This chapter covers the following subjects:
• Understanding the Cisco IOS Software DMVPN architecture: Covers the integration of several technologies to deploy DMVPNs to hub and spoke routers.
• Plan the deployment of a Cisco IOS Software DMVPN: Discusses the environmental information that you must analyze to successfully deploy a DMVPN.
• Configure and verify Cisco IOS Software GRE tunnels: Learn to configure GRE tunnels and integrate them with other Cisco technologies to create secure communication paths.
• Configure and verify a Cisco IOS Software NHRP client and server: Covers the deployment of NHRP on both the client (spoke router) and server (hub router).
• Configure and verify a Cisco IOS Software DMVPN hub: Discusses successfully deploying DMVPNs on the hub router.
• Configure and verify a Cisco IOS Software DMVPN spoke: Discusses successfully deploying DMVPNs on the spoke routers.
• Configure and verify dynamic routing in a Cisco IOS Software DMVPN: Covers adding dynamic routing to a DMVPN.
• Troubleshoot a Cisco IOS Software DMVPN: Discusses troubleshooting the configuration of a DMVPN.
Dynamic Multipoint Virtual Private Networks (DMVPN) are a feature of Cisco IOS Software that makes the deployment of large hub-and-spoke, partial mesh, and full mesh VPN topologies much easier.
The “Do I Know This Already?” quiz helps you decide whether you really need to read the entire chapter. If you already intend to read the entire chapter, you do not necessarily need to answer these questions now.
The ten-question quiz, derived from the major sections in the “Foundation Topics” portion of this chapter, helps you determine how to spend your limited study time.
Table 17-1 outlines the major topics discussed in this chapter and the “Do I Know This Already?” quiz questions that correspond to those topics.
Table 17-1. “Do I Know This Already?” Foundation Topics Section-to-Question Mapping

The goal of self-assessment is to gauge your mastery of the topics in this chapter. If you do not know the answer to a question or are only partially sure of the answer, you should mark this question wrong for purposes of the self-assessment. Giving yourself credit for an answer that you correctly guess skews your self-assessment results and might provide you with a false sense of security.
1. Which mechanism provides a scalable multiprotocol tunneling framework with optional dynamic routing?
a. NHRP
b. IPsec
c. GRE
d. 802.1X
e. None of these answers are correct.
2. Which mechanism provides dynamic mutual discovery of spoke devices?
a. GRE
b. IKE
c. NHRP
d. DHCP
e. Expired Certificate List
3. Which mechanism provides key management and transmission protection?
a. NHRP
b. GRE
c. mGRE
d. IDS/IPS
e. IKE + IPsec
4. To integrate PKI-based authentication with site-to-site VPNs, which protocol must be configured to use PKI-based authentication?
a. IKE
b. GRE
c. AAA
d. RSA
e. VPN
5. DMVPNs can use pre-shared keys or PKI-based IKE authentication. Either choice is acceptable for a hub-and-spoke network, but which of the following is recommended for a fully meshed network?
a. IPsec
b. DH group 14
c. Pre-shared keys
d. PKI-based authentication
6. GRE uses which IP protocol in combination with IPsec VPNs to pass routing information between connected networks?
a. 89
b. 50
c. 47
d. 51
e. None of these answers are correct.
7. When a spoke router initially connects to a DMVPN, it registers its inner (tunnel) and outer (physical interface) IP address with which of the following?
a. NHRP server
b. DHCP server
c. Cisco ACS Server
d. Cisco Security Manager
e. None of these answers are correct.
8. What Cisco IOS Software command designates the tunnel interface as multipoint GRE mode?
a. tunnel source
b. tunnel destination
c. tunnel mode gre multipoint
d. interface gre 0/0 multipoint
9. If a DMVPN spoke router is configured with a point-to-point GRE interface, the spoke will only participate in which type of topology?
a. Strict hub-and-spoke
b. Partial mesh
c. Full mesh
d. Token ring
e. None of these answers are correct.
10. On the hub, what is the main factor that determines whether the DMVPN will operate as strict hub-and-spoke or as partially/full mesh?
a. Routing protocol functions
b. Network administrator preference
c. Bandwidth to the hub
d. Cisco router hardware model
e. Cisco IOS Software Release
The answers to the “Do I Know This Already?” quiz are found in Appendix A. The suggested choices for your next step are as follows:
• 8 or less overall score: Read the entire chapter. This includes the “Foundation Topics” and the “Exam Preparation” sections.
• 9 or 10 overall score: If you want more review on these topics, skip to the “Exam Preparation” section. Otherwise, move on to Chapter 18, “Deploying High Availability in Tunnel-Based IPsec VPNs.”
This chapter covers the integration of multipoint Generic Routing Encapsulation (mGRE) with Next Hop Resolution Protocol (NHRP) and IPsec to deploy Dynamic Multipoint Virtual Private Networks (DMVPN) on hub-and-spoke devices. Verification and troubleshooting of DMVPN operation will also be covered. Upon completing this chapter, you will be able to configure and troubleshoot a Cisco IOS Software DMVPN.
The Cisco DMVPN feature allows administrators to deploy scalable IPsec VPNs for both small and large networks. The Cisco DMVPN feature combines the features and benefits of mGRE tunnels, IPsec encryption, and the Next Hop Resolution Protocol (NHRP) to provide a solution that easily provisions VPN peers. DMVPN also supports dynamically addressed spoke routers as long as appropriate peer authentication is used.

DMVPNs can be deployed using two models:
• Hub-and-spoke: A hub-and-spoke DMVPN requires that each branch (spoke) have a point-to-point GRE interface that is used to build a tunnel to the hub router. All traffic between spokes must flow through the hub router. This model provides a scalable configuration on the hub router but does not provide direct spoke-to-spoke communication.
• Spoke-to-spoke: A spoke-to-spoke DMVPN requires that each branch (spoke) have an mGRE interface through which dynamic spoke-to-spoke tunnels are used for spoke-to-spoke traffic. This model provides a scalable configuration for all involved devices and also provides direct spoke-to-spoke communication. Be aware that DMVPN does not immediately produce a mesh (partial or full) topology. It initially establishes a hub-and-spoke topology from which a partial or full mesh is generated dynamically as traffic patterns dictate.

The primary benefits of implementing a DMVPN are as follows:
• Hub router configuration reduction: Traditional IPsec VPNs involved configuring the crypto map characteristics, the crypto access control list (ACL), and the GRE tunnel interface on each spoke router. The DMVPN feature allows you to create a single mGRE tunnel interface and a single IPsec profile, and does not require crypto ACLs on the hub router. This makes the size of the configuration on the hub router relatively the same, even as spoke routers are added to the network.
• Automatic IPsec initiation: GRE relies on NHRP to configure and resolve peer destination addresses. This allows IPsec to immediately trigger to dynamically create point-to-point GRE tunnels and begin negotiation of IPsec sessions without traditional IPsec configuration.
• Support for spoke routers configured with dynamic addressing: Traditional IPsec hub-and-spoke VPNs required that all sites be configured with static IP addresses on spoke router interfaces because that information is required when configuring the GRE tunnel destination IP address and the IPsec peer address. DMVPNs allow spoke routers to have dynamic interface IP addresses and to use NHRP to register the spoke router’s interface IP address with the hub router.
DMVPNs use the benefits of other Cisco features to provide the best of hub-and-spoke and full mesh VPN topologies. The Cisco features that DMVPNs combine are as follows:
• mGRE: mGRE allows a single Generic Routing Encapsulation (GRE) interface to support multiple GRE tunnels and makes the configuration much easier. Using GRE tunnels provides support for IP multicast and non-IP protocols to traverse the interface as well. Support for IP multicast enables the use of dynamic routing protocols such as Enhanced IGRP (EIGRP) or Open Shortest Path First (OSPF) to update routing tables and create redundant VPN paths if needed. All DMVPN members use either GRE or mGRE interfaces to build tunnels between other peer devices.
• NHRP: Next Hop Resolution Protocol (NHRP) is a client and server protocol where the hub acts as the NHRP server and the spokes are the NHRP clients. The NHRP database maintains mappings between the router (public, physical interface) and the tunnel (inside the tunnel interface) IP addresses of each spoke. Each spoke registers its public and internal tunnel addresses when it boots and queries the NHRP database for the addresses of other spokes. NHRP is used to make the configuration and management of a partial or full mesh environment less complex, like that found in a hub-and-spoke network.
• IPsec: IPsec provides transmission protection for GRE tunnels. DMVPNs form a permanent hub-and-spoke IPsec VPN that can dynamically reconfigure itself into a partial or full mesh configuration as needed.
The DMVPN cloud topology can support either a hub-and-spoke or a spoke-to-spoke deployment model. The hub-and-spoke deployment model requires that each hub router be configured with an mGRE interface and each branch router (spoke) be configured with a point-to-point GRE interface. The spoke-to-spoke deployment model requires that both the hub router and branch routers (spokes) are configured with mGRE interfaces to forward traffic to peers. A DMVPN cloud is a collection of routers that are configured with either an mGRE interface or a point-to-point GRE interface (or a combination of the two) and that share the same subnet.
Because the typical VPN is used to secure private or corporate traffic over a shared public infrastructure, like the Internet, there are two distinct IP address domains:
• The enterprise addressing space, which is sometimes referred to as the private or inside addresses. These are the addresses that are assigned to the (m)GRE interfaces and are registered with the NHRP server by each spoke.
• The infrastructure addressing space, which is sometimes referred to as the service provider, public, or outside addresses. These are the addresses on the router’s physical interfaces and used as addresses in tunnel envelopes (the source and destination addresses in the GRE/IPsec packet headers).
Figure 17-1 illustrates a DMVPN where the top router (hub) and the other three spoke routers are configured to connect to the hub using NHRP, GRE, and IPsec. Initially, all spokes build a GRE/IPsec tunnel with the hub and register with NHRP on the hub.
Figure 17-1. DMPVN: Hub-and-Spoke Model


In its initial state, the network is purely hub-and-spoke and can stay that way if desired. The initial network properties are
• The hub knows the outer and inner IP addresses of each spoke in its NHRP database.
• Three spoke-to-hub GRE/IPsec tunnels are created.
• Any traffic from a spoke (whether to a hub or another spoke) must travel through the hub.
In a spoke-to-spoke DMVPN topology, spokes can locate each other and dynamically build tunnels directly to each other.

In Figure 17-1, consider a PC on R2’s inside network sending traffic to a server on R4’s internal network:
Step 1: Dynamic routing (already configured over the permanent hub-to-spoke GRE/IPsec tunnels) populates each spoke’s routing table so that each spoke knows about the subnets behind the other spokes.
Step 2: R2 consults its routing table for a route to the subnet behind R4. The routing table provides a next-hop IP address of R4’s tunnel IP.
Step 3: R2 consults its local NHRP cache for R4’s tunnel IP address and finds no entry. It sends an NHRP query to the NHRP server (the DMVPN hub router) to resolve the inner (tunnel) address to an external physical IP address on R4.
Step 4: The NHRP server, which maintains inner (tunnel) and outer (physical) addresses for spoke routers, sends an NHRP reply to R2, which informs it that the inner tunnel IP on R4 is reachable through its outer (physical) IP address.
Step 5: R2 receives the response from the server and enters it into the local NHRP cache. IPsec is triggered on R2 to create an IPsec tunnel directly to the outer (physical) address on R4. R2 initiates an IKE session to the outer (physical) address of R4 using its outer (physical) address and then negotiates IPsec security associations (SA) for a spoke-to-spoke GRE tunnel.
Step 6: After the IPsec tunnel is created, R2 will create a GRE tunnel to the outer (physical) address on R4, and all traffic between R2 and R4 will now flow through this tunnel with no more participation by the hub router.
Step 7: At this point, only traffic from R2 to R4 (one direction) has been enabled.
Step 8: For two-way traffic, R4 needs next-hop information for the subnet behind R2. R4 queries the NHRP server. After it receives the NHRP response from the NHRP server, the return path is mapped to the newly built GRE tunnel and the response traffic is now sent directly to R2, encapsulated in the GRE/IPsec tunnel.
When the NHRP entries age out, IPsec will be triggered to tear down the dynamically built spoke-to-spoke tunnel. Timers can be tuned to have these tunnels tear down sooner.
Table 17-2 outlines the benefits and limitations of DMVPNs.

Table 17-2. DMVPN Benefits and Limitations

A successful DMVPN deployment requires that you gather information about your current environment. The following sections discuss gathering and analyzing that information.
Information to be gathered as input when deploying a DMVPN includes the following:
• Hardware and software on existing routers: This is needed to determine the performance of the routers and whether the Cisco IOS Software release supports DMVPN features.
• Cryptographic standards and security requirements: Information concerning local security requirements and security policies will dictate information such as cryptographic algorithms, key lengths, and tunneling protocols.
Deploying a DMVPN requires completing the following configuration tasks:
Task 1. Configure Internet Key Exchange (IKE) sessions between each DMVPN spoke and the hub, including Internet Security Association and Key Management Protocol (ISAKMP) policies and authentication information.
Task 2. Configure NHRP sessions between each DMVPN spoke and the hub.
Task 3. Configure mGRE tunnels and IPsec profiles on the DMVPN hub.
Task 4. Configure GRE (for pure hub-and-spoke DMVPNs) or mGRE (for partial or full mesh DMVPNs) tunnels and IPsec profiles on DMVPN spokes.
Task 5. Configure dynamic routing over DMVPN tunnels.
The process of configuring DMVPNs comes with several deployment choices. The deployment choices and helpful information is provided next:
• Pre-shared keys or PKI-based IKE authentication: Either choice is acceptable for a hub-and-spoke, but public key infrastructure (PKI)–based authentication is recommended for scalability in an environment with fully meshed DMVPNs.
• Single or multiple hubs: This depends on redundancy requirements. Dual hubs and links can provide both device and path redundancy.
• Hub-and-spoke or full mesh topology: Use full mesh if there is a significant amount of spoke-to-spoke traffic; otherwise, consider a hub-and-spoke topology.

Consider the following deployment guidelines for DMVPN implementations:
• Use DMVPNs to deploy large hub-and-spoke or full mesh VPN networks.
• Use dynamic routing protocols to route traffic across the DMVPNs.
• DMVPN is a suitable technology for providing secure paths over insecure networks such as the Internet because it hides internal IP addressing.
Generic Routing Encapsulation (GRE) was designed to carry multiprotocol and IP multicast traffic between sites that might not have IP connectivity. Cisco developed GRE as a protocol that can encapsulate a wide variety of protocol packet types inside IP tunnels.
GRE uses IP protocol 47 and is used with IPsec VPNs to pass routing information between connected networks. A tunnel interface supports header types for each of the following:
• A passenger protocol or encapsulated protocol, such as IP, AppleTalk, DECnet, or IPX
• A carrier protocol (GRE in this case)
• A transport protocol (IP only in this case)
GRE has many useful features and is supported by many vendors, so interoperability is usually not an issue for basic GRE point-to-point tunnels. GRE tunnels support multiple network protocols and IP multicast. One of the significant advantages of GRE tunneling over (non-VTI) IPsec tunnels is that GRE uses Cisco IOS Software interfaces that can utilize QoS features.

GRE does have some limitations:
• GRE provides no cryptographic protection for traffic and must be combined with IPsec to provide it.
• There is no standard way to determine the end-to-end state of a GRE tunnel. Cisco IOS Software provides proprietary GRE keepalives for this purpose.
• It can be CPU intensive on some platforms.
• Tunnels can sometimes cause maximum transmission unit (MTU) and IP fragmentation-related issues.
Most systems installed today leave the default MTU set to 1500 bytes and, in most cases, generate packets at this size limit. The tunneling adds a 24-bit GRE header to the packet, which then makes the packet larger than the MTU size. If an application sets the Don’t Fragment (DF) bit, this causes a problem when the tunneling system causes the packet to be fragmented.
GRE tunnels can be deployed as point-to-point tunnels or point-to-multipoint GRE tunnels. This section provides a comparison of the two deployment options.

The main characteristics of GRE point-to-point configurations are as follows:
• They are typically used as simple point-to-point tunnels that act like point-to-point WAN links or as the connections from spoke to hub router in hub-and-spoke VPNs.
• There is a separate GRE tunnel interface configured on each peer for each GRE peer.
• They do not require NHRP because other peers (their destination tunnel address) are statically configured.
• They provide support for unicast, multicast, and broadcast traffic.
The main characteristics of GRE point-to-multipoint configurations are as follows:
• Only one tunnel interface must be configured for a router to support multiple GRE peers. For a hub-and-spoke network, a single mGRE tunnel interface on the hub can be used for many spoke peers.
• Devices taking advantage of mGRE interfaces require the use of NHRP to build dynamic GRE tunnels. This also provides the option to have peers (usually spokes) that are dynamically addressed.
• mGRE interfaces also support unicast, multicast, and broadcast traffic.
Figure 17-2 shows a DMVPN with both types of GRE tunnels. The solid lines are point-to-point, hub-and-spoke GRE tunnels, and the dashed arrow shows how mGRE permits tunnels to be dynamically built between two spokes. Using mGRE on the spokes as well combined with NHRP provides a topology in which devices can dynamically create partial or full meshed VPN networks.
Figure 17-2. GRE Point-to-Point and Point-to-Multipoint Tunnels

Example 17-1 provides a sample configuration for a basic point-to-point GRE tunnel. Figure 17-3 shows the example.
Example 17-1. Configure Basic Point-to-Point GRE Tunnel
R1(config)# interface tunnel0
R1(config-if)# tunnel mode gre ip
R1(config-if)# tunnel source 192.168.0.1
R1(config-if)# tunnel destination 192.168.0.2
R1(config-if)# ip address 10.0.0.1 255.255.255.255
R2(config)# interface tunnel0
R2(config-if)# tunnel mode gre ip
R2(config-if)# tunnel source 192.168.0.2
R2(config-if)# tunnel destination 192.168.0.1
R2(config-if)# ip address 10.0.0.2 255.255.255.255
Figure 17-3. Point-to-Point Tunnel Configuration Example

To configure GRE tunneling in hub-and-spoke GRE networks, complete the following tasks:
Task 1: Configure a multipoint GRE interface on the hub.
Task 2: Configure a point-to-point GRE interface on each spoke that connects to the hub mGRE interface.
The configuration scenario involves creating a hub-and-spoke network with an mGRE interface on the hub and a point-to-point GRE interface on the spoke. The hub has an outer physical IP address of 172.17.0.1 and a tunnel IP address of 10.1.1.1. The spoke has an outer physical IP address of 172.17.2.4 and a tunnel IP address of 10.1.1.2.
Task 1 is to create a Cisco IOS Software tunnel interface using the interface tunnel global configuration command and make it an mGRE interface with the tunnel mode gre multipoint command. The task also includes enabling the IP process on the interface by giving it an IP address with the ip address command. Also, you need to specify the tunnel endpoints with the tunnel source and tunnel destination commands, as shown in Example 17-2.
Using the multipoint interface requires a subnet large enough to handle all the spoke tunnels because the multipoint and the spoke tunnel interfaces will be a part of the same logical IP subnet.

Example 17-2. [Hub] Configure an mGRE Interface
Hub(config)# interface tunnel0
Hub(config-if)# tunnel mode gre multipoint
Hub(config-if)# tunnel source 172.17.0.1
Hub(config-if)# tunnel destination 172.17.2.4
Hub(config-if)# ip address 10.1.1.1 255.255.0.0
Task 2 is to create a Cisco IOS Software GRE tunnel interface as a partner to the hub’s multipoint GRE interface using the interface tunnel global configuration command and the tunnel mode gre ip command. The task also includes enabling the IP process on the interface by giving it an IP address that is included in the hub’s IP subnet with the ip address command. Also, you need to specify the tunnel endpoints with the tunnel source and tunnel destination commands. Example 17-3 shows the commands in use.

Example 17-3. [Spoke] Configure a GRE Interface
Spoke(config)# interface tunnel0
Spoke(config-if)# tunnel mode gre ip
Spoke(config-if)# tunnel source 172.17.2.4
Spoke(config-if)# tunnel source 172.17.0.1
Spoke(config-if)# tunnel destination 172.17.0.1
After a hub-and-spoke GRE tunnel is configured, the show interface tunnel command can be used to verify its state. The line protocol of a GRE tunnel is up as long as there is a route to the tunnel destination. This is not the case if GRE keepalives are configured (not recommended if using dynamic routing protocols). The line protocol on mGRE tunnel interfaces is always up. Example 17-4 shows the output from a tunnel interface in the up/down state.
Example 17-4. show interface tunnel Command Output
Router# show interfaces tunnel 0
Tunnel4 is up, line protocol is down
Hardware is Routing Tunnel
MTU 1500 bytes, BW 9 Kbit, DLY 500000 usec, rely 255/255, load 1/255
Encapsulation TUNNEL, loopback not set, keepalive set (10 sec)
Tunnel source 0.0.0.0, destination 0.0.0.0
Tunnel protocol/transport GRE/IP, key disabled, sequencing disabled
Last input never, output never, output hang never
Last clearing of "show interface" counters never
Output queue 0/0, 0 drops; input queue 0/75, 0 drops
Five minute input rate 0 bits/sec, 0 packets/sec
Five minute output rate 0 bits/sec, 0 packets/sec
0 packets input, 0 bytes, 0 no buffer
Received 0 broadcasts, 0 runts, 0 giants
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored, 0 abort
0 packets output, 0 bytes, 0 underruns
0 output errors, 0 collisions, 0 interface resets, 0 restarts
To quickly verify that a GRE tunnel is passing traffic, use the ping command and specify the tunnel IP address of the remote GRE peer. Also, observe routing protocol neighbor relationships. If they have established successfully, the tunnel must be passing traffic. Note that both methods require NHRP because mGRE is in use.
NHRP is used by routers to determine the IP address of the next hop in IP tunneling networks. When a spoke router initially connects to a DMVPN, it registers its inner (tunnel) and outer (physical interface) IP address with the hub router (the NHRP server) whose IP address is configured on the spoke router (NHRP client).
The hub router places an entry in its NHRP cache and returns a registration reply. The spoke router now sees the hub router as a valid Next Hop Server (NHS) and queries it when trying to locate any other spokes and networks in the NHRP domain. The NHRP client registration enables the mGRE interface on the hub to build a dynamic GRE tunnel back to the spoke, which in turn creates a mapping between the tunnel IP and physical IP address for each spoke that registers with the hub.
As stated, NHRP maps tunnel IP addresses to outer transport IP addresses. This means that in a hub-and-spoke DMVPN deployment, no GRE or IPsec information about the spoke is configured on the hub router. The spoke router is configured with information about the hub using NHRP commands. When the spoke router starts up, it automatically initiates an IPsec tunnel with the hub router. It then notifies the NHRP server (hub router) of its current physical interface IP address. This notification is beneficial because
• The hub router configuration is drastically shortened and simplified because it does not need to know any GRE or IPsec information about the spoke router. It is learned through NHRP.
• Adding a new spoke router to the DMVPN requires no configuration on the hub. The spoke is configured with the hub information and dynamically registers with the hub router. Dynamic routing protocols distribute information about the spoke to the hub, which in turn propagates the information to the other spokes.

To configure NHRP, complete the following tasks on the hub and spokes in a DMVPN:
Task 1: Configure an NHRP server on the hub.
Task 2: Configure NHRP clients on the network spokes.
The configuration scenario involves creating a basic hub-and-spoke network with an mGRE interface on the hub and a point-to-point GRE interface on the spoke. The hub has an outer physical IP address of 172.17.0.1 and a tunnel IP address of 10.1.1.1. The spoke has an outer physical IP address of 172.17.2.4 and a tunnel IP address of 10.1.1.2. The NHRP network ID is 1.
Task 1 is to create an NHRP server on the hub router. Create a new NHRP server on the tunnel interface using the ip nhrp network-id interface command. The NHRP network ID must be the same on the NHRP server and its NHRP clients. If NHRP is being used over an untrusted network, configure NHRP authentication with the ip nhrp authentication command.
To support dynamic routing protocols, enable support of IP multicast traffic with the ip nhrp map multicast dynamic interface command. This allows each spoke to register as a receiver of multicast traffic, causing the hub to replicate and forward multicast packets to the spoke routers. See Example 17-5 for the command usage.

Example 17-5. [Hub] Configure an NHRP Server
Hub(config)# interface tunnel0
Hub(config-if)# ip nhrp network-id 1
Hub(config-if)# ip nhrp authentication ADFqeqrDA
Hub(config-if)# ip nhrp map multicast dynamic
Task 2 is to create an NHRP client on each spoke router. As configured on the hub, configure the NHRP network ID and NHRP authentication string with the ip nhrp network-id and ip nhrp authentication commands. Then, specify the location of the NHRP NHS with the ip nhrp nhs interface command. To allow the spoke to register its multicast capability with the hub, use the ip nhrp map multicast command. Finally, specify a static NHRP map that enables the spoke to reach the NHRP server over its address. Example 17-6 displays the command sequence being used.

Example 17-6. Configure NHRP Client
Spoke(config)# interface tunnel0
Spoke(config-if)# ip nhrp network-id 1
Spoke(config-if)# ip nhrp authentication ADFqeqrDA
Spoke(config-if)# ip nhrp nhs 10.1.1.1
Spoke(config-if)# ip nhrp map multicast 172.17.0.1
Spoke(config-if)# ip nhrp map 10.1.1.1 172.17.0.1
The show ip nhrp command can be used to show NHRP mapping information for a device, as demonstrated in Example 17-7.
On the hub (NHRP server), this command displays NHRP mappings for all registered spokes.
On the spoke (NHRP clients), this command should show a static mapping for the hub router, a local mapping for the local spoke, and dynamic mappings for any spokes to which the local spoke is sending traffic.
Example 17-7. show ip nhrp Command Output
Router# show ip nhrp
10.0.0.2/32 via 10.0.0.2, Tunnel0 created 00:17:49, expire 00:01:30
Type: dynamic, Flags: unique registered used
NBMA address: 172.17.0.2
Group: test-group-0
The show ip nhrp nhs command displays the NHRP next-hop server information and can be used to show NHRP mapping information for a device, as demonstrated in Example 17-8. The show ip nhrp nhs detail command helps find issues with NHRP registrations. If the req-failed field shows a number greater than 0, there might be a network ID or authentication mismatch. It can also indicate a connectivity problem with the NHS.
Router# show ip nhrp nhs
Legend: E=Expecting replies, R=Responding, W=Waiting
Tunnel0:
192.0.2.1 W priority = 2 cluster = 0
192.0.2.2 RE priority = 0 cluster = 0
192.0.2.3 RE priority = 1 cluster = 0
Router# debug nhrp packet
NHRP activity debugging is on
Router#
NHRP: Send Purge Request via ATM3/0.1, packet size: 72
src: 135.206.58.55, dst: 135.206.58.56
(F) afn: NSAP(3), type: IP(800), hop: 255, ver: 1
shtl: 20(NSAP), sstl: 0(NSAP)
(M) flags: "reply required", reqid: 2
src NBMA: 47.0091810000000002ba08e101.525354555355.01
src protocol: 135.206.58.55, dst protocol: 135.206.58.56
(C-1) code: no error(0)
prefix: 0, mtu: 9180, hd_time: 0
addr_len: 0(NSAP), subaddr_len: 0(NSAP), proto_len: 4, pref: 0
client protocol: 135.206.58.130
NHRP: Receive Purge Reply via ATM3/0.1, packet size: 72
(F) afn: NSAP(3), type: IP(800), hop: 254, ver: 1
shtl: 20(NSAP), sstl: 0(NSAP)
(M) flags: "reply required", reqid: 2
src NBMA: 47.0091810000000002ba08e101.525354555355.01
src protocol: 135.206.58.55, dst protocol: 135.206.58.56
(C-1) code: no error(0)
prefix: 0, mtu: 9180, hd_time: 0
addr_len: 0(NSAP), subaddr_len: 0(NSAP), proto_len: 4, pref: 0
client protocol: 135.206.58.130
To troubleshoot even deeper, use the debug nhrp packet command to see registration failures.
Dynamic Multipoint Virtual Private Networks (DMVPN) are a feature found in Cisco IOS Software that makes it much easier to deploy VPN tunnels. The following sections cover configuring the hub routers in a DMVPN.

Perform the following configuration sequence to configure a DMVPN hub:
Task 1: (Optional) Configure an IKE policy. Recall that if one is not created, the hub router will use default IKE policies.
Task 2: Generate/configure spoke authentication credentials. This typically involves enrolling into a PKI to obtain a certificate for authentication purposes.
Task 3: Configure an IPsec profile with an optional transform set. If a transform set is not configured, the router will use default IPsec transform sets.
Task 4: Create an mGRE tunnel interface.
Task 5: Configure an NHRP server on the mGRE interface.
Task 6: Associate the IPsec profile with the mGRE interface to configure tunnel protection.
Task 7: Configure an IP address and IP fragmentation/segmentation parameters on the mGRE interface.
The configuration scenario involves creating a basic hub-and-spoke network with an mGRE interface on the hub and a point-to-point GRE interface on the spoke. The hub has an outer physical IP address of 172.17.0.1 and a tunnel IP address of 10.1.1.1. The spoke has an outer physical IP address of 172.17.2.4 and a tunnel IP address of 10.1.1.2. The NHRP network ID is 1.
The assumption is that there is full IP connectivity among all physical interfaces.
Task 1 is to optionally create an IKE policy with the crypto isakmp policy global configuration command. Example 17-9 shows an IKE policy specifying Rivest, Shamir, and Adelman (RSA) signatures for authentication and Diffie-Hellman group 14 to protect the key exchange command.

Example 17-9. [Hub] Configure an Optional IKE Policy
Hub(config)# crypto isakmp policy 10
Hub(config-isakmp)# auth rsa-sig
Hub(config-isakmp)# group 14
Task 2 of this configuration sequence specifies that certificates be used for authentication. Refer to the configuration process covered in previous sections of this book and enroll the hub router into the PKI and configure other PKI-relevant items.
Task 3 creates a named IPsec profile using the crypto ipsec profile command that allows the transform set and encapsulation mode to be specified. Example 17-10 shows an empty IPsec profile, which indicates that the default transform sets will be used to protect GRE traffic.
Example 17-10. Configure an IPsec Profile
Hub(config)# crypto ipsec profile IPsec-PROFILE
Task 4 creates the mGRE tunnel interface. Enter the interface tunnel command and then configure basic GRE parameters. The tunnel mode gre multipoint command designates the tunnel interface as mGRE, and the tunnel source command specifies the physical interface to which the GRE tunnel is bound. The tunnel key command is required and must match the tunnel key configured on the spokes. This command allows network administrators to run more than one DMVPN at a time on the same router. The GRE tunnel key therefore uniquely identifies the DMVPN. Example 17-11 shows the sequence of commands.

Example 17-11. Create an mGRE Tunnel Interface
Hub(config)# interface tunnel0
Hub(config-if)# tunnel mode gre multipoint
Hub(config-if)# tunnel source GigabitEthernet0/0
Hub(config-if)# tunnel key 13579
Task 5 configures an NHRP server on the hub router inside the mGRE tunnel interface. The ip nhrp network-id command specifies the NHRP network ID, and the ip nhrp authentication command configures authentication for NHRP traffic. Finally, the ip nhrp map multicast dynamic command registers the multicast capabilities of the device. Example 17-12 shows the sequence of commands.

Example 17-12. Configure the NHRP Server
Hub(config-if)# ip nhrp network-id 1
Hub(config-if)# ip nhrp authentication WERQRQ$
Hub(config-if)# ip nhrp map multicast dynamic
Task 6 specifies the IPsec protection profile created in Step 3 as the tunnel protection mechanism with the tunnel protection ipsec profile command. Example 17-13 shows the profile being applied to the mGRE interface.

Example 17-13. Apply the IPsec Profile to the mGRE Interface
Hub(config)# interface tunnel0
Hub(config-if)# tunnel protection ipsec profile IPsec-PROFILE
Task 7 specifies the IP address for the mGRE tunnel interface. This must be a large enough subnet to accommodate all spoke routers. Optionally, to avoid problems with fragmentation of user packets, set the IP MTU side with the ip mtu command and set the TCP maximum segment size (MSS) value using the ip tcp adjust-mss interface command. Example 17-14 shows the profile being applied to the mGRE interface.

Example 17-14. Configure IP Parameters on the mGRE Interface
Hub(config)# interface tunnel0
Hub(config-if)# ip address 10.1.1.1 255.255.0.0
Hub(config-if)# ip mtu 1400
Hub(config-if)# ip tcp adjust-mss 1360
Cisco Configuration Professional (CCP) allows administrators to avoid the CLI configuration for the DMVPN hub and configure the hub and spoke by manually filling in the settings or by utilizing the DMVPN hub configuration wizard. Refer to the Cisco Configuration Professional User Guide, v2.4 for detailed information about using CCP to configure the DMVPN hub. This guide can be found at www.cisco.com/en/US/docs/net_mgmt/cisco_configuration_professional/v2_4/olh/ccp.pdf.
One of the first troubleshooting steps should be to verify spoke registrations on the hub router. As demonstrated in Example 17-15, the show dmvpn command provides a snapshot view of the spoke registration state.
The output shows entries for all connected spokes and provides the following information:
• The spoke nonbroadcast multiaccess (NBMA) address. This is either the physical interface or a loopback to which the tunnel is attached.
• The spoke tunnel address as it is configured on the GRE or mGRE tunnel.
• The tunnel state.
• The tunnel uptime.
Example 17-15. show dmvpn Command Output
Router# show dmvpn
Legend: Attrb --> S - Static, D - Dynamic, I - Incomplete
N - NATed, L - Local, X - No Socket
# Ent --> Number of NHRP entries with same NBMA peer
! The line below indicates that the sessions are being displayed for Tunnel1.
! Tunnel1 is acting as a spoke and is a peer with three other NBMA peers.
Tunnel1, Type: Spoke, NBMA Peers: 3,
# Ent Peer NBMA Addr Peer Tunnel Add State UpDn Tm Attrb
----- --------------- --------------- ----- -------- -----
2 192.0.2.21 192.0.2.116 IKE 3w0d D
1 192.0.2.102 192.0.2.11 NHRP 02:40:51 S
1 192.0.2.225 192.0.2.10 UP 3w0d S
The show dmvpn detail command reveals additional information about DMVPN peers, as demonstrated in Example 17-16. The detailed output shows IKE SA information and IPsec statistics, such as the number of packets the router has decrypted and encrypted on this tunnel and the target network.
Example 17-16. show dmvpn detail Command Output
Router# show dmvpn detail
Legend: Attrb --> S - Static, D - Dynamic, I - Incomplete
N - NATed, L - Local, X - No Socket
# Ent --> Number of NHRP entries with same NBMA peer
-------------- Interface Tunnel1 info: --------------
Intf. is up, Line Protocol is up, Addr. is 192.0.2.5
Source addr: 192.0.2.229, Dest addr: MGRE
Protocol/Transport: "multi-GRE/IP", Protect "gre_prof",
Tunnel VRF "" ip vrf forwarding ""
NHRP Details: NHS: 192.0.2.10 RE 192.0.2.11 E
Type: Spoke, NBMA Peers: 4
# Ent Peer NBMA Addr Peer Tunnel Add State UpDn Tm Attrb Target Network
----- --------------- --------------- ----- -------- ----- -----------------
2 192.0.2.21 192.0.2.116 UP 00:14:59 D 192.0.2.118/24
IKE SA: local 192.0.2.229/500 remote 192.0.2.21/500 Active
Capabilities:(none) connid:1031 lifetime:23:45:00
Crypto Session Status: UP-ACTIVE
fvrf: (none)
IPSEC FLOW: permit 47 host 192.0.2.229 host 192.0.2.21
Active SAs: 2, origin: crypto map
Inbound: #pkts dec'ed 1 drop 0 life (KB/Sec) 4494994/2700
Outbound: #pkts enc'ed 1 drop 0 life (KB/Sec) 4494994/2700
Outbound SPI : 0xD1EA3C9B, transform : esp-3des esp-sha-hmac
Socket State: Open

Consider the following DMVPN hub implementation guidelines when deploying the DMVPN hub:
• For IKE peering, each hub-to-spoke peering is the same as a classic IPsec VPN peering. However, if using dynamic spoke-to-spoke tunnels, these too need to be authenticated. This necessitates using a scalable authentication method, such as PKI.
• The hub and all spokes must be on the same IP subnet on the tunnel interface. This is required for the routing protocols to advertise the DMVPN as a single IP subnet. This is necessary so that all next-hop addresses appear to directly connect to the tunnel interface.
• If using redundant interfaces on the hub router, the DMVPN can be attached to a loopback interface. The loopback must be included in routing processes in this case.
DMVPN features have been released in phases by Cisco. Basic DMVPN is supported in Release 12.2(13)T and higher. In these releases, spoke-to-spoke traffic must be sent over the hub.
Cisco IOS Software Release 12.3(9)a added support for dynamic spoke-to-spoke tunnels. Reduced latency, hub redundancy, and hierarchical hub design was introduced with Release 12.4(6)T. Release 15.0(1)M provides support for DMVPN tunnel health monitoring and recovery.
DMVPNs are a VPN solution for hub-and-spoke, partial mesh, and full mesh networks that involve much less configuration than traditional VPN implementations. The following sections cover configuring the DMVPN spoke routers.

Perform the following configuration sequence to configure a DMVPN spoke:
Task 1: (Optional) Configure an IKE policy. Recall that if one is not created, the hub router will use default IKE policies.
Task 2: Generate/configure spoke authentication credentials. This typically involves enrolling into a PKI to obtain a certificate for authentication purposes.
Task 3: As with configuring the hub, an IPsec profile with an optional transform set can be configured. If a transform set is not configured, the router will use default IPsec transform sets.
Task 4: Create an mGRE tunnel interface (or a GRE interface for strict hub-and-spoke DMVPNs).
Task 5: Configure NHRP client parameters on the mGRE interface.
Task 6: Associate the IPsec profile with the mGRE interface to configure tunnel protection.
Task 7: Configure an IP address and IP fragmentation/segmentation parameters on the mGRE interface.
The configuration scenario involves creating a basic hub-and-spoke network with an mGRE interface on the hub and a point-to-point GRE interface on the spoke. The hub has an outer physical IP address of 172.17.0.1 and a tunnel IP address of 10.1.1.1. The spoke has an outer physical IP address of 172.17.2.4 and a tunnel IP address of 10.1.1.2. The NHRP network ID is 1.
The assumption is that there is full IP connectivity among all physical interfaces.
Task 1 is to optionally create an IKE policy with the crypto isakmp policy global configuration command. Example 17-17 shows an IKE policy specifying RSA signatures for authentication and Diffie-Hellman group 14 to protect the key exchange command.
Example 17-17. [Spoke] Configure an Optional IKE Policy
Spoke(config)# crypto isakmp policy 10
Spoke(config-isakmp)# auth rsa-sig
Spoke(config-isakmp)# group 14
Task 2 of this configuration sequence involves specifying that certificates be used for authentication. Refer to the configuration process covered in the section, “Enroll a Cisco IOS Software VPN Router into a PKI and Troubleshoot the Enrollment Process,” in Chapter 16, “Deploying Scalable Authentication in Site-to-Site IPsec VPNs” and enroll the hub router into the PKI and configure other PKI-relevant items.
Task 3 creates a named IPsec profile using the crypto ipsec profile command that allows the transform set and encapsulation mode to be specified. Example 17-18 shows an empty IPsec profile, which indicates that the default transform sets will be used to protect GRE traffic.

Example 17-18. Configure an IPsec Profile
Spoke(config)# crypto ipsec profile IPsec-PROFILE
Task 4 creates the mGRE tunnel interface. Enter the interface tunnel command and then configure basic GRE parameters. The tunnel mode gre multipoint command designates the tunnel interface as mGRE and the tunnel source command specifies the physical interface to which the GRE tunnel is bound. The tunnel key command is required and must match the tunnel key configured on the spokes. This command allows network administrators to run more than one DMVPN at a time on the same router. The GRE tunnel key therefore uniquely identifies the DMVPN. Example 17-19 shows the sequence of commands.

Example 17-19. Create an mGRE Tunnel Interface
Spoke(config)# interface tunnel0
Spoke(config-if)# tunnel mode gre multipoint
Spoke(config-if)# tunnel source GigabitEthernet0/0
Spoke(config-if)# tunnel key 13579
Task 5 configures an NHRP client on the spoke router inside the mGRE tunnel interface. The ip nhrp network-id command specifies the NHRP network ID, the ip nhrp authentication command configures authentication for NHRP traffic, and the ip nhrp nhs command specifies the hub as the NHRP server. Finally, the ip nhrp map multicast dynamic command registers the multicast capabilities of the device. Example 17-20 shows the sequence of commands.

Example 17-20. Configure the NHRP Client
Spoke(config-if)# ip nhrp network-id 1
Spoke(config-if)# ip nhrp authentication WERQRQ$
Spoke(config-if)# ip nhrp nhs 10.1.1.1
Spoke(config-if)# ip nhrp map multicast 172.17.0.1
Spoke(config-if)# ip nhrp map 10.1.1.1 172.17.0.1
Task 6 specifies the IPsec protection profile created in Step 3 as the tunnel protection mechanism with the tunnel protection ipsec profile command. Example 17-21 shows the profile being applied to the mGRE interface.
Example 17-21. Apply the IPsec Profile to the mGRE Interface
Spoke(config)# interface tunnel0
Spoke(config-if)# tunnel protection ipsec profile IPsec-PROFILE
Task 7 specifies the IP address for the mGRE tunnel interface. This must be an IP address in the same network as the hub’s mGRE subnet. Optionally, to avoid problems with fragmentation of user packets, set the IP MTU side with the ip mtu command and set the TCP MSS value using the ip tcp adjust-mss interface command. Example 17-22 shows the profile being applied to the mGRE interface.

Example 17-22. Configure IP Parameters on the mGRE Interface
Spoke(config)# interface tunnel0
Spoke(config-if)# ip address 10.1.1.2 255.255.0.0
Spoke(config-if)# ip mtu 1400
Spoke(config-if)# ip tcp adjust-mss 1360
The show dmvpn detail command provides a snapshot view of the state of the spoke’s association with the hub.
The output displays entries for all connected spokes and provides the following information:
• Session information (NBMA address, tunnel address, state, uptime, and target network)
• State of IKE SAs
• Crypto session status
• Traffic counters to show that user traffic is being protected
Some factors must be taken into consideration when using dynamic routing protocols across a DMVPN. These considerations are even more important when implementing a spoke-to-spoke design because the DMVPN cloud is like an NBMA network.
Many routing protocols use IP multicast to discover other participating nodes. For this reason, NHRP multicast maps must be configured on the spoke routers to register their multicast capability with the hub. The hub router can be configured with a dynamic multicast map that will replicate multicast traffic to the spokes that have registered to receive IP multicast. This permits the hub and spoke routers to forward multicast and broadcast, but it does not permit spokes to receive broadcasts from other spokes.
In a DMVPN, routing neighbor relationships only happen between the hub and each spoke. There is never any spoke-to-spoke routing peering. Thus, the hub will propagate information from each spoke network to the other spokes.
Note that routing protocol functionality on the hub is the main factor in determining whether a DMVPN will remain as a strict hub-and-spoke network or expand to a partial or full mesh by propagation of routes between spokes. The other factor is whether spokes use a GRE interface (only allows a strict hub-and-spoke topology) or an mGRE interface, which permits mesh topologies.
EIGRP can be used in either strict hub-and-spoke networks or in partial/full mesh networks.
Configuring EIGRP in a strict hub-and-spoke DMVPN deployment is fairly straightforward. Generally speaking, you should disable automatic summarization on the hub and disable EIGRP split horizon so that the hub will propagate information about spoke networks to other spokes. Disable split horizon with the no ip split-horizon eigrp AS-number interface configuration command.
Mesh DMVPNs cause a problem for route propagation because spoke routers cannot directly exchange information with one another, even though they are on the same logical subnet. This requires the hub router to advertise the subnets from the spokes on the same subnet, and the advertised route must contain the original next hop as it was learned by the hub router from the originating spoke. The no ip next-hop-self eigrp AS-number interface configuration command takes care of this requirement. Examples 17-23 and 17-24 show the configuration for the use of EIGRP with both DMVPN topologies.
Example 17-23. EIGRP with Hub-and-Spoke
router(config)# router eigrp 1
router(config-router)# no auto-summary
router(config-router)# exit
!
router(config)# interface tunnel 0
router(config-if)# no ip split-horizon eigrp 1
Example 17-24. EIGRP with Full Mesh
router(config)# router eigrp 1
router(config-router)# no auto-summary
router(config-router)# exit
!
router(config)# interface tunnel 0
router(config-if)# no ip next-hop-self eigrp
Routet(config-if)# no ip split-horizon eigrp 1
Do not include the outside address space of the tunnel in a routing protocol running inside the tunnel.
Running OSPF in a DMVPN presents the same challenges as running OSPF over other networks. The hub should be configured as a designated router (DR) because it is in direct communication with all the spokes. Typically, there is no backup designated router (BDR).
In strict hub-and-spoke DMVPNs, this can be simplified by including the tunnel interface in the OSPF router process and configuring the tunnel interface on the hub as a point-to-multipoint OSPF network type and on the spokes as point-to-point network types. There is no need for a BDR, and this makes the branches consider the hub as the only path off the subnet. This will greatly simplify the Dijkstra algorithm process for the OSPF area.
For mesh DMVPNs, configure the mGRE tunnel on the hub as an OSPF broadcast network and each spoke router with an OSPF priority of 0 to prevent a spoke from becoming a DR. Examples 17-25 and 17-26 show OSPF configurations for DMVPN topologies.

Example 17-25. OSPF with Hub-and-Spoke
router(config)# interface tunnel 0
router(config-if)# ip ospf network point-to-multipoint
router(config-if)# ip ospf priority 10

Example 17-26. OSPF with Full Mesh
router(config)# interface tunnel 0
router(config-if)# ip ospf network broadcast
router(config-if)# ip ospf priority 10
Verifying routing in a strict hub-and-spoke DMVPN can be accomplished by running the show ip route command on a spoke router to ensure that all spoke networks are reachable through the hub, as demonstrated in Example 17-27. The next hop for each spoke prefix should be the hub tunnel IP address. In addition to verifying routes, verify that IKE SAs have established with the show crypto isakmp sa command and look for a state of QM_IDLE, as demonstrated in Example 17-28.
Example 17-27. show ip route Command Output
Router# show ip route
Codes: I - IGRP derived, R - RIP derived, O - OSPF derived,
C - connected, S - static, E - EGP derived, B - BGP derived,
* - candidate default route, IA - OSPF inter area route,
i - IS-IS derived, ia - IS-IS, U - per-user static route,
o - on-demand routing, M - mobile, P - periodic downloaded static route,
D - EIGRP, EX - EIGRP external, E1 - OSPF external type 1 route,
E2 - OSPF external type 2 route, N1 - OSPF NSSA external type 1 route,
N2 - OSPF NSSA external type 2 route
Gateway of last resort is 10.119.254.240 to network 10.140.0.0
O E2 172.150.0.0 [160/5] via 10.119.254.6, 0:01:00, Ethernet2
E 172.17.10.0 [200/128] via 10.119.254.244, 0:02:22, Ethernet2
O E2 172.70.132.0 [160/5] via 10.119.254.6, 0:00:59, Ethernet2
O E2 10.130.0.0 [160/5] via 10.119.254.6, 0:00:59, Ethernet2
E 172.30.0.0 [200/128] via 10.119.254.244, 0:02:22, Ethernet2
E 10.129.0.0 [200/129] via 10.119.254.240, 0:02:22, Ethernet2
E 172.80.129.0 [200/128] via 10.119.254.244, 0:02:22, Ethernet2
Example 17-28. show crypto isakmp sa Command Output
Router# show crypto isakmp sa
f_vrf/i_vrf dst src state conn-id slot
/vpn2 172.21.114.123 10.1.1.1 QM_IDLE 13 0
Verifying routing in a full mesh DMVPN can be accomplished by running the show ip route command on a spoke router to ensure that all spoke networks are reachable directly through the spoke router that advertises the network.
You must consider the hub and the spoke routers when troubleshooting connectivity issues between the hub and a spoke router in a DMVPN. For spoke-to-spoke issues between two spokes, you must troubleshoot on the hub as well as the two spoke routers.
The following steps are a suggested troubleshooting sequence in the event that there is an issue with DMVPN deployments:
Step 1: Verify an IKE peering between the hub and spokes (or between spoke and spoke if that is where the issue is) by using the show crypto isakmp sa command. If it is not established, verify connectivity between the two hosts using ping and verifying routes for connectivity issues.
Step 2: Verify the operation of NHRP, GRE, and IPsec. Use the debug dmvpn detail all command to see helpful output.
Step 3: Verify routing over the DMVPN. Check to see whether the routing protocol has established spoke-to-hub neighbor relationships inside the DMVPN using the show ip route command. Ensure that all spoke networks are reachable from the hub and at the correct next hops by pinging the hub from the spoke routers or pinging the hub router from each spoke. If there is a problem, check the NHRP multicast mapping.
As mentioned in the section, “How to Use This Book,” in the Introduction, you have several choices for exam preparation: the exercises here, the memory tables in Appendix C, the final exam preparation chapter, and the exam simulation questions on the CD-ROM. The following questions present a bigger challenge than the exam itself because they use an open-ended question format. By using this more difficult format, you exercise your memory better and prove your conceptual and factual knowledge of this chapter. You can find the answers to these questions in Appendix A, “Answers to the DIKTA Quizzes and Fill in the Blanks Questions.”
Review the most important topics in the chapter, noted with the Key Topics icon in the margin of the page. Table 17-3 lists a reference of these key topics and the page numbers on which each is found.


Print a copy of Appendix C, “Memory Tables” (found on the CD), or at least the section for this chapter, and complete the tables and lists from memory. Appendix D, “Memory Table Answers,” also on the CD, includes completed tables and lists to check your work. Although there are no memory tables in this chapter, you will find some memory tables for other chapters.
Define the following key terms from this chapter, and check your answers in the Glossary:
Encapsulating Security Protocol
multipoint Generic Routing Encapsulation (mGRE)
Next Hop Resolution Protocol (NHRP)
Cisco Design Zone for WAN/MAN, IPsec VPN WAN Design Overview, www.cisco.com/application/pdf/en/us/guest/netsol/ns171/c649/ccmigration_09186a008074f22f.pdf.