), the operations layer (such as get-config and edit-config) and the content layer (the configuration and state data).

Why the other options are wrong:

  • B. NETCONF has an operations layer, not a configuration layer.
  • C. The lowest layer is called the secure transport layer, and there is no configuration layer.
  • D. The lowest layer is called the secure transport layer, not the encrypted transport layer.

Question 14

Huawei Open Programmability System (OPS) provides openness and programmability capabilities for network devices and enables users to perform secondary development to implement customized functions.

Which of the following commands can be used to view information about the OPS script installed on a network device?

  • A. display ops script
  • B. display ops file
  • C. check ops script
  • D. check ops file
Show answer and explanation ▾

Correct answer: A

The display ops script command shows information about the OPS scripts installed on the device.

Why the other options are wrong:

  • B. display ops file is not the command that shows installed OPS scripts.
  • C. check ops script is not a command for viewing installed OPS scripts.
  • D. check ops file is not a command for viewing installed OPS scripts.

Question 15

In the SSH algorithm negotiation phase, the SSH server and client exchange the list of algorithms that hey support. The list of symmetric encryption algorithms of the SSH server is as follows:

aes256-ctr, aes192-ctr, aes128-ctr, and aes256-cbc.

The list of symmetric encryption algorithms of the SSH client is as follows:

aes128-ctr, aes192-ctr, aes256-ctr, and aes256-cbc.

Which of the following symmetric encryption algorithms can be used by both the SSH server and client?

  • A. aes256-ctr
  • B. aes192-ctr
  • C. aes128-ctr
  • D. aes256-cbc
Show answer and explanation ▾

Correct answer: C

During SSH negotiation the chosen algorithm is the first one in the client's list that the server also supports. The client lists aes128-ctr first and the server supports it, so aes128-ctr is used.

Why the other options are wrong:

  • A. aes256-ctr is first in the server's list, but selection follows the client's order, where it comes third.
  • B. aes192-ctr comes after aes128-ctr in the client's list, so it is not selected.
  • D. aes256-cbc is last in both lists, so it is not selected.

Question 16

YANG is a data modeling language.

Which of the following statements is incorrect about YANG?

  • A. The YANG model is presented as a .yang file.
  • B. The YANG model can be converted to an XML syntax-based YANG Independent Notation (YIN) model without any loss.
  • C. YANG is a modular language representing data structures in a hierarchical tree format.
  • D. YANG only supports built-in data type, while extended data types are not allowed.
Show answer and explanation ▾

Correct answer: D

YANG has built-in data types and also lets modelers define their own derived types with the typedef statement, so saying that only built-in types are supported is incorrect.

Why the other options are wrong:

  • A. This is accurate: a YANG model is stored as a .yang file.
  • B. This is accurate: a YANG module can be converted to the XML-based YIN format without loss.
  • C. This is accurate: YANG is modular and describes data as a hierarchical tree.

Question 17

Which of the following types of LSAs does OSPF use to support SR-MPLS?

  • A. Type10 Opaque LSA
  • B. Type1 Router LSA
  • C. Type2 Network LSA
  • D. Type7 NSSA External LSA
Show answer and explanation ▾

Correct answer: A

OSPF carries Segment Routing information such as the SRGB, prefix SIDs and adjacency SIDs in Type 10 Opaque LSAs, which are flooded within an area.

Why the other options are wrong:

  • B. Router LSAs describe a router's links and are not extended to carry SR information.
  • C. Network LSAs describe the routers on a broadcast segment and do not carry SR information.
  • D. Type 7 LSAs carry external routes in an NSSA and do not carry SR information.

Question 18

Which of the following rules does not apply to SR-MPLS Policy design?

  • A. Associate service traffic with tunnels to achieve some degree of path visualization.
  • B. Ensure that the traffic on each tunnel is not too heavy. Otherwise, bandwidth optimization will be difficult.
  • C. Deploy as many tunnels as possible, as more tunnels result in more refined service isolation and service quality assurance.
  • D. Consider possible network expansion in the future.
Show answer and explanation ▾

Correct answer: C

Tunnel design aims for a manageable number of tunnels. Deploying as many tunnels as possible increases device load and makes path computation and maintenance harder, so it is not a design rule.

Why the other options are wrong:

  • A. This is a valid design rule: mapping services to tunnels makes paths visible.
  • B. This is a valid design rule: lightly loaded tunnels are easier to optimize.
  • D. This is a valid design rule: the design should leave room for future expansion.

Question 19

A VPDN is a VPN constructed on a public network.

Which of the following is a VPDN?

  • A. L2TP
  • B. VPLS
  • C. GRE VPN
  • D. IPsec
Show answer and explanation ▾

Correct answer: A

A virtual private dial-up network gives dial-in users access to a private network over a public network, and L2TP is the typical VPDN tunneling protocol because it tunnels PPP sessions.

Why the other options are wrong:

  • B. VPLS is a Layer 2 VPN that emulates a LAN over an MPLS network, not a dial-up VPN.
  • C. GRE is a general tunneling protocol for site-to-site links, not a dial-up VPN.
  • D. IPsec protects IP packets with encryption and authentication and is not a dial-up tunneling protocol.

Question 20

Which of the following BGP route attributes cannot be used to control BGP route selection?

  • A. Community
  • B. AS_Path
  • C. MED
  • D. Originator ID
Show answer and explanation ▾

Correct answer: A

This attribute only tags routes so that routing policies can match them. None of the BGP route selection rules compares it, so on its own it cannot decide which route is preferred.

Why the other options are wrong:

  • B. AS_Path length is compared directly during route selection, and a shorter path is preferred.
  • C. MED is compared directly during route selection, and a lower value is preferred.
  • D. Originator_ID is used in route selection in place of the router ID for reflected routes.

Question 21

Which of the following fields is added to BGP4+ Update packets to advertise IPv6 routes?

  • A. A new NLRI type: IPv6_NLRI, which carries the IPv6 NLRI prefix length and IPv6 NLRI prefix.
  • B. A new path attribute: MP_REACH_NLRI
  • C. A new path attribute: IPv6_REACH_NLRI
  • D. A new path attribute: MP_UNREACH_NLRI
Show answer and explanation ▾

Correct answer: B

BGP4+ advertises IPv6 routes with the MP_REACH_NLRI path attribute, which carries the address family, the next hop and the reachable IPv6 prefixes.

Why the other options are wrong:

  • A. IPv6 prefixes are carried inside MP_REACH_NLRI, not in a new NLRI type called IPv6_NLRI.
  • C. There is no path attribute named IPv6_REACH_NLRI.
  • D. MP_UNREACH_NLRI is used to withdraw routes, not to advertise them.

Question 22

In BGP4+, what is the content of the next hop network address field carried in the MP_REACH_NLRI attributed of an Update packet?

  • A. It must be a link-local address.
  • B. It must be a global unicast address.
  • C. It can be only a link-local address.
  • D. The link-local address and global unicast address can be carried at the same time.
Show answer and explanation ▾

Correct answer: D

The next hop field in MP_REACH_NLRI is 16 bytes when it holds a global unicast address, or 32 bytes when it holds a global unicast address followed by a link-local address, so both can be carried together.

Why the other options are wrong:

  • A. The field can carry a global unicast address, so it does not have to be link-local.
  • B. The field can also carry a link-local address along with the global address.
  • C. The field is not limited to a link-local address.

Question 23

Which of the following operations is not involved in the middle phase of a migration?

  • A. Service test
  • B. Migration preparation
  • C. On-site monitoring
  • D. Migration implementation
Show answer and explanation ▾

Correct answer: C

The middle phase of a migration covers migration preparation, migration implementation and the service test. On-site monitoring takes place in the late phase, after the cutover has been completed.

Why the other options are wrong:

  • A. The service test is part of the middle phase.
  • B. Migration preparation is part of the middle phase.
  • D. Migration implementation is part of the middle phase.

Question 24

Which of the following statements is incorrect about a BGP/MPLS IP VPN?

  • A. A BGP/MPLS IP VPN consists of customer edges (CEs), provider edges (PEs), and providers (Ps). PEs and Ps are carrier devices, whereas CEs are BGP/MPLS IP VPN customer devices.
  • B. Inter-site communication can be implemented through VPN. A site can only belong to only one VPN.
  • C. In most cases, CEs are unaware of VPNs and do not need to support MPLS or MP-BGP.
  • D. Ps only need to provide basic MPLS forwarding capabilities and do not need to maintain VPN-related information.
Show answer and explanation ▾

Correct answer: B

A site is not limited to one VPN. By importing and exporting several VPN targets, a site can belong to multiple VPNs at the same time, so option B is the incorrect statement.

Why the other options are wrong:

  • A. This is accurate: the model is made of CEs, PEs and Ps, with PEs and Ps in the carrier network.
  • C. This is accurate: CEs normally run only ordinary routing and are unaware of the VPN.
  • D. This is accurate: P devices only forward labeled packets and hold no VPN routes.

Question 25

Which of the following security protection actions is not used to safeguard MAC addresses?

  • A. Restrict
  • B. Protect
  • C. Re-mark
  • D. Shutdown
Show answer and explanation ▾

Correct answer: C

Port security has three protection actions: restrict, protect and shutdown. Re-mark is a QoS action that changes packet priority and has nothing to do with protecting MAC addresses.

Why the other options are wrong:

  • A. Restrict is a port security action that discards the offending packets and reports an alarm.
  • B. Protect is a port security action that discards the offending packets.
  • D. Shutdown is a port security action that sets the interface to error-down.

Question 26

What is the largest number of CPEs that a site can support in the Huawei SD-WAN Solution?

  • A. 1
  • B. 2
  • C. 3
  • D. 4
Show answer and explanation ▾

Correct answer: B

A site in the Huawei SD-WAN Solution has either one CPE (single-gateway site) or two CPEs (dual-gateway site), so the largest number is two.

Why the other options are wrong:

  • A. A site can also be a dual-gateway site with two CPEs.
  • C. A site supports at most two CPEs, not three.
  • D. A site supports at most two CPEs, not four.

Question 27

Which of the following Internet access modes can be used to implement centralized security control over Internet access traffic?

  • A. Local Internet access
  • B. Centralized Internet access
  • C. Hybrid Internet access
  • D. Priority-based Internet access
Show answer and explanation ▾

Correct answer: B

With centralized Internet access, all sites send their Internet traffic through a central site, so security policies for that traffic can be applied and maintained in one place.

Why the other options are wrong:

  • A. Local Internet access sends traffic out at each site, so control is spread across sites.
  • C. Hybrid Internet access lets part of the traffic break out locally, so control is not fully centralized.
  • D. Priority-based Internet access is not the mode used for centralized security control.

Question 28

Assume that a company's headquarters network and branch networks are connected through the Internet.

Which of the following VPN technologies is most capable of ensuring secure and reliable data transmission between the headquarters and branches?

  • A. SSL VPN
  • B. IPsec VPN
  • C. L2TP VPN
  • D. MPLS VPN
Show answer and explanation ▾

Correct answer: B

IPsec VPN builds site-to-site tunnels over the Internet and protects the traffic with encryption, integrity checks and peer authentication, which makes it the best fit for connecting headquarters and branches securely.

Why the other options are wrong:

  • A. SSL VPN is designed for remote users reaching internal applications, not for linking whole sites.
  • C. L2TP alone does not encrypt the data it tunnels.
  • D. MPLS VPN runs on a carrier's MPLS backbone rather than the Internet and does not encrypt traffic.

Question 29

In VXLAN, which of the following is similar to a VLAN ID on a traditional network and identifies a VXLAN segment?

  • A. VRF
  • B. VNI
  • C. BD
  • D. VTEP
Show answer and explanation ▾

Correct answer: B

The VXLAN network identifier (VNI) identifies a VXLAN segment in the same way a VLAN ID identifies a VLAN. It is 24 bits long, so it supports far more segments.

Why the other options are wrong:

  • A. A VRF is a Layer 3 routing instance, not a segment identifier.
  • C. A bridge domain is the Layer 2 broadcast domain that a VNI is mapped to on the device.
  • D. A VTEP is the tunnel endpoint that encapsulates and decapsulates VXLAN packets.

Question 30

Which of the following is used to implement inter-subnet communication on a VXLAN network and allow access from a VXLAN network to an external non-VXLAN network?

  • A. VLAN IF interface
  • B. NVE interface
  • C. Layer 2 VXLAN gateway
  • D. Layer 3 VXLAN gateway
Show answer and explanation ▾

Correct answer: D

A Layer 3 VXLAN gateway routes between different VXLAN subnets and between the VXLAN network and external non-VXLAN networks.

Why the other options are wrong:

  • A. A VLANIF interface is the gateway of a VLAN on a traditional network, not of a VXLAN segment.
  • B. An NVE interface is the logical interface that sets up VXLAN tunnels, not a gateway.
  • C. A Layer 2 VXLAN gateway connects users to the VXLAN network and handles communication within one subnet only.

Question 31

Which of the following statements is incorrect about configuring fabric access management on iMaster NCE-Campus?

  • A. Create a server template and then an authentication template.
  • B. Only one authentication mode can be specified in an authentication template.
  • C. iMaster NCE-Campus can function as a RADIUS server or Portal server.
  • D. Authentication templates need to be bound to wired and wireless access points.
Show answer and explanation ▾

Correct answer: B

An authentication template is not limited to one mode. It can combine several authentication modes, such as 802.1X, MAC address and Portal authentication, so option B is the incorrect statement.

Why the other options are wrong:

  • A. This is accurate: the server template is created first and then referenced by the authentication template.
  • C. This is accurate: iMaster NCE-Campus can act as the RADIUS server or the Portal server.
  • D. This is accurate: authentication templates are bound to the wired and wireless access points.

Question 32

In a scenario where a VXLAN tunnel is dynamically established through BGP EVPN, which of the following BGP EVPN routes is used to advertise MAC addresses, ARP entries, and IRB routes?

  • A. Type 1
  • B. Type 2
  • C. Type 3
  • D. Type 4
Show answer and explanation ▾

Correct answer: B

The EVPN Type 2 route is the MAC/IP advertisement route. It advertises host MAC addresses, host ARP entries and integrated routing and bridging (IRB) host routes between VTEPs.

Why the other options are wrong:

  • A. Type 1 is the Ethernet auto-discovery route and does not advertise host addresses.
  • C. Type 3 is the inclusive multicast route, used to discover VTEPs and build the BUM forwarding list.
  • D. Type 4 is the Ethernet segment route, used for designated forwarder election.

Question 33

When an NVE encapsulates a data frame into a VXLAN packet, information of which of the following protocols is encapsulated between the outer IP header and the VXLAN header?

  • A. TCP
  • B. UDP
  • C. IP
  • D. Ethernet
Show answer and explanation ▾

Correct answer: B

VXLAN uses MAC-in-UDP encapsulation. The original frame gets a VXLAN header, then a UDP header, then the outer IP header, so the UDP header sits between the outer IP header and the VXLAN header.

Why the other options are wrong:

  • A. VXLAN is carried over UDP, not TCP.
  • C. No second IP header is placed between the outer IP header and the VXLAN header.
  • D. The Ethernet headers are outside the outer IP header and inside the VXLAN header, not between them.

Question 34

Huawei Open Programmability System (OPS) provides openness and programmability capabilities for network devices and enables users to perform secondary development. To facilitate script compilation, Huawei OPS provides the Python script template.

Which of the following statements about the Python script template are correct? (Choose three.)

  • A. The OPSConnection class is used to invoke RESTful APIs. This class defines methods for establishing HTTP connections. You do not need to modify the methods.
  • B. The initialization method of the OPSConnection class is def__init__(self, host, port=80). When the OPSConnection class is initialized in the main() function, the host parameter needs to be set to the IP address of the management interface of the network device to be managed.
  • C. Users need to search for the corresponding URI and the XML content of the HTTP packet body in the RESTful API Reference based on the managed object to be accessed, and modify the Python script template accordingly.
  • D. The OPSConnection class has four methods: create, delete, set, and get, which encapsulate the PUT, DELETE, POST and GET methods of HTTP.
Show answer and explanation ▾

Correct answer: A, B, C

The OPSConnection class wraps the HTTP connection and RESTful calls and is used as supplied. Its constructor takes the device address with port 80 as the default, and the user adapts the template with the URI and XML body taken from the RESTful API reference.

Why the other options are wrong:

  • D. In the template, create sends POST and set sends PUT, so the mapping listed in this option is not right.

Question 35

Which of the following are key technologies used to implement SRv6 Policies based on Huawei’s CloudWAN solution architecture?

  • A. BGP-LS
  • B. BGP IPv6 SR Policy
  • C. PCEP
  • D. NETCONF
Show answer and explanation ▾

Correct answer: B

In the CloudWAN architecture the controller computes SRv6 Policy paths and delivers them to the headend routers through BGP IPv6 SR Policy, which makes it the technology that implements SRv6 Policies.

Why the other options are wrong:

  • A. BGP-LS reports topology and link information to the controller. It does not deliver SRv6 Policies.
  • C. PCEP is not the protocol used to deliver SRv6 Policies in this architecture.
  • D. NETCONF delivers device and service configuration, not the SRv6 Policy paths computed by the controller.

Question 36

Which of the following statements regarding MPLS label encapsulation format are correct? (Choose three.)

  • A. The total length of one layer of MPLS label is 4 bytes.
  • B. The TTL field in an MPLS label has the similar meaning as that in an IP packet, and can prevent loops.
  • C. The S field in an MPLS label has 1 bit and is used to identify whether the label is the bottom-of-stack label. If the value is 1, the label is a penultimate-layer label.
  • D. For Ethernet and PPP packets, a label stack is located between the Layer 2 packet header and data. If a VLAN tag is available, the label stack is placed between Ethernet header and the VLAN tag.
Show answer and explanation ▾

Correct answer: A, B, C

One MPLS label entry is 4 bytes: a 20-bit label, a 3-bit EXP field, a 1-bit S field and an 8-bit TTL. The TTL works like the IP TTL to stop loops, and the 1-bit S field marks the bottom of the label stack.

Why the other options are wrong:

  • D. The label stack follows the entire Layer 2 header, so with a VLAN tag it comes after the tag, not between the Ethernet header and the tag.

Question 37

Which of the following types of authorization information can be delivered by a RADIUS server after a client passes 802.1 X authentication?

(Choose three.)

  • A. MAC address
  • B. ACL
  • C. VLAN ID
  • D. UCL group
Show answer and explanation ▾

Correct answer: B, C, D

After a user passes 802.1X authentication, the RADIUS server can deliver authorization attributes including an ACL, a VLAN ID and a UCL group. The access device applies them to control what the user can reach.

Why the other options are wrong:

  • A. A MAC address identifies the terminal. It is not an authorization attribute delivered by the server.

Question 38

Which of the following deployment modes are recommended for LAN-side APs on small and midsize campus networks based on the Huawei CloudCampus Solution? (Choose two.)

  • A. Deployment through the registration query
  • B. Email-based deployment
  • C. DHCP option-based deployment
  • D. CLI-based deployment
Show answer and explanation ▾

Correct answer: A, C

LAN-side APs get their addresses from DHCP inside the campus, so they can learn the controller address from a DHCP option or by querying the registration center. Both give plug-and-play onboarding without on-site configuration.

Why the other options are wrong:

  • B. Email-based deployment is intended for egress devices such as AR routers, not for LAN-side APs.
  • D. CLI-based deployment requires manual configuration of every AP, so it is not a recommended mode.

Question 39

MPLS supports forwarding equivalence class (FEC).

Which of the following cannot be used as a FEC allocation standard?

  • A. Fragment offset
  • B. Destination address
  • C. Application protocol
  • D. Class of Service (CoS)
Show answer and explanation ▾

Correct answer: A

A FEC groups packets that are forwarded in the same way, and it can be defined by criteria such as destination address, application protocol or class of service. The fragment offset only describes IP fragmentation and is not a FEC criterion.

Why the other options are wrong:

  • B. Destination address is the most common basis for defining a FEC.
  • C. Application protocol can be used as a basis for defining a FEC.
  • D. Class of service can be used as a basis for defining a FEC.

Question 40

Which of the following statements is incorrect?

  • A. LSPs are classified into static and dynamic ISPs. A static LSP is manually configured by an administrator. A dynamic LSP is dynamically established using routing and label distribution protocols.
  • B. When labels are allocated manually, ensure that the outgoing label of an upstream node is the same as the incoming label of the downstream node.
  • C. LSPs created by static Distribution labels can be dynamically adjusted according to network topology changes, without the need for administrator intervention.
  • D. Dynamic LSPs are established using label distribution protocols such as MP-BGP RSVP-TE and LDP.
Show answer and explanation ▾

Correct answer: C

A static LSP is built from manually assigned labels and cannot react to topology changes. An administrator has to change it by hand, so the claim that it adjusts dynamically is incorrect.

Why the other options are wrong:

  • A. This is accurate: LSPs are either static, set manually, or dynamic, set up by protocols.
  • B. This is accurate: the outgoing label of the upstream node must equal the incoming label of the downstream node.
  • D. This is accurate: MP-BGP, RSVP-TE and LDP all distribute labels for dynamic LSPs.

Question 41

Which of the following statements regarding MPLS label space is incorrect?

  • A. The label space 16-1023 is shared by static LSPs and static CR-LSPs.
  • B. The label space 1024 and higher is shared by dynamic signaling protocols, such as LDP, RSVPTE and MP-BGP.
  • C. When the last hop receives packets with the label value 0, it directly drops the packets.
  • D. When the penultimate-hop LSR performs label switching, if it finds that the label value after the switching is 3, then it pops out the label and sends the packet to the last hop.
Show answer and explanation ▾

Correct answer: C

Label 0 is the IPv4 explicit null label. The last hop that receives it pops the label and forwards the packet based on its IPv4 header, so the packet is not dropped.

Why the other options are wrong:

  • A. This is accurate: labels 16 to 1023 are shared by static LSPs and static CR-LSPs.
  • B. This is accurate: labels from 1024 upward are used by dynamic protocols such as LDP, RSVP-TE and MP-BGP.
  • D. This is accurate: label 3 is the implicit null label, which tells the penultimate hop to pop the label.

Question 42

What is the function of the isis ipv6 cost 50 command on an interface?

  • A. Sets the IS-ISI Pv6 Level-1 cost of the interface to 50.
  • B. Sets the IS- IS Pv6 Level-2 cost of the interface to 50.
  • C. Sets both the IS-IS IPv6 Level-1 and Level-2 costs of the interface to 50.
  • D. Sets both the IS-IS IPv6 Level-1 and Level-2 costs of the interface to 60.
Show answer and explanation ▾

Correct answer: C

When the isis ipv6 cost command is entered without a level, it applies to both Level-1 and Level-2, so both IPv6 costs of the interface are set to 50.

Why the other options are wrong:

  • A. With no level specified, the command is not limited to Level-1.
  • B. With no level specified, the command is not limited to Level-2.
  • D. The command sets the cost to 50, not 60.

Question 43

Which of the following types of routes is used by EVPN to implement fast convergence and split horizon?

  • A. Ethernet A-D route
  • B. MAC/IP advertisement route
  • C. Inclusive multicast route
  • D. Ethernet segment route
Show answer and explanation ▾

Correct answer: A

Ethernet auto-discovery (A-D) routes let a PE withdraw reachability for a whole Ethernet segment at once, which gives fast convergence, and they carry the ESI label that is used for split horizon.

Why the other options are wrong:

  • B. MAC/IP advertisement routes advertise host MAC and IP addresses.
  • C. Inclusive multicast routes are used to set up the forwarding of broadcast, unknown unicast and multicast traffic.
  • D. Ethernet segment routes are used to discover PEs on the same segment and elect the designated forwarder.

Question 44

Which of the following roles is not a core role in Huawei's free mobility solution?

  • A. Authentication device
  • B. Policy enforcement device
  • C. Policy control device
  • D. iMaster NCE-Campus
Show answer and explanation ▾

Correct answer: D

Free mobility is described with three roles: the authentication device, the policy enforcement device and the policy control device. iMaster NCE-Campus is the product that fills the policy control role, not a separate role.

Why the other options are wrong:

  • A. The authentication device is one of the three roles in free mobility.
  • B. The policy enforcement device is one of the three roles in free mobility.
  • C. The policy control device is one of the three roles in free mobility.

Question 45

Huawei CloudCampus Solution provides location-based services (LBS), obtains Wi-Fi-based terminal location data, and reports the data to developers' LBS applications. In this case, terminal locations within a certain range can be calculated. Which of the following statement is incorrect about the interactions between the LBS application and iMaster NCE?

  • A. The LBS application generates a verification code. You need to manually copy the verification code to iMaster NCE.
  • B. iMaster NCE sends a response code verification request to the LBS application and the LBS application responds to the request.
  • C. After the verification code is verified, Master NCE sends terminal location data to the LBS application.
  • D. The LBS application parses the received terminal location data.
Show answer and explanation ▾

Correct answer: A

The verification code is not generated by the LBS application for an administrator to copy into iMaster NCE, so this description of the first step is the incorrect one. The other three statements follow the flow: a verification request and response, then location data, then parsing by the application.

Why the other options are wrong:

  • B. This is accurate: iMaster NCE sends the verification request and the LBS application answers it.
  • C. This is accurate: location data is sent only after verification succeeds.
  • D. This is accurate: the LBS application parses the location data it receives.

Question 46

EVPN supports multiple service modes.

Which of the following service modes allows an interface to be used only by a single user?

  • A. Port-based mode
  • B. VLAN-based mode
  • C. VLAN bundle mode
  • D. VLAN-aware bundle mode
Show answer and explanation ▾

Correct answer: A

In port-based mode the whole physical interface is bound to a single EVPN instance, so the interface serves one user only and cannot be shared among users by VLAN.

Why the other options are wrong:

  • B. VLAN-based mode binds individual VLANs to EVPN instances, so one interface can serve several users.
  • C. VLAN bundle mode maps several VLANs to one EVPN instance and does not dedicate the interface to one user.
  • D. VLAN-aware bundle mode carries several VLANs in one EVPN instance and does not dedicate the interface to one user.

Question 47

During the migration test, which of the following indicates that the test is completed?

  • A. The network is running properly.
  • B. Tested network services are running property.
  • C. Tested upper-layer application services are running properly on the network.
  • D. Command outputs are correct.
Show answer and explanation ▾

Correct answer: C

A migration exists to carry the business, so the test is complete only when the upper-layer application services that run over the network are verified to work properly. Network and service checks come first, and the application check closes the test.

Why the other options are wrong:

  • A. A properly running network is an early check and does not prove that applications work.
  • B. Network services running properly still does not confirm that the applications above them work.
  • D. Correct command outputs only show device state and do not confirm service or application health.

Question 48

Which of the following is the default port used by iMaster NCE-Campus northbound open APIs?

  • A. 18008
  • B. 18006
  • C. 18002
  • D. 18000
Show answer and explanation ▾

Correct answer: C

iMaster NCE-Campus exposes its northbound open RESTful APIs on port 18002 by default, so API clients build their request URLs with that port.

Why the other options are wrong:

  • A. 18008 is not the default port of the northbound open APIs.
  • B. 18006 is not the default port of the northbound open APIs.
  • D. 18000 is not the default port of the northbound open APIs.

Question 49

Which of the following parameters is not included in an IPsec security association (SA)?

  • A. Security parameter index (SPI)
  • B. Destination IP address
  • C. Source IP address
  • D. Security protocol ID (AH or ESP)
Show answer and explanation ▾

Correct answer: C

An IPsec SA is unidirectional and is uniquely identified by three parameters: the security parameter index, the destination IP address, and the security protocol ID (AH or ESP). The source IP address is not part of that identifier.

Why the other options are wrong:

  • A. The SPI is one of the three parameters that identify an SA.
  • B. The destination IP address is one of the three parameters that identify an SA.
  • D. The security protocol ID, AH or ESP, is one of the three parameters that identify an SA.

Question 50

Which of the following statements about VXLAN principles is incorrect?

  • A. VXLAN uses MAC-in-UDP encapsulation to encapsulate Ethernet packets into UDP packets and thereby extend Layer 2 networks.
  • B. VXLAN can be deployed on campus networks to implement Layer 2 and Layer 3 communication.
  • C. VXLAN packets are forwarded through routes on the underlay network, and the MAC address of the terminal in the inner data frame is not considered during underlay forwarding.
  • D. When VXLAN is deployed, in addition to devices at both ends of a VXLAN tunnel, intermediate forwarding devices are also required to support VXLAN. Otherwise, VXLAN packets cannot be forwarded.
Show answer and explanation ▾

Correct answer: D

Only the VTEPs at the two ends of a tunnel encapsulate and decapsulate VXLAN packets. Intermediate devices forward them as ordinary IP and UDP packets based on the outer header, so they do not need to support VXLAN.

Why the other options are wrong:

  • A. VXLAN does use MAC-in-UDP encapsulation to extend Layer 2 networks, so this statement is accurate.
  • B. VXLAN is deployed on campus networks for both Layer 2 and Layer 3 communication, so this statement is accurate.
  • C. Underlay forwarding uses only the outer IP header and ignores the inner terminal MAC address, so this statement is accurate.

Question 51

Which of the following Layer 3 logical interfaces on a Layer 3 VXLAN gateway is used to implement communication between different VXLAN networks?

  • A. VLANIF interface
  • B. VBDIF interface
  • C. NVE interface
  • D. Layer 2 sub-interface
Show answer and explanation ▾

Correct answer: B

A VBDIF interface is a Layer 3 logical interface created for a bridge domain. Configured with an IP address on the Layer 3 VXLAN gateway, it routes traffic between different VXLAN segments and between VXLAN and non-VXLAN networks.

Why the other options are wrong:

  • A. A VLANIF interface is the Layer 3 interface of a VLAN, not of a VXLAN bridge domain.
  • C. An NVE interface is the logical interface that sets up VXLAN tunnels and does not act as a gateway interface.
  • D. A Layer 2 sub-interface connects users to a bridge domain and has no Layer 3 function.

Question 52

Which of the following is a data encryption algorithm?

  • A. MD5
  • B. SHA1
  • C. SHA-256
  • D. DES
Show answer and explanation ▾

Correct answer: D

DES is a symmetric encryption algorithm that turns plaintext into ciphertext with a shared secret, so the data can later be decrypted. The other three are hash algorithms that produce a fixed-length digest for integrity checks and cannot be reversed.

Why the other options are wrong:

  • A. MD5 is a hash algorithm used for integrity checks, not for encrypting data.
  • B. SHA1 is a hash algorithm used for integrity checks, not for encrypting data.
  • C. SHA-256 is a hash algorithm used for integrity checks, not for encrypting data.

Question 53

When a client invokes the iMaster NCE-Campus RESTful API, it sends an HTTP request. The server then returns the status code 401.

Which of the following errors has occurred?

  • A. Access denied.
  • B. Unauthorized.
  • C. The requested resource does not exist.
  • D. The service is unavailable.
Show answer and explanation ▾

Correct answer: B

HTTP status code 401 means Unauthorized: the request lacks valid authentication credentials, for example a missing or expired token. The client must authenticate again and resend the request.

Why the other options are wrong:

  • A. Access denied corresponds to status code 403, not 401.
  • C. A resource that does not exist returns status code 404.
  • D. An unavailable service returns status code 503.

Question 54

Which types of SIDs can be allocated by BGP EPE to inter-AS paths? (Choose all that apply.)

  • A. Peer-Node SID
  • B. Peer-Adj SID
  • C. Peer-Set SID
  • D. Peer-Prefix SID
Show answer and explanation ▾

Correct answer: A, B, C

BGP egress peer engineering allocates three kinds of peering SIDs to inter-AS paths: a Peer-Node SID identifies a peer node, a Peer-Adj SID identifies one adjacency to a peer, and a Peer-Set SID identifies a group of peers.

Why the other options are wrong:

  • D. Peer-Prefix SID is not one of the SID types that BGP EPE allocates.

Question 55

Which of the following devices provide access for wired and wireless users in the three-layer networking architecture of small and midsize campus networks? (Choose all that apply.)

  • A. AP
  • B. Firewall
  • C. Switch - D.AR
Show answer and explanation ▾

Correct answer: A, C

In the three-layer campus architecture the access layer connects end users: access switches provide wired access and APs provide wireless access.

Why the other options are wrong:

  • B. A firewall protects the network border and does not provide user access.

Question 56

The architecture of the Huawei SD-WAN Solution consists of the management layer, control layer, and network layer.

Which of the following devices are deployed at the network layer? (Choose all that apply.)

  • A. RR
  • B. EDGE
  • C. Firewall
  • D. GW
Show answer and explanation ▾

Correct answer: B, D

The network layer of the Huawei SD-WAN Solution is made up of the forwarding devices: EDGE devices at enterprise sites and GW devices that connect the SD-WAN network to legacy networks.

Why the other options are wrong:

  • A. The RR distributes routes between sites and belongs to the control layer.
  • C. A firewall is not one of the device roles defined for the network layer of this architecture.

Question 57

gRPC is a language-neutral, platform-neutral, and open source remote RPC system and supports the service methods of unary RPCs and streaming RPCs.

Which of the following service methods are not supported by gRPC? (Choose all that apply.)

  • A. rpc Cancel(CancelArgs) returns(Cancel Reply) {};
  • B. rpc Subscribe(SubsArgs) stream returns(Subs Reply) {};
  • C. rpc stream LotsOfGreetings(HelloRequest) returns (HelloResponse) {};
  • D. rpc dataPublish(stream serviceArgs) returns(stream serviceArgs) {};
Show answer and explanation ▾

Correct answer: B, C

gRPC supports unary, server streaming, client streaming, and bidirectional streaming methods, and the word stream is only valid inside the parentheses in front of the request or response type. The Subscribe and LotsOfGreetings definitions place stream outside the parentheses, so they are not supported method forms.

Why the other options are wrong:

  • A. A method with a single request and a single response is a unary RPC, which gRPC supports.
  • D. A method with stream on both the request and the response is a bidirectional streaming RPC, which gRPC supports.

Question 58

During the evolution from OSPFv2 to OSPFv3, LSA formats and functions remain unchanged, except that the network layer addresses in LSAs are changed from IPv4 to IPv6.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

OSPFv3 changes the LSAs themselves. It adds Link-LSAs and Intra-Area-Prefix-LSAs, removes prefix information from Router-LSAs and Network-LSAs, renames the Type 3 and Type 4 LSAs, and introduces flooding scopes.

Why the other options are wrong:

  • A. LSA types, formats and functions all change in OSPFv3, not only the address family.

Question 59

BGP is an EGP routing protocol. Therefore, BGP routing policies can be deployed only between EBGP peers. BGP routing policies deployed between IBGP peers are invalid.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

BGP routing policies can be applied to routes exchanged with any peer, IBGP or EBGP. Import and export policies toward IBGP peers take effect and are widely used, for example to set Local_Pref.

Why the other options are wrong:

  • A. Routing policies applied to IBGP peers are valid and take effect.

Question 60

SSH connections are usually established based on TCP/IP connections. In specific cases. SSH connections can be established based on UDP connections.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

SSH runs over a reliable, connection-oriented transport and uses TCP, with port 22 by default. It is not established over UDP.

Why the other options are wrong:

  • A. SSH connections are not established over UDP.

Question 61

NETCONF uses SSL to implement secure data transmission, and RPC to implement client-server communication.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

NETCONF uses SSH as its secure transport protocol, not SSL, and exchanges RPC messages between the client and the server over that SSH session.

Why the other options are wrong:

  • A. The secure transport that NETCONF uses is SSH, so the statement is wrong.

Question 62

The VNI field of VXLAN packets is 24 bits long.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

The VXLAN header contains a 24-bit VXLAN network identifier, which allows about 16 million isolated segments compared with 4096 VLANs.

Why the other options are wrong:

  • B. The VNI field is 24 bits long, so the statement is correct.

Question 63

The SBFD state machine running on the initiator has only the up and down states and can alternate only between the two states.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

The SBFD state machine on the initiator has only the Up and Down states and switches only between them. The session goes Up when a reflected packet carrying the Up state arrives, and goes Down when an AdminDown packet arrives or no reflected packet is received before the timer expires.

Why the other options are wrong:

  • B. The initiator has no other session states than Up and Down, so the statement is correct.

Question 64

MPLS LDP can be enabled for GRE tunnels.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

MPLS and MPLS LDP can be enabled on a GRE tunnel interface, so an LDP LSP can be set up across the tunnel. This carries MPLS services across a network segment that does not support MPLS.

Why the other options are wrong:

  • B. A GRE tunnel interface does support MPLS LDP.

Question 65

To prevent hackers from attacking user devices or networks using MAC addresses, you can configure MAC addresses of untrusted users as blackhole MAC addresses to filter out such invalid MAC addresses.

When receiving a packet whose source or destination MAC address is a blackhole MAC address on a device, the device discards the packet.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

A blackhole MAC address entry tells the device to discard any packet whose source or destination MAC address matches it. Adding the MAC addresses of untrusted users as blackhole entries filters their traffic.

Why the other options are wrong:

  • B. Packets with a matching source or destination MAC address are discarded, as the statement says.

Question 66

An NVE is a network entity that implements network virtualization and must be a physical switch.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

A network virtualization edge is the entity that implements network virtualization, and it can be a hardware switch or a software device such as a virtual switch on a server.

Why the other options are wrong:

  • A. An NVE does not have to be a physical switch.

Question 67

How many bits does an MPLS label field have?

  • A. 8
  • B. 3
  • C. 1
  • D. 20
Show answer and explanation ▾

Correct answer: D

The 32-bit MPLS header consists of a 20-bit label value, a 3-bit Exp field, a 1-bit bottom of stack flag, and an 8-bit TTL.

Why the other options are wrong:

  • A. 8 bits is the length of the TTL field.
  • B. 3 bits is the length of the Exp field.
  • C. 1 bit is the length of the bottom of stack flag.

Question 68

Which of the following statements about traffic policing is correct?

  • A. Traffic policing can be used only in the outbound direction.
  • B. Traffic policing can be used only in the inbound direction.
  • C. Traffic policing discards excess traffic.
  • D. Traffic policing buffers excess traffic.
Show answer and explanation ▾

Correct answer: C

Traffic policing measures traffic against a committed rate and drops, or re-marks, the packets that exceed it. It does not queue the excess traffic.

Why the other options are wrong:

  • A. Traffic policing can be applied in both the inbound and the outbound direction.
  • B. Traffic policing can be applied in both the inbound and the outbound direction.
  • D. Buffering excess traffic is what traffic shaping does.

Question 69

On a virtualized network deployed using iMaster NCE-Campus, which of the following is not one of the operations performed by the administrator in the network planning phase?

  • A. Create an authentication template.
  • B. Deploy an underlay automation resource pool.
  • C. Deploy a global resource pool for the fabric.
  • D. Deploy an overlay access resource pool.
Show answer and explanation ▾

Correct answer: A

In the network planning phase the administrator prepares the resource pools used later by the underlay, the fabric and the overlay. Authentication templates are created when access management is configured, not during network planning.

Why the other options are wrong:

  • B. The underlay automation resource pool is planned in the network planning phase.
  • C. The global resource pool for the fabric is planned in the network planning phase.
  • D. The overlay access resource pool is planned in the network planning phase.

Question 70

If multiple types of service packets are transmitted over just one link, the traffic of high-priority applications is preferentially processed when congestion occurs, ensuring a smooth user experience for high-priority applications. To achieve this, which of the following intelligent traffic steering policies can be used?

  • A. Link quality-based traffic steering
  • B. Link bandwidth-based traffic steering
  • C. Link priority-based traffic steering
  • D. Application priority-based traffic steering
Show answer and explanation ▾

Correct answer: D

Application priority-based traffic steering handles several services that share one link. When the link is congested, the traffic of high-priority applications is served first so their experience is protected.

Why the other options are wrong:

  • A. Link quality-based steering selects a link by delay, jitter and loss and does not rank applications on one link.
  • B. Link bandwidth-based steering distributes traffic by bandwidth usage across links.
  • C. Link priority-based steering prefers one link over another and does not rank applications on the same link.

Question 71

Which of the following languages is integrated into the OPS?

  • A. Python
  • B. C programming language
  • C. Java
  • D. C++
Show answer and explanation ▾

Correct answer: A

The Open Programmability System integrates a Python runtime environment, so users write Python scripts that call the OPS RESTful APIs to manage the device.

Why the other options are wrong:

  • B. The C programming language is not the script language integrated into the OPS.
  • C. Java is not the script language integrated into the OPS.
  • D. C++ is not the script language integrated into the OPS.

Question 72

The Python Paramiko library implements the SSH protocol. In the Python Paramiko module, which of the following methods can be used to enable an interactive SSH session?

  • A. from_transport()
  • B. connect()
  • C. Transport()
  • D. invoke_shell()
Show answer and explanation ▾

Correct answer: D

In Paramiko, invoke_shell() opens an interactive shell channel on an established SSH connection. Commands are then sent and output is read through that channel, which is how scripts drive a network device CLI.

Why the other options are wrong:

  • A. from_transport() builds an SFTP client from an existing transport and does not open an interactive shell.
  • B. connect() establishes the SSH connection and authenticates, but does not start an interactive session.
  • C. Transport() creates the underlying transport object and does not start an interactive shell.

Question 73

Which of the following technologies can be used to isolate users in the same VLAN?

  • A. Ethernet port security
  • B. Port isolation
  • C. Super VLAN
  • D. IPSG
Show answer and explanation ▾

Correct answer: B

Port isolation places ports in an isolation group, and ports in the same group cannot exchange Layer 2 traffic even though they belong to the same VLAN.

Why the other options are wrong:

  • A. Port security limits the MAC addresses learned on a port and does not isolate users from each other.
  • C. A super VLAN aggregates sub-VLANs under one gateway and separates users by placing them in different sub-VLANs.
  • D. IPSG checks source addresses against a binding table to stop spoofing and does not isolate users.

Question 74

In the firewall hot standby scenario, which of the following protocols is used to back up the session table between the active and standby firewalls?

  • A. VRRP
  • B. VGMP
  • C. BFD
  • D. HRP
Show answer and explanation ▾

Correct answer: D

The Huawei Redundancy Protocol backs up dynamic state such as the session table, along with important configuration, from the active firewall to the standby firewall over the heartbeat link, so sessions survive a switchover.

Why the other options are wrong:

  • A. VRRP provides the virtual gateway address and does not back up sessions.
  • B. VGMP manages the VRRP groups so they switch together and does not back up the session table.
  • C. BFD detects link and peer faults quickly and does not back up sessions.

Question 75

Compared with traditional WANs, SD-WAN provides better application experience.

Which of the following is the main reason for SD-WAN?

  • A. SD-WAN ensures VRF-based isolation of services between multiple departments.
  • B. SD-WAN provides different routing policies and QoS policies for different applications.
  • C. SD-WAN uses IPsec for encrypted packet transmission.
  • D. SD-WAN supports multiple networking modes, such as full mesh, hub-spoke, and partial mesh.
Show answer and explanation ▾

Correct answer: B

SD-WAN identifies applications and applies a routing policy and a QoS policy per application. Important applications are steered to links that meet their quality needs and get priority treatment, which is what improves application experience.

Why the other options are wrong:

  • A. VRF-based isolation separates departments and does not by itself improve application experience.
  • C. IPsec encryption protects the traffic and does not improve application experience.
  • D. Flexible networking modes concern topology, not how each application is treated.

Question 76

Portal authentication, also known as web authentication, is used to authenticate user identities on a web authentication page.

Which of the following statements about Portal authentication is incorrect?

  • A. Before Portal authentication, the authentication client must obtain the IP address and be able to communicate with the Portal server.
  • B. In most cases, Portal authentication does not require installation of dedicated client software. Therefore, Portal authentication is mainly used in scenarios where no client software is available.
  • C. When Portal authentication is deployed on the network, clients to be authenticated must support the Portal protocol.
  • D. Portal authentication can authenticate users based on the combination of user name, VLAN, IP address, and MAC address.
Show answer and explanation ▾

Correct answer: C

Portal authentication only needs a web browser on the client. The Portal protocol runs between the access device and the Portal server, so the client itself does not have to support it, which makes this statement the incorrect one.

Why the other options are wrong:

  • A. This is accurate: the client needs an IP address and a path to the Portal server before the authentication page can open.
  • B. This is accurate: a browser is enough, so Portal suits users who have no dedicated client software.
  • D. This is accurate: Portal can combine the user name with the VLAN, IP address and MAC address when authenticating a user.

Question 77

Assume that the SRH of an SRv6 packet has five segment IDs.

When the packet arrives at the third endpoint, what is the SL value in the SRH?

  • A. 5
  • B. 3
  • C. 2
  • D. 1
Show answer and explanation ▾

Correct answer: C

With five segment IDs the source sets SL to 4, pointing at the first segment. Each endpoint decrements SL by 1 before forwarding, so the packet reaches the second endpoint with SL 3 and the third endpoint with SL 2.

Why the other options are wrong:

  • A. With five segment IDs the highest SL value is 4, so SL is never 5.
  • B. SL is 3 when the packet arrives at the second endpoint.
  • D. SL is 1 when the packet arrives at the fourth endpoint.

Question 78

In Huawei's NAC solution, which of the following cannot be used for Portal authentication?

  • A. User name and password
  • B. User's MAC address
  • C. SMS verification code
  • D. Passcode
Show answer and explanation ▾

Correct answer: B

Portal authentication checks credentials that the user submits on a web page, such as a user name and password, an SMS verification code or a passcode. A terminal's MAC address is the credential of MAC address authentication, which is a separate NAC method.

Why the other options are wrong:

  • A. A user name and password is the standard credential entered on the Portal page.
  • C. An SMS verification code is a supported Portal credential, commonly used for guests.
  • D. A passcode is a supported Portal credential, commonly used for guest access.

Question 79

OSPFv2 is an IGP running on IPv4 networks, whereas OSPFv3 is an IGP running on IPv6 networks. OSPFv3 and OSPFv2 use the same types of packets, including Hello, DD, LSR, LSU, and LSAck packets.

Which of the following statements is correct about OSPFv3 packets?

  • A. An OSPFv3 Hello packet carries the IPv6 address prefix of a router interface.
  • B. OSPFv3 uses the link-local address as the source address for sending packets. Packets can be forwarded beyond the originating link.
  • C. OSPFv3 uses the authentication field in the packet header for packet authentication.
  • D. OSPFv3 uses IPv6 multicast addresses FF02::5 and FF02::6 to send OSPFv3 packets.
Show answer and explanation ▾

Correct answer: D

OSPFv3 sends packets to the IPv6 multicast address FF02::5 for all OSPF routers and FF02::6 for the DR and BDR, the counterparts of 224.0.0.5 and 224.0.0.6 in OSPFv2.

Why the other options are wrong:

  • A. An OSPFv3 Hello packet carries no address prefix; prefixes are advertised in LSAs.
  • B. Packets sourced from a link-local address are never forwarded beyond the originating link.
  • C. The OSPFv3 packet header has no authentication field; authentication relies on IPv6 mechanisms.

Question 80

An enterprise has various services on the bearer WAN. To ensure that protocol packets are properly forwarded and processed even during network congestion, which of the following priority should be set for protocol packets?

  • A. CS6/CS7
  • B. EF
  • C. AF4
  • D. BE
Show answer and explanation ▾

Correct answer: A

CS6 and CS7 are the highest service classes and are reserved for network control and protocol packets, such as routing protocol and management traffic. Using them keeps protocol packets forwarded and processed first when the network is congested.

Why the other options are wrong:

  • B. EF is planned for low-delay service traffic such as voice, not for protocol packets.
  • C. AF4 is planned for service traffic such as real-time video, not for protocol packets.
  • D. BE is the default best-effort class and gives no protection during congestion.

Question 81

During the implementation of iMaster NCE-Campus Wi-Fi location-based services (LBS), terminal location data is sent to an LBS platform through HTTP requests.

Which of the following methods is used in such HTTP requests?

  • A. GET
  • B. PUT
  • C. POST
  • D. DELETE
Show answer and explanation ▾

Correct answer: C

Terminal location data is pushed to the LBS platform in the body of HTTP POST requests. POST is the HTTP method used to submit data to a server for processing.

Why the other options are wrong:

  • A. GET retrieves a resource and is not the method used to push the location data.
  • B. PUT is not the method used for this location data report.
  • D. DELETE removes a resource and does not report data.

Question 82

WAN VPN can be classified as L2VPN or L3VPN depending on the layer where it is implemented.

Which of the following statements about WAN VPN is incorrect?

  • A. Traditional L2VPN includes VPLS and VPWS, which can use LDP or BGP to establish virtual links.
  • B. Traditional L3VPN uses VPN instances to isolate services and BGP VPNv4/v6 to transmit Layer 3 information.
  • C. EVPN can function as the control plane protocol for L2VPN and L3VPN.
  • D. On a Huawei device, EVPN provides different address families for L2VPN and L3VPN services.
Show answer and explanation ▾

Correct answer: D

On Huawei devices EVPN carries both Layer 2 and Layer 3 services in one BGP EVPN address family and tells them apart by route type. It does not use different address families for L2VPN and L3VPN, so this statement is the incorrect one.

Why the other options are wrong:

  • A. This is accurate: VPLS and VPWS are the traditional L2VPN services and can be signaled with LDP or BGP.
  • B. This is accurate: traditional L3VPN isolates services with VPN instances and carries routes in BGP VPNv4 or VPNv6.
  • C. This is accurate: EVPN serves as a unified control plane for L2VPN and L3VPN.

Question 83

Which of the following user logout modes are supported by 802.1X authentication? (Choose three.)

  • A. A user logs out manually.
  • B. An access device forces a user to log out.
  • C. An authentication server forces a user to log out.
  • D. An accounting server forces a user to log out.
Show answer and explanation ▾

Correct answer: A, B, C

An 802.1X user can log out manually, in which case the client sends an EAPOL-Logoff packet. The access device can also force a user offline, and the authentication server can force a logout by sending a disconnect request to the access device.

Why the other options are wrong:

  • D. An accounting server records usage and is not one of the parties that forces an 802.1X user to log out.

Question 84

When you create a fabric on iMaster NCE-Campus, which of the following parameters need to be set in the fabric global resource pool? (Choose three.)

  • A. VLAN
  • B. Interconnection IP address
  • C. VXLAN network identifier
  • D. Bridge domain
Show answer and explanation ▾

Correct answer: A, C, D

The fabric global resource pool on iMaster NCE-Campus holds the VLAN, VXLAN network identifier and bridge domain ranges. The controller draws from these pools automatically when it creates virtual networks and access resources on the fabric.

Why the other options are wrong:

  • B. Interconnection IP addresses are not part of the fabric global resource pool.

Question 85

Which of the following statements about URIs and uniform resource locators (URLs) are correct? (Choose three.)

  • A. A URI is used to uniquely identify a resource.
  • B. A URL is a URI that identifies a resource and specifies how to locate the resource.
  • C. REST determines resources distributed over the entire network by a URI. Applications on clients obtain resources through URLs.
  • D. http://www.ietf.org/rfc/rfc2396.txt and ftp://ftp.is.co.za/rfc/rfc1808.txt are URLs.
Show answer and explanation ▾

Correct answer: A, B, D

A URI uniquely identifies a resource. A URL is a kind of URI that also says how to locate the resource, for example by scheme and host, and the http and ftp addresses listed are both URLs.

Why the other options are wrong:

  • C. REST identifies and accesses resources through URIs, so this split between URIs and URLs does not describe REST correctly.

Question 86

Which of the following statements are correct about the security protocols used by IPsec? (Choose three.)

  • A. AH performs integrity check for an entire IP packet.
  • B. ESP encrypts the entire IP packet.
  • C. Both AH and ESP support tunnel encapsulation.
  • D. ESP is an IP-based transport-layer protocol, with the protocol number 50.
Show answer and explanation ▾

Correct answer: A, C, D

AH checks the integrity of the entire IP packet, including the IP header. AH and ESP both support tunnel mode as well as transport mode, and ESP runs directly over IP with protocol number 50.

Why the other options are wrong:

  • B. ESP encrypts the payload it protects, not the entire IP packet; the outer IP header is never encrypted.

Question 87

Which of the following intelligent O&M functions are supported by the Huawei CloudCampus solution? (Choose four.)

  • A. Uses telemetry technology to monitor the wireless network quality in terms of the AP, radio, and user.
  • B. Provides visualized monitoring of network quality based on network health.
  • C. Analyzes network faults based on individual and group faults.
  • D. Offers predictive intelligent optimization capabilities based on big data and AI.
Show answer and explanation ▾

Correct answer: A, B, C, D

Intelligent O&M in the CloudCampus solution uses telemetry to monitor wireless quality per AP, radio and user, shows network health visually, analyzes both individual and group faults, and uses big data and AI for predictive optimization.


Question 88

Which of the following can be used to set the internal priority of packets in behavior aggregate classification (priority mapping)? (Choose two.)

  • A. DSCP value
  • B. MPLS EXP value
  • C. Source or destination IP address
  • D. Traffic class
Show answer and explanation ▾

Correct answer: A, B

Behavior aggregate classification maps the priority field already carried in a packet to an internal priority. The DSCP value of an IP packet and the EXP value of an MPLS packet are both such fields.

Why the other options are wrong:

  • C. Source and destination IP addresses are matched in multi-field classification, not in priority mapping.
  • D. Traffic class is not the packet field named for priority mapping here.

Question 89

Which of the following items are included in static information collection and analysis? (Choose three.)

  • A. Device types
  • B. Interface types
  • C. License
  • D. Packet loss rate
Show answer and explanation ▾

Correct answer: A, B, C

Static information is data that does not change while the network runs, such as device types, interface types and license details. It is collected once and analyzed to understand what is deployed.

Why the other options are wrong:

  • D. Packet loss rate changes with traffic, so it is dynamic running information.

Question 90

On small and midsize campus networks, stacking and inter-device link aggregation can be deployed on switches to implement link reliability.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

Stacking combines several switches into one logical device, and inter-device link aggregation bundles uplinks that end on different member switches. Together they keep links available when one member or one link fails.

Why the other options are wrong:

  • B. Stacking with inter-device link aggregation is a standard reliability design for small and midsize campus networks.

Question 91

Huawei Open Programmability System (OPS) enables users to run Python scripts on local PCs to invoke open RESTful APIs of devices.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

OPS runs Python scripts on the network device itself. The scripts are installed on the device and call its open RESTful APIs locally, so the statement about scripts running on local PCs is false.

Why the other options are wrong:

  • A. OPS scripts are stored and executed on the device, not on a local PC.

Question 92

Traffic policing can only be applied to the inbound direction. Its purpose is to monitor the traffic entering the device and ensure that the upstream device does not unnecessarily consume network resources.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

Traffic policing can be applied in both the inbound and the outbound direction of an interface. It limits traffic to the committed rate and drops or re-marks the excess in either direction.

Why the other options are wrong:

  • A. Traffic policing is not limited to the inbound direction.

Question 93

In the inter-AS MPLS L3VPN Option B solution, the RT filtering function must be disabled on ASBRs. Otherwise, VPNv4 routes cannot be transmitted.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

Disabling RT filtering is one way to make an ASBR keep VPNv4 routes, but not the only one. If the ASBR has VPN instances whose import RTs match the routes, the routes are accepted and transmitted with RT filtering still enabled.

Why the other options are wrong:

  • A. RT filtering does not have to be disabled when matching VPN instances exist on the ASBR.

Question 94

If public key authentication is used in the SSH user authentication phase, the public key must be manually configured on the SSH server in advance.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

With this method the server verifies the client's signature against the peer's public information that is stored locally and bound to the SSH user. That information has to be configured on the server before the user can log in.

Why the other options are wrong:

  • B. The server cannot verify the client's signature unless the client's public information was configured on it beforehand.

Question 95

OSPFv3 and OSPFv2 use the same route advertisement mode. Specifically, routes are advertised by running the network command in the OSPFv3 area view.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

OSPFv3 runs per link and is enabled directly on each interface, which is assigned to a process and an area in the interface view. It has no network command in the area view, unlike OSPFv2.

Why the other options are wrong:

  • A. OSPFv3 does not advertise routes with a network command in the area view.

Question 96

In Huawei's CloudWAN solution, the IGP is a supporting protocol that can collect and flood Layer 3 topology information on the entire network. In addition, the IGP can work with TWAMP/iFIT to collect SLA information such as link delay on the network.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

In the CloudWAN solution the IGP collects and floods the Layer 3 topology of the network. With TWAMP or iFIT measuring the links, it can also flood SLA information such as link delay for path computation.

Why the other options are wrong:

  • B. The statement correctly describes the supporting role of the IGP in the CloudWAN solution.

Question 97

A fault is any phenomenon that adversely affects services for users.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

In network troubleshooting a fault is defined from the user's point of view: any phenomenon that adversely affects the services users rely on counts as a fault.

Why the other options are wrong:

  • B. The statement matches the troubleshooting definition of a fault.

Question 98

The enterprise WAN provides interconnection between enterprise headquarters and branches, between enterprises and clouds, and between clouds.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: A

An enterprise WAN connects sites over long distances. It links headquarters with branches, connects the enterprise to clouds, and interconnects the clouds themselves.

Why the other options are wrong:

  • B. The statement lists the interconnection scenarios an enterprise WAN covers.

Question 99

OSPFv3 link LSAs are advertised in the entire area.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

An OSPFv3 Link LSA has link-local flooding scope. It is sent only to routers attached to the same link and is never flooded through the whole area.

Why the other options are wrong:

  • A. Link LSAs are flooded only on the local link, not in the entire area.

Question 100

In the Huawei SD-WAN Solution, the topologies of different VNs must be the same.

  • A. True
  • B. False
Show answer and explanation ▾

Correct answer: B

In the Huawei SD-WAN Solution each VN can be given its own overlay topology, for example hub-spoke for one VN and full mesh for another. Different VNs are not required to share the same topology.

Why the other options are wrong:

  • A. Topologies are planned per VN and can differ from one VN to another.

Question 101

Which of the following inter-AS MPLS L3VPN solutions needs ASBRs to transmit VPNv4 routes?

  • A. Option A
  • B. Option C
  • C. Option B
  • D. Option D
Show answer and explanation ▾

Correct answer: C

In Option B the ASBRs set up an MP-EBGP session with each other and exchange the VPNv4 routes they receive from the PEs in their own ASs, so the ASBRs themselves transmit VPNv4 routes.

Why the other options are wrong:

  • A. In Option A the ASBRs treat each other as CEs and exchange ordinary IPv4 routes per VPN instance.
  • B. In Option C the PEs exchange VPNv4 routes directly over a multihop session, and the ASBRs do not carry them.
  • D. Option D is not the solution in which the ASBRs transmit VPNv4 routes.

Question 102

Which of the following statements are correct about the OSPFv3 packet format? (Choose all that apply.)

  • A. Hello packets do not contain address information but carry interface IDs.
  • B. The Authentication and Auth Type fields are removed.
  • C. The OSPF version number is changed from 2 to 3.
  • D. The Options field in the Hello packet does not change.
Show answer and explanation ▾

Correct answer: A, B, C

OSPFv3 Hello packets carry an interface ID instead of address information. The Authentication and Auth Type fields are removed from the packet header, and the version number changes from 2 to 3.

Why the other options are wrong:

  • D. The Options field in an OSPFv3 Hello packet is extended to 24 bits, so it does change.

Question 103

Which of the following IPv6 transition technologies are address translation technologies? (Choose all that apply.)

  • A. 6PE
  • B. IVI
  • C. IPv4/IPv6 dual stack
  • D. NAT64
Show answer and explanation ▾

Correct answer: B, D

IVI and NAT64 both translate between IPv6 and IPv4 addresses and packet headers, so IPv6-only and IPv4-only nodes can communicate. They are address translation technologies.

Why the other options are wrong:

  • A. 6PE is a tunneling technology that carries IPv6 traffic across an IPv4 MPLS backbone.
  • C. Dual stack runs IPv4 and IPv6 side by side on the same device and translates nothing.

Question 104

Which of the following statements is correct about the result of the ipv6 enable topology ipv6 command configured in the IS-IS view?

  • A. IPv6 SPF calculation is not supported.
  • B. IPv4 and IPv6 use the same topology for SPF calculation.
  • C. It does not have any meaning.
  • D. SPF calculation is performed in IPv4 and IPv6 topologies separately.
Show answer and explanation ▾

Correct answer: D

The ipv6 enable topology ipv6 command enables IPv6 for the IS-IS process in a separate IPv6 topology. IS-IS then runs SPF for the IPv4 topology and the IPv6 topology independently.

Why the other options are wrong:

  • A. The command enables IPv6 route calculation; it does not remove it.
  • B. A shared topology is the result of topology standard, not topology ipv6.
  • C. The command has a clear effect: it selects a separate IPv6 topology.

Question 105

If a BFD session goes down, a direct link becomes disconnected.

  • A. TRUE
  • B. FALSE
Show answer and explanation ▾

Correct answer: B

A BFD session going down only means BFD packets were not received within the detection time. The cause can be packet loss, congestion, a fault on a multihop path or a problem on the peer, so the direct link is not necessarily disconnected.

Why the other options are wrong:

  • A. A BFD session can go down while the direct physical link is still up.

Question 106

On traditional campus networks, network admission control (NAC), VLAN, and ACL technologies are used to control users' network access rights. This results in heavy network O&M workload. Which of the following improvements does the free mobility solution of iMaster NCE-campus make to traditional campus networks in terms of policy management and control? (Choose all that apply.)

  • A. Policy enforcement points on a network can subscribe to IP-security group entries through iMaster NCE-Campus to obtain the mapping between users and IP addresses on the entire network.
  • B. Administrators can define service policies based on security groups to decouple service policies from IP addresses.
  • C. Administrators can use iMaster NCE-campus to centrally manage service policies on network-wide policy enforcement points.
  • D. Administrators need to configure service policies multiple times for each policy enforcement point through iMaster NCE-Campus, which then automatically delivers the policies to the corresponding policy enforcement points.
Show answer and explanation ▾

Correct answer: A, B, C

Free mobility defines policies between security groups, which decouples them from IP addresses. iMaster NCE-Campus manages the policies centrally for all policy enforcement points, and those points can subscribe to IP-security group entries to learn the user-to-IP mappings of the whole network.

Why the other options are wrong:

  • D. Policies are defined once on iMaster NCE-Campus and delivered to all enforcement points, not configured repeatedly per point.

Question 107

On a small- or medium-sized campus network deployed based on the Huawei CloudCampus Solution, an AR functioning as the egress gateway supports web-based network management, registration center query, and DHCP Option 148-based deployment, but does not support CLI-based deployment.

  • A. TRUE
  • B. FALSE
Show answer and explanation ▾

Correct answer: B

An AR used as the egress gateway can be deployed through the web system, the registration center, DHCP Option 148 and also the CLI. The claim that CLI-based deployment is not supported makes the statement false.

Why the other options are wrong:

  • A. CLI-based deployment is also available for the AR egress gateway.

Question 108

Which of the following statements are correct about network deployment design for a large or medium-sized virtualized campus network?

(Choose all that apply.)

  • A. The service configurations for egress devices on the campus network are similar. It is recommended that the DHCP Option mode be used to implement plug-and-play for these devices.
  • B. Huawei CloudCampus Solution requires that devices are onboarded on iMaster NCE-Campus before service planning.
  • C. An independent DHCP server needs to be planned on the network to allocate IP addresses to end users.
  • D. When DHCP Option 148 is used to deploy switches, the DHCP server needs to provide the IP address and port number of iMaster NCECampus to the switches.
Show answer and explanation ▾

Correct answer: A, B, D

Egress devices have similar service configurations, so DHCP Option based plug-and-play suits them. Devices must be onboarded on iMaster NCE-Campus before services are planned, and with DHCP Option 148 the DHCP server gives switches the IP address and port number of iMaster NCE-Campus.

Why the other options are wrong:

  • C. An independent DHCP server for end users is not a required part of this deployment design.

Question 109

If the number of MAC addresses learned on an interface enabled with port security reaches the upper limit, which of the following actions may the switch take? (Choose all that apply.)

  • A. Sets the interface state to error-down and generates an alarm.
  • B. Discards packets with unknown source MAC addresses without generating any alarm.
  • C. Discards packets with unknown source MAC addresses and generates an alarm.
  • D. Sets the interface state to error-down without generating any alarm.
Show answer and explanation ▾

Correct answer: A, B, C

Port security has three protective actions once the MAC address limit is reached. Restrict discards packets from new source MAC addresses and reports an alarm, protect discards them without an alarm, and shutdown sets the interface to error-down and reports an alarm.

Why the other options are wrong:

  • D. The shutdown action always reports an alarm when it sets the interface to error-down.

Question 110

In a scenario where a VXLAN tunnel is dynamically established through BGP EVPN, compared with host ARP entry advertisement, which of the following is added in BGP EVPN Type 2 routes used to advertise host IP routes?

  • A. L3VNI
  • B. L2VNI
  • C. MAC Address
  • D. IP Address
Show answer and explanation ▾

Correct answer: A

A Type 2 route that advertises a host IP route carries the Layer 3 VNI of the tenant VPN instance in addition to the Layer 2 VNI, so the receiving VTEP can install the host route in the right VPN instance and forward routed traffic over the VXLAN tunnel.

Why the other options are wrong:

  • B. The L2VNI is already carried when the route advertises a host ARP entry.
  • C. The host MAC address is already present in a Type 2 route that advertises an ARP entry.
  • D. The host IP address is already present in a Type 2 route that advertises an ARP entry.

Question 111

When you configure external networks connected to a fabric on iMaster NCE-Campus, which of the following parameters need to be set? (Choose all that apply.)

  • A. Interconnection IP address
  • B. Loopback interface address
  • C. Interconnection port
  • D. External service IP address
Show answer and explanation ▾

Correct answer: A, C, D

An external network defines how the border node hands traffic to an egress device. The administrator selects the interconnection port, sets the interconnection IP addresses for that link, and specifies the external service addresses that the fabric reaches through it.

Why the other options are wrong:

  • B. A loopback interface address is not a parameter of the external network that connects the border node to the egress device.

Question 112

In Huawei SD-WAN Solution, a CPE can be managed by iMaster NCE-WAN only after it establishes a control channel with iMaster NCE-WAN.

  • A. TRUE
  • B. FALSE
Show answer and explanation ▾

Correct answer: A

iMaster NCE-WAN delivers configurations to a CPE and collects its status over a control channel. Until the CPE registers and that channel is up, the controller has no path to the device and cannot manage it.

Why the other options are wrong:

  • B. The controller cannot push configuration to or monitor a CPE that has not set up its control channel.

Question 113

If the K bit in the GRE header is set to 1, a 4-byte Key field is inserted into the GRE header. The sender and receiver then authenticate each other based on this key.

  • A. TRUE
  • B. FALSE
Show answer and explanation ▾

Correct answer: B

With the K bit set, the GRE header does carry an extra 4-byte field, but its job is a simple check: the receiving end compares the value in each arriving packet with its own setting and discards packets that do not match. It is not a mutual authentication exchange between the two ends.

Why the other options are wrong:

  • A. The field lets the receiver verify incoming tunnel packets, and the two ends do not authenticate each other with it.

Question 114

Which of the following intelligent traffic steering policies can be used to fully utilize link bandwidth and implement link bandwidth-based load balancing on an enterprise network with multiple links?

  • A. Traffic bandwidth-based traffic steering
  • B. Link priority-based traffic steering
  • C. Link quality-based traffic steering
  • D. Application priority-based traffic steering
Show answer and explanation ▾

Correct answer: A

Bandwidth-based traffic steering distributes application traffic across the available links according to their bandwidth and current usage, so all links carry load and none sits idle while another is congested.

Why the other options are wrong:

  • B. Link priority-based steering sends traffic over the preferred link and keeps the others mainly as backups.
  • C. Link quality-based steering moves traffic when delay, jitter or packet loss crosses a threshold, not to balance bandwidth.
  • D. Application priority-based steering protects high-priority applications when links are congested, not to balance bandwidth.

Question 115

Which of the following steps is not mandatory when manually configuring an OSPF-based SR-MPLS BE tunnel?

  • A. Enable segment routing globally.
  • B. Enable MPLS globally.
  • C. Configure the LSR ID for routers.
  • D. Configure the adjacency SID for each link.
Show answer and explanation ▾

Correct answer: D

An SR-MPLS BE tunnel is built from prefix or node SIDs along the shortest path that the IGP calculates, so adjacency SIDs are not required. Adjacency SIDs matter for explicit paths, as in SR-MPLS TE.

Why the other options are wrong:

  • A. Segment routing must be enabled globally before the IGP can advertise SIDs.
  • B. MPLS must be enabled globally because SR-MPLS forwards packets with MPLS labels.
  • C. An LSR ID must be configured on each router before MPLS can be enabled.

Question 116

Prefix segments and adjacency segments are globally visible and unique.

  • A. TRUE
  • B. FALSE
Show answer and explanation ▾

Correct answer: B

A prefix segment is globally visible and globally unique within the SR domain. An adjacency segment is advertised to other nodes, but it is valid only on the node that allocated it, so its value is locally significant and can be reused on other nodes.

Why the other options are wrong:

  • A. Adjacency segments are only locally valid, so the statement does not hold for both segment types.

Question 117

If IS-IS is used to transmit SIDs on an SR network, the node SIDs advertised by a Level-2 router are transmitted only within the same area.

  • A. TRUE
  • B. FALSE
Show answer and explanation ▾

Correct answer: B

Level-2 routers form the IS-IS backbone, and Level-2 LSPs are flooded to all Level-2 routers across areas. Node SIDs advertised by a Level-2 router therefore travel through the whole Level-2 domain and are not confined to one area.

Why the other options are wrong:

  • A. Level-2 LSPs, including the SIDs they carry, are flooded between areas across the Level-2 backbone.

Question 118

If an engineer wants to remotely log in to a Huawei device, the engineer uses the Paramiko library to compile the following code when compiling the SSH Python script:

cli = ssh.invoke_shell() cli.send('screen-length 0 temporary\n')

What is the function of "screen-length 0 temporary"?

  • A. Canceling split-screen display
  • B. Performing automatic word wrap
  • C. Performing split-screen display
  • D. Canceling automatic word wrap
Show answer and explanation ▾

Correct answer: A

The screen-length 0 temporary command turns off split-screen display for the current session, so long command output is returned in one piece. Without it, the device pauses at each screen and the script would have to answer the More prompt.

Why the other options are wrong:

  • B. The command controls how many lines are shown per screen and has nothing to do with word wrap.
  • C. A value of 0 disables split-screen display instead of enabling it.
  • D. The command does not change word wrap behavior.

Question 119

Which of the following are HTTP features? (Choose all that apply.)

  • A. Media independent
  • B. Connectionless
  • C. Stateless
  • D. UDP encapsulation
Show answer and explanation ▾

Correct answer: A, C

HTTP is media independent, because any type of data can be carried as long as both ends know how to handle the content type. It is also stateless: the protocol itself keeps no memory of earlier requests, so each request is handled on its own.

Why the other options are wrong:

  • B. HTTP runs over TCP, which is a connection-oriented transport protocol.
  • D. HTTP messages are carried over TCP, not encapsulated in UDP.

Question 120

In the firewall hot standby scenario, which of the following information is not carried by VGMP packets?

  • A. Software version of the local device
  • B. Status of the local VGMP group
  • C. Priority of the local VGMP group
  • D. Whether the local device is busy
Show answer and explanation ▾

Correct answer: D

VGMP packets let the two firewalls compare their VGMP groups and decide which device is active, so they carry information such as the status and priority of the local VGMP group. A flag that says whether the local device is busy is not part of that exchange.

Why the other options are wrong:

  • A. The software version of the local device is among the information carried in VGMP packets.
  • B. The status of the local VGMP group is carried so the peer knows whether the device is active or standby.
  • C. The priority of the local VGMP group is carried so the two devices can compare priorities.

Question 121

In scenarios other than virtual links, which type of IPv6 address is the source IPv6 address of OSPFv3 Hello packets?

  • A. Unique local address (ULA)
  • B. Global unicast address
  • C. Link-local address
  • D. IPv6 anycast address
Show answer and explanation ▾

Correct answer: C

OSPFv3 runs per link and sends its packets with the link-local address of the outgoing interface as the source address on every interface except virtual links. Virtual links cross several hops, so they need an address with a wider scope.

Why the other options are wrong:

  • A. A unique local address is not used as the source of OSPFv3 Hello packets on ordinary links.
  • B. A global unicast address is used as the source only on virtual links.
  • D. An anycast address is never used as the source address of a packet.

Question 122

Which of the following statements is correct about the differences between NSR and NSF?

  • A. NSF does not require the collaboration of neighboring routers.
  • B. Both NSR and NSF require the collaboration of neighboring routers.
  • C. NSR requires the collaboration of neighboring routers.
  • D. NSF requires the collaboration of neighboring routers.
Show answer and explanation ▾

Correct answer: D

NSF relies on graceful restart: during a switchover the neighbors act as helpers, keep the adjacency and help the restarting router rebuild its routing information. NSR backs up protocol state to the standby main board, so the switchover is invisible to neighbors and needs no help from them.

Why the other options are wrong:

  • A. NSF depends on neighboring routers acting as graceful restart helpers.
  • B. NSR completes the switchover with state backed up inside the device, without neighbor involvement.
  • C. NSR hides the switchover from neighbors, so their collaboration is not needed.

That was 122 of 572.

The full Huawei H12-891 HCIE-Datacom pack has all 572 questions, each with the answer, the explanation and why the other options are wrong, plus a questions-only copy for timed runs. $39.00 USD, paid once, with free monthly updates and a pass-or-your-money-back guarantee.

Get the full pack

Not ready yet? Get the first 154 free as one PDF by email. Get the Half Pack →

About the H12-891 exam

These 122 Huawei H12-891 HCIE-Datacom questions are free here. The free PDF has 154, free by email.

They are drawn from the same bank as the full H12-891 pack, which has 572 questions in total.

Back to blog