Answers and Explanations

A1:	Answer: C. Your environment has non-Cisco routers. Although some newer vendors support Cisco's version of frame relay encapsulation, most do not. You have to use the standards-based IETF (Internet Engineering Task Force—the forum that oversees the evolution of the Internet architecture and the smooth connectivity across the Internet). A is incorrect, as you have non-Cisco routers. B is incorrect, as Q.933a is an LMI signaling standard developed by the CCITT (Comité Consultatif International Téléphonique et Télégraphique—the forum that defines many standards for data communication). D is incorrect, as ANSI (American National Standards Institute—an organization that administers the voluntary standardization of many different standards) is also an LMI signaling standard.
A2:	Answer: C. A DLCI identifies a logical circuit between the router and the frame relay switch. A is incorrect, as your router either autosenses the signaling, or you assign it manually. B is incorrect, as the DLCI does not decide the signaling. D is incorrect, as a DLCI is used by frame relay encapsulation, not used to define the encapsulation.
A3:	Answer: B. The Cisco LMI type is turned on by default. It is not proprietary to Cisco, as Northern Telecom and other companies helped develop it. A is incorrect, as the ANSI T1.617.D is not the default LMI type. D is incorrect, as Q.933a is not the default type. E is incorrect, as modified cut-through is a frame-forwarding convention.
A4:	Answer: A. Typically, you have different LMI types on different vendors. B is incorrect, as the LMI provides signaling between the router and the frame relay switch. It also assigns the DLCI. C is incorrect, as frame relay does not handle authentication. D is incorrect, as you typically have Inverse ARP handling IP address mapping.
A5:	Answer: D. The problem with split-horizon is its refusal to broadcast routes out the physical interface it came in on. When logical sub-interfaces are created, the router thinks the sub-interfaces are "separate" interfaces. A is incorrect, as frame relay maps help map DLCIs to IP addresses. B is incorrect, as there are no such things as frame relay proxies. C is incorrect, as LMI is a signaling standard similar to keepalives.
A6:	Answer: A. Frame relay uses packet-switched technology. B is incorrect, as circuit-switched technology is used by asynchronous serial and ISDN connections. C is incorrect, as leased lines are dedicated TDM lines that are given dedicated bandwidth and dedicated connections. D is incorrect, as LANs do not run frame relay.
A7:	Answer: D. Frame relay primarily works at Layer 2 of the OSI Model, which is the Data-link layer. A is incorrect, as frame relay is not involved with the Application layer. B is incorrect, as frame relay is not involved at the Transport layer. C is incorrect, as frame relay is not involved at the Session layer. E is incorrect as well; although frame relay does have involvement at the Physical layer, encapsulation does not occur there.
A8:	Answer: B. The Local Management Interface (LMI) is responsible for maintaining the connection between the router and the frame relay switch. A is incorrect, as the DLCI is responsible for identifying the virtual circuit between a router and the frame relay switch. C is incorrect, as the permanent virtual circuit is the logical path that data travels over in a frame relay environment. D is incorrect, as the Internet Engineering Task Force is not a signaling standard; it's a standards committee.
A9:	Answer: E. A full mesh topology has virtual circuits to all other routers involved in the topology. A is incorrect, as the star topology has a central router that has single virtual circuits to the other routers. B is incorrect, as the hub-and-spoke topology is the same as a star topology. C is incorrect, as the bus topology is sometimes called peer-to-peer and has a connection between two routers. D is incorrect, as a partial mesh topology has some redundant links, but not to every destination.
A10:	Answer: D. Split-horizon prevents routing updates (broadcasts) from going back out the interface they came in on. If you have multiple virtual circuits coming in on a physical interface, this causes issues. A is incorrect, as route poisoning still functions. B is incorrect, as hold-down timers do not affect frame relay. C is incorrect, as triggered updates still occur.
A11:	Answer: B. Static map statements need to be entered if Inverse ARP is not supported or is not working correctly. A is incorrect, as static routes do not give you correct DLCI-to-Layer 3 address mappings. C is incorrect, as DHCP assigns IP addresses to host interfaces. D is incorrect, as there is no such thing as LMI mapping. LMI does assist in the Inverse ARP process, however.
A12:	Answer: D. The Cisco IOS command is frame relay map [protocol] [network address] [dlci number] [broadcasts allowed or not (optional)]. A is incorrect, as you do not specify the DLCI before the IP address. B is incorrect, as RARP is not involved. C is incorrect, as the LMI is not involved in the frame relay map command.
A13:	Answer: C. The command shows all the current entries, dynamic or static, in the frame relay map table. A is incorrect, as there is no such command. B is incorrect, as this command shows you the routing table entries. D is incorrect, as this command shows you an overview of settings on all IP interfaces.
A14:	Answers: C and D. IPX and AppleTalk do not allow you to turn off split- horizon, and split-horizon is in place to prevent routing loops from occurring. A is incorrect, as you can turn off split-horizon in an IP network. B is incorrect, as routing updates are still transferred.
A15:	Answer: C. You do not want to have an IP address configured on the physical interface. A is incorrect, as you do need to specify frame relay on the physical interface. B is incorrect, as each sub-interface needs a DLCI to identify the virtual circuit assigned to the sub-interface. D is incorrect, as you must specify whether a sub-interface is multipoint or point-to-point.
A16:	Answers: A, B, and C. Forward Explicit Congestion Notifications (FECNs) and Backward Explicit Congestion Notifications (BECNs) are used to notify routers of congestion occurring on the frame relay circuit. Discard Eligible (DE) bits are used to mark packets that exceed the CIR and are dropped if congestion occurs. D is incorrect, as LMI maintains the circuit between the router and the frame relay switch. E is incorrect, as the Committed Information Rate does nothing to control congestion but merely states the promised bandwidth from the service provider.
A17:	Answer: A. LMI is the protocol used between a DCE and DTE to manage the connection. Signaling messages for SVCs, PVC status messages, and keepalives are all LMI messages. B is incorrect, as BECN is the bit in the frame relay header that signals to anyone receiving the frame that congestion is occurring in the backward direction of the frame. Switches and DTEs can react by slowing the rate by which data is sent in that direction. C is incorrect, as FECN is the same as BECN, only in a different direction. D is incorrect, as DLCI is the frame relay address used in the headers to identify the virtual circuit.
A18:	Answer: C. Frame Relay Traffic Shaping (FRTS) allows you to control the flow of traffic over your virtual circuits. It is included as part of the Cisco IOS and does not cost any additional money. A is incorrect, as you already have a slow link at the branch office. B is incorrect, as the criteria state you cannot spend more money, and faster links require more money. D is incorrect as well because you have to purchase each PVC, and this would require spending more money.
A19:	Answer: A. The Committed Information Rate is the value that sets the maximum data rate that the frame relay network tries to deliver under "normal conditions." B is incorrect, as this identifies the DLCI. C is incorrect, as this defines Inverse ARP. D is incorrect, as it defines LMI.
A20:	Answer: D. You have added two offices to your existing eight, bringing the total number of nodes to 10. Using the formula of n(n – 1)/2 to figure out the total number of PVCs, you arrive at 45. A is incorrect, as this works only for a star topology. B is incorrect, as this is only a partial mesh. C is incorrect; this provides only a partial mesh, as there are not enough connections for a full mesh.
A21:	Answer: B. Frame relay improved on X.25 by relying on applications, or upper layers, to provide error correction and reliability of transmission. A is incorrect, as frame relay is faster than X.25 because of the lack of overhead. C is incorrect, as X.25 is packet-switched as well. D is incorrect, as both X.25 and frame relay are standards-based.
A22:	Answer: D. Switched virtual circuits (SVCs) are temporary connections. A is incorrect, as VCs are multiplexed logical data conversations. B is incorrect, as SVCs are not widely supported by all frame relay providers. C is incorrect, as PVCs are used for always-on permanent connections.
A23:	Answers: A, B, and D. A PVC must be in place with LMI operating correctly. To use Inverse ARP, you must be running Cisco IOS 11.2 or higher. C is incorrect, as you do not need your routing protocol to support Inverse ARP for it to work.
A24:	Answer: A. The frame relay map command allows you to specify a different encapsulation (IETF or Cisco) to be used rather than the general encapsulation given in the interface command encapsulation frame relay [ietf/cisco]. B is incorrect, as you are using Cisco equipment elsewhere in your environment, and it is better to use the Cisco version of frame relay if possible. C is incorrect, as it is unnecessary to purchase new equipment. D is incorrect, as there is no such thing as a frame relay proxy.
A25:	Answer: A. The correct syntax of the frame relay map command is frame-relay map ip [remote_ip_address] [local_dlci] broadcast. The broadcast keyword allows routing protocol updates to function. In this case, Arizona is trying to reach the remote IP address 10.1.1.2 in Michigan and uses the local DLCI of 512 to get there. All other answers are incorrect because they use either the wrong DLCI or IP address.
A26:	Answer: A. Three primary PVC states indicate the status of the line. ACTIVE means there are no problems. Answer B is incorrect, as INACTIVE means that there is a problem with the remote router; DELETED means that there is a problem with your local router. Typically, this is caused by using the incorrect DLCI information. Answer D is eliminated because multipoint and point-to-point designs use DLCI information in the same way. If the DLCI shows up as DELETED under a multipoint configuration, it shows up as DELETED under a point-to-point configuration. Finally, if you are physically disconnected from the service provider, you do not see DLCI information (because LMI is used to send the DLCI status to your router); thus, answer C is incorrect as well.
A27:	Answer: B. The show interface serial [port] command gives you information about the Physical and Data-link layer on your interface. A is incorrect, as this command displays the status of each configured connection, but more importantly, it shows the traffic statistics. C is incorrect, as there is no such command. D is incorrect, as this command gives you the existing configuration of the router that is currently running in RAM.
A28:	Answer: A. The circuit number is the local loop identifier given by the service provider. It is typically on the label attached to your CSU/DSU. B is incorrect, as the DLCI identifies the virtual circuit to the router and is given by the LMI. C is incorrect, as the LMI is the signaling technology that maintains the virtual circuits. D is incorrect, as the CIR is the minimum agreed bandwidth between the provider and your company.
A29:	Answer: C. The output shown is from the show frame relay lmi command. A is incorrect, as this command is ambiguous and needs another parameter. B is incorrect, as this command displays the status of each configured connection with traffic statistics. D is incorrect, as there is no such command.
A30:	Answer: D. The debug frame relay lmi command gives you the output shown. A is incorrect, as it is ambiguous and requires another parameter. B is incorrect, as this command displays the LMI traffic statistics in a static manner. C is incorrect, as it gives you all frame relay statistics in real time, not just the output shown. E is incorrect, as you never issue this command on a production router.
A31:	Answer: A. LMI multicasts are sent every 10 seconds by default. B, C, and D are incorrect timers for LMI signaling.
A32:	Answers: B and C. FRF.5 allows for frame relay to cross ATM networks. FRF.8 allows a frame relay user to communicate with an ATM user. A is incorrect, as FRF.1 defines UNI implementations. D is incorrect, as FRF.11 defines voice-over-frame relay.
A33:	Answer: A. Both the ITU-T and ANSI groups help define frame relay. B is incorrect, as frame relay operates at the lower two layers of the OSI Model. C is incorrect, as frame relay allows for star, partial mesh, and full mesh topologies. D is incorrect, as VLANs are a Layer 2 Switch capability to segment your network into logical sub-networks or virtual LANs.
A34:	Answer: D. Frame relay's core aspects function at the two lowest layers of the OSI Reference Model: Physical and Data-link. A is incorrect, as frame relay does not have aspects at all layers. B is incorrect, as well, because despite frame relay relying on upper layers for functionality of data, it does not operate there. C is incorrect, as frame relay does not operate at the Session and Transport layers of the OSI Model.
A35:	Answer: B. The show frame relay pvc command provides you with statistics of each configured connection, as well as traffic statistics. A is incorrect, as the output is not for LMI statistics. C and D are incorrect, as there are no such commands.
A36:	Answer: C. You need only a connection from the hub to each of the spokes. A is incorrect, as each router needs a separate connection for each and every other router in the topology. This would be quite expensive in a large topology. B is incorrect, as you still have more connections than a hub-and-spoke topology. D is incorrect, as nonbroadcast multi-access (NBMA) does not affect the cost of frame relay.
A37:	Answer: D. A service provider marks extra frames with a Discard Eligible (DE) bit in the header that allows the service provider to discard frames if necessary when there is congestion. A is incorrect, as the service provider marks extra traffic. B is incorrect, as providers allow occasional "bursting," as long as it doesn't affect the network. C is incorrect, as service providers typically do not charge extra fees when there is bursting, depending on the type of contract you have.
A38:	Answer: A. A frame relay switch sends Forward Explicit Congestion Notification (FECN) marked frames to the destination DTE if there is congestion on the frame network. B is incorrect, as BECNs are sent back to the source DTE after FECNs are received. C is incorrect, as Discard Eligible (DE) frames are discarded if the network becomes too congested. D is incorrect, as the CIR sets the maximum average data rate that the network undertakes to deliver under "normal conditions."
A39:	Answers: A, B, D, and E. For a frame relay connection to be in place, you must have a virtual circuit identified by a DLCI. The DLCI is given to you via LMI, which also maintains the circuit. The service provider wants a CIR for each of the virtual circuits. C is incorrect, as you do not need Inverse ARP if you use static frame relay maps.
A40:	Answer: C. The cable that connects you to the DCE is specified by the service provider. It can be a form of EIA/TIA, V.35, or X.21. A is incorrect, as the signaling standard is specified by the LMI. B is incorrect, as the encapsulation for crossing the cloud is either IETF or Cisco. D is incorrect, as Cisco routers use a DB-60 connector on their end.
A41:	Answer: A and D. DTE is on the customer end and is the data terminal equipment. The DCE is on the service provider's equipment, typically called data communication equipment. It's usually a device like a CSU/DSU. B and C are incorrect, as these describe equipment not used in frame relay or any serial communication. E is incorrect as well; even though the DCE does provide clocking, it is not called data clocking equipment. F is also incorrect; although these devices help provide the logical circuits, they are not called data circuit equipment.
A42:	Answers: A, B, and C. A is correct, as you can use a full mesh topology, which solves the issue. B is correct, as you can use sub-interfaces. C is also correct; however, you must be running a routing protocol that uses split-horizon. D is incorrect because RIPv2 still has split-horizon mechanisms, although it does improve over RIPv1 by using multicasts to update other routers and is a classless routing protocol.
A43:	Answer: C. Sub-interfaces allow you to overcome split-horizon issues if you use point-to-point sub-interfaces. A is incorrect, as you can use more than one physical interface if necessary for more frame relay connections. B is incorrect, as sub-interfaces don't require you to use more than one physical interface. D is incorrect, as you still need to assign IP addresses to the sub-interfaces. In fact, you increase the number of IP addresses.
A44:	Answer: B. A virtual circuit is "inactive" when the remote side is not sending, but you are okay on your side of the connection. A is incorrect, as this represents a working connection, or active state. C is incorrect, as this represents a deleted state. D is incorrect, as a shutdown interface shows up as administratively down.
A45:	Answer: D. This describes a virtual circuit in an active state. A is incorrect, as an inactive state shows a problem on the opposite side of the connection. B is incorrect, as this shows a circuit that is not receiving any LMI signaling. C is incorrect, as this is a port state in Spanning-Tree Protocol.
A46:	Answers: A, B, and D. They are all valid signaling types for Local Management Interface (LMI). C is incorrect, as the ITU-T is a standards organization that actually created the Q.933a standard.
A47:	Answer: A. Without the broadcast keyword, broadcast and multicast-based routing updates are not sent across the link. Answer B is incorrect because the RIP version is irrelevant to getting RIP working across frame relay. Answer C is incorrect because Inverse ARP has nothing to do with getting your routing updates across a frame relay network. Finally, answer D is incorrect because RIP is activated on the interfaces. When you enter in the network statements under the RIP configuration mode, it automatically enables RIP on the interfaces where those networks reside.
A48:	Answer: B. Routers send Inverse ARP messages every 60 seconds by default. The rest of the answers are not the default timer values for Inverse ARP messages.
A49:	Answer: C. The frame relay switch checks the incoming frame's inbound DLCI number. A is incorrect, as it is the last step. B is incorrect, as it is part of the last step. D is incorrect, as it is the second step.
A50:	Answer: A. There are redundant links between the routers; however, there are not redundant links between all of the routers. B is incorrect, as there are not redundant links between all of the routers. C is incorrect, as even though the connections form a ring as shown in the figure, if it is in the cloud, it does not look like this. D is incorrect, as it is not a star or hub-and-spoke topology; there is no central point