CCNP Route Lab 2-4, EIGRP Frame Relay Hub-and-Spoke: Router Used as a Frame Relay Switch

CCNP Route Lab 2-4, EIGRP Frame Relay Hub-and-Spoke: Router Used as a Frame Relay Switch




  • Review a basic configuration of EIGRP on a serial interface.
  • Configure EIGRP over Frame Relay hub-and-spoke.
  • Configure a router as a Frame Relay switch.
  • Configure the ip bandwidth-percent command.
  • Disable split horizon.
  • Use EIGRP in non-broadcast mode.
  • Enable EIGRP manual summarization in topologies with discontiguous major networks.

You are responsible for configuring and testing the new network that connects your company’s headquarters (HQ) and EAST and WEST branches. The three locations are connected over hub-and-spoke Frame Relay,using the company headquarters as the hub. In this lab, you model each branch office’s network with multiple loopback interfaces on each router and configure EIGRP to allow full connectivity between all departments.

To simulate the Frame Relay WAN connections, use a router with three serial ports to act as a Frame Relay switch. The configuration of the router as a Frame Relay switch is described in Step 2.

When accessing a Frame Relay service, a modem or a CSU/DSU is used at the customer premises to connect the router to the local loop and over the local loop to the Frame Relay switch. The modem or CSU/DSU provides the Layer 1 DCE (clocking) functions to the router. However, in this lab, R2 interface serial 0/0/1 is configured as the DCE for compatibility with other labs. If you are uncertain which side of the connection is the DCE, use the show controllers serial interface-number command:

Note: In this lab, Router R3 acts as the Frame Relay switch and requires two serial interface cards. If you are using an Adtran as a Frame Relay switch, see Appendix A for the Adtran configuration. When using the Adtran, the clock (DCE) is provided for each serial link.

Instructor note: You can also use Lab 2.4b from CCNP version 5 in the instructor lab manual, if available.

Note: This lab uses Cisco 1841 routers with Cisco IOS Release 12.4(24)T1 and the Advanced IP Services image c1841 -advipservicesk9-mz.124-24.T1 .bin. You can use other routers (such as 2801 or 2811) and Cisco IOS Software versions if they have comparable capabilities and features. Depending on the router and Cisco IOS Software version, the commands available and output produced might vary from what is shown in this lab.

Required Resources

  • 3 routers (Cisco 1841 with Cisco IOS Release 12.4(24)T1 Advanced IP Services or comparable)
  • 1 router acting as a Frame Relay switch (Cisco 1841 with Cisco IOS Release 12.4(24)T1 Advanced IP Services or comparable)
  • Serial and console cables

Step 1: Configure loopback addressing.
Using the addressing scheme in the diagram, apply IP addresses to the loopback interfaces on HQ, EAST, and WEST. You can paste the following configurations into your routers to begin.

Router R1 (HQ)

Router R2 (EAST)

Router R4 (WEST)

Step 2: Configure the Frame Relay switch.
Use a fourth Cisco router with three serial interfaces as a Frame Relay switch, and cable the routers according to the diagram. Paste the following configuration into the router that is simulating the Frame Relay

Note: If you are using an Adtran as a Frame Relay switch, see Appendix A for the Adtran configuration and which cables to use between the routers and the Adtran.

Router R3 (FRS)

Note: You do not need to configure the LMI type as cisco because it is the default. In addition, the HQ, EAST, and WEST routers are able to automatically determine the LMI type. However, you could configure the Frame Relay switch with a different LMI type on each of its interfaces to demonstrate support for the existing types (cisco, ansi, q933a) and to show that they interoperate cleanly without requiring any particular configuration on the Frame Relay switch or end routers. Additionally, the ietf keyword is meaningful only on Frame Relay end devices, not on Frame Relay switches. It is configured on R3 here for clarity.

Step 3: Configure the Frame Relay endpoints.
You will be configuring HQ to be the Frame Relay hub, with EAST and WEST as the spokes. Check the topology diagram for the data-link connection identifiers (DLCIs) to use in the Frame Relay maps. You will also be turning off Frame Relay Inverse Address Resolution Protocol (InARP) for all interfaces. Inverse ARP allows a Frame Relay network to discover the IP address associated with the virtual circuit. This is sometimes a desirable trait in a production network. However, in the lab, Inverse ARP is turned off because you are using static maps.

a. Assign the Frame Relay interface of each router an IP address in the Frame Relay subnet /29, as indicated in the topology diagram.

b. Enable Frame Relay encapsulation, disable Frame Relay Inverse ARP, and map the other IPs in the subnet to DLCIs using the frame-relay map ip address dlci broadcast command. The broadcast keyword is important because without it, no broadcast or multicast packets including EIGRP messages are sent through the Frame Relay cloud. Bring up the interfaces with the no shutdown command.

Note: It is unnecessary and actually undesirable to specify the mappings from EAST to WEST and from WEST to EAST with the broadcast option. Specifying this option duplicates all broadcasts and multicasts sent from EAST or WEST routers. Because of the hub-and-spoke topology, HQ will also receive all broadcasts and multicasts twice. Specify the broadcast option if a direct PVC is going toward the mapped IP, as is the case between EAST and HQ and between WEST and HQ.

Note: It is good practice to specify the IETF encapsulation format, which is an open format. Using the IETF frame format helps ensure interoperability between different vendors.

You will configure the bandwidth for the serial links in Step 4.

c. Verify that you have connectivity across the Frame Relay network by pinging the remote routers from each of the Frame Relay endpoints.

The only interface that the Frame Relay interface is unable to communicate with is itself. This is not a significant problem in Frame Relay networks. To allow the Frame Relay interface to ping itself, you can map the local IP address to be forwarded out a PVC. The remote router at the other end of the PVC can then forward the IP address back based on its Frame Relay map statements. This solution is so that the Tcl scripts used for testing return successful echo replies under all circumstances. You do not need the broadcast keyword on this DLCI, because it is not important to forward broadcast and multicast packets (such as EIGRP hellos) to your own interface.

d. Configure the local mappings.

HQ now forwards packets destined for its own serial interface ( to the EAST serial interface ( and then back. This allows HQ to ping its own serial interface in the lab Frame Relay network. In a production network in which a company is billed based on per-PVC usage, this is not a preferred configuration. However, in this lab network, it helps ensure full ICMP connectivity in the Tcl scripts.

Step 4: Set interface-level bandwidth.
a. On the three routers, set the Frame Relay serial interface bandwidth with the bandwidth bandwidth command in interface configuration mode. Specify the bandwidth in kilobits per second (kb/s). For HQ,
use 128 kb/s. On EAST and WEST, use 64 kb/s.

Recall from Lab 2-1 that, by default, EIGRP limits its bandwidth usage to 50 percent of the value specified by the bandwidth parameter. The default bandwidth for a serial interface is 1544 kb/s. This means that each neighbor for which this is an outbound interface has a traffic limit of a fraction of that 50 percent,represented by 1/N, where N is the number of neighbors out that interface.

The HQ serial interface divides its total EIGRP bandwidth into fractional amounts according to the number of neighbors out that interface.

How much bandwidth on HQ serial 0/0/1 is reserved for EIGRP traffic to EAST?
On HQ, up to 32 Kb/s of bandwidth can be utilized for EIGRP traffic to EAST, which represents one fourth of the total bandwidth of that interface.

b. You can control both the bandwidth and the EIGRP bandwidth percentage on a per-interface basis. On HQ, limit the bandwidth used by EIGRP to 40 percent without changing the bandwidth parameter on the interface using the interface-level command ip bandwidth-percent eigrp as_number percent.

Step 5: Configure EIGRP routing.
a. Configure EIGRP AS 1 on HQ, EAST, and WEST.
The network represented in the diagram is a discontiguous network ( configured on all three routers. If you enabled auto-summarization, HQ sends and receives summaries for with both EAST and WEST. Auto-summarization causes considerable routing disruption in this network, because HQ does not know which of the two spokes is the correct destination for subnets of For this reason, turn off auto-summarization on each router.

b. Add your network statements to EIGRP on all three routers. The two major networks being used are for the loopbacks and for the Frame Relay cloud.

c. Issue the show ip eigrp topology command on EAST.

Which networks are missing from the topology database?
All the subnets in the range are missing from the EAST topology table. These subnets are the remote loopback interfaces on the WEST router.

What do you think is responsible for this problem?
This problem is caused by the split horizon functionality on HQ. HQ will not send route updates from EAST to WEST because the incoming interface and outgoing interface are the same.

d. HQ needs the no ip split-horizon eigrp as_number command on its serial Frame Relay interface. This command disables split horizon for an EIGRP autonomous system. If split horizon is enabled (the default), route advertisements from EAST to HQ do not travel to WEST and vice versa, as shown in the above output.

e. Verify that you see the correct EIGRP adjacencies with the show ip eigrp neighbors command.

f. Verify that you have IP routes on all three routers for the entire topology with the show ip route command.

g. Run the following Tcl script on all routers to verify full connectivity.

You should get ICMP echo replies for every address pinged.

Step 6: Configure non-broadcast EIGRP mode.
Currently, you are using EIGRP in its default mode, which multicasts packets to the link-local address However, not all Frame Relay configurations support multicast. EIGRP supports unicasts to remote destinations using non-broadcast mode on a per-interface basis. This mode is analogous to configuring RIPv2 with a passive interface and statically configuring neighbors out that interface.

a. Implement packet unicasts to neighbors.

b. HQ now has two neighbor statements, and the other two routers have one. After you configure neighbor statements for an interface, EIGRP automatically stops multicasting packets out that interface and starts unicasting packets instead. Use the show ip eigrp neighbors command to verify the changes.

Step 7: Implement EIGRP manual summarization.
a. Implement EIGRP manual summarization on each router. Each router should advertise only one network summarizing all of its loopbacks. Using the commands you learned in EIGRP Lab 2-3, configure the summary address on the serial interfaces.

What is the length of the network mask that is used to summarize all the loopbacks on each router?
Summarize all the loopbacks at the 16-bit boundary for each router.

b. Look at the simplified EIGRP topology table on each router using the show ip eigrp topology command.

Router Interface Summary Table

Router Interface Summary
Router Model Ethernet Interface
Ethernet Interface
Serial Interface
Serial Interface
1700 Fast Ethernet 0
Fast Ethernet 1
Serial 0 (S0) Serial 0/0/1
1800 Fast Ethernet 0/0
Fast Ethernet 0/1
Serial 0/0/0
Serial 0/0/1
2600 Fast Ethernet 0/0
Fast Ethernet 0/1
Serial 0/0 (S0/0) Serial 0/1 (S0/1)
2800 Fast Ethernet 0/0
Fast Ethernet 0/1
Serial 0/0/0
Serial 0/0/1
Note: To find out how the router is configured, look at the interfaces to identify the type of router and how many interfaces the router has. Rather than list all combinations of configurations for each router class, this table includes identifiers for the possible combinations of Ethernet and serial interfaces in the device. The table does not include any other type of interface, even though a specific router might contain one. For example, for an ISDN BRI interface, the string in parenthesis is the legal abbreviation that can be used in Cisco IOS commands to represent the interface.

If an Adtran Atlas 550 is used for the Frame Relay switch, connect the serial cable from each router interface
in the topology diagram to the Adtran interface indicated in the table below. The Adtran is preconfigured to
simulate a Frame Relay service that provides the following PVCs.

HQ S0/0/1 DTE port 1/1 102 201 East
HQ S0/0/1 DTE port 1/1 103 301 West
East S0/0/1 DTE port 1/2 201 102 HQ
West S0/0/0 DTE port 2/1 301 103 HQ

Frame Relay Switching Configuration

The Adtran Frame Relay switch interfaces all provide the DCE clock. Be sure to use the appropriate cable
between each router and the Adtran. All the router interfaces are DTE, and the cable to the Adtran interface
should be serial to V.35 DCE. Use the show controllers command to verify which cable type is connected to
a given router interface.

Device Configurations (Instructor version)




More Resources

About the author


Leave a Comment