CCNP Route Lab 2-2, EIGRP Load Balancing

CCNP Route Lab 2-2, EIGRP Load Balancing

Topology

ccnp-route-lab-2-2-eigrp-load-balancing

Objectives

  • Review a basic EIGRP configuration.
  • Explore the EIGRP topology table.
  • Identify successors, feasible successors, and feasible distances.
  • Use show and debug commands for the EIGRP topology table.
  • Configure and verify equal-cost load balancing with EIGRP.
  • Configure and verify unequal-cost load balancing with EIGRP.

Background
As a senior network engineer, you are considering deploying EIGRP in your corporation and want to evaluate its ability to converge quickly in a changing environment. You are also interested in equal-cost and unequalcost load balancing because your network contains redundant links. These links are not often used by other link-state routing protocols because of high metrics. Because you are interested in testing the EIGRP claims that you have read about, you decide to implement and test on a set of three lab routers before deploying EIGRP throughout your corporate network.

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 a 2801 or 2811) and Cisco IOS Software versions if they have comparable capabilities and features. Depending on the router model 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)
  • Serial and console cables

Step 1: Configure the addressing and serial links.
a. Create three loopback interfaces on each router and address them as 10.1.X.1/30, 10.1.X.5/30, and 10.1.X.9/30, where X is the number of the router. Use the following table or the initial configurations located at the end of the lab.

Router Interface IP Address/Mask

Router Interface IP Address/Mask
R1 Loopback11 10.1.1.1/30
R1 Loopback15 10.1.1.5/30
R1 Loopback19 10.1.1.9/30
R2 Loopback21 10.1.2.1/30
R2 Loopback25 10.1.2.5/30
R2 Loopback29 10.1.2.9/30
R3 Loopback31 10.1.3.1/30
R3 Loopback35 10.1.3.5/30
R3 Loopback39 10.1.3.9/30
b. Specify the addresses of the serial interfaces as shown in the topology diagram. Set the clock rate to 64 kb/s, and manually configure the interface bandwidth to 64 kb/s.

Note: If you have WIC-2A/S serial interfaces, the maximum clock rate is 128 kb/s. If you have WIC-2T serial interfaces, the maximum clock rate is much higher (2.048 Mb/s or higher depending on the hardware), which is more representative of a modern network WAN link. However, this lab uses 64 kb/s and 128 kb/s settings.

c. Verify connectivity by pinging across each of the local networks connected to each router.

d. Issue the show interfaces description command on each router. This command displays a brief listing of the interfaces, their status, and a description (if a description is configured). Router R1 is shown as an example.

e. Issue the show protocols command on each router. This command displays a brief listing of the interfaces, their status, and the IP address and subnet mask configured (in prefix format /xx) for each interface. Router R1 is shown as an example.

Step 2: Configure EIGRP.
a. Enable EIGRP AS 100 for all interfaces on R1 and R2 using the commands used in the previous EIGRP lab. Do not enable EIGRP yet on R3. For your reference, these are the commands which can be used:

b. Use the debug ip eigrp 100 command to watch EIGRP install the routes in the routing table when your routers become adjacent. You get output similar to the following.

Essentially, the EIGRP DUAL state machine has just computed the topology table for these routes and installed them in the routing table.

c. Check to see that these routes exist in the routing table with the show ip route command.

d. After you have full adjacency between the routers, ping all the remote loopbacks to ensure full connectivity or use the following Tcl script. If you have never used Tcl scripts or need a refresher, see Lab 1 –1.

You should receive ICMP echo replies for each address pinged. Make sure that you run the Tcl script on each router and verify connectivity before you continue with the lab.

e. Verify the EIGRP neighbor relationships with the show ip eigrp neighbors command.

Step 3: Examine the EIGRP topology table.
a. EIGRP builds a topology table containing all successor routes. The course content covered the vocabulary for EIGRP routes in the topology table. What is the feasible distance of route 10.1.1.0/30 in the R3 topology table in the following output?
_______________________________________________________________________________
_______________________________________________________________________________

The feasible distance (FD) for the 10.1.1.0/30 route is 40640000.

b. The most important thing is the two successor routes in the passive state on R3. R1 and R2 are both advertising their connected subnet of 10.1.102.0/30. Because both routes have the same feasible distance of 41024000, both are installed in the topology table. This distance of 41024000 reflects the composite metric of more granular properties about the path to the destination network. Can you view the metrics before the composite metric is computed?
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________

Yes, the EIGRP route advertisements and updates indicate each of the individual path metrics that EIGRP uses. These path metrics can be displayed with the show ip eigrp topology network/mask command.

c. Use the show ip eigrp topology 10.1.102.0/29 command to view the information that EIGRP has received about the route from R1 and R2.

The output of this command shows the following information regarding EIGRP:

  • The bandwidth metric represents the minimum bandwidth among all links comprising the path to the destination network.
  • The delay metric represents the total delay over the path.
  • The minimum MTU represents the smallest MTU along the path.
  • If you do not have full knowledge of your network, you can use the hop count information to check how many Layer 3 devices are between the router and the destination network.

Step 4: Observe equal-cost load balancing.
EIGRP produces equal-cost load balancing to the destination network 10.1.102.0/29 from R1. Two equal-cost paths are available to this destination per the topology table output above.

a. Use the traceroute 10.1.102.1 command to view the hops from R3 to this R1 IP address. Notice that both R1 and R2 are listed as hops because there are two equal-cost paths and packets can reach this network via either link.

Recent Cisco IOS releases enable Cisco Express Forwarding (CEF), which, by default, performs perdestination load balancing. CEF allows for very rapid switching without the need for route processing. However, if you were to ping the destination network, you would not see load balancing occurring on a packet level because CEF treats the entire series of pings as one flow.

CEF on R3 overrides the per-packet balancing behavior of process switching with per-destination load balancing.

b. To see the full effect of EIGRP equal-cost load balancing, temporarily disable CEF and route caching so that all IP packets are processed individually and not fast-switched by CEF.

Note: Typically, you would not disable CEF in a production network. It is done here only to illustrate load balancing. Another way to demonstrate per-packet load balancing, that does not disable CEF, is to use the per-packet load balancing command ip load-share per-packet on outgoing interfaces S0/0/0 and S0/0/1.

c. Verify load balancing with the debug ip packet command, and then ping 10.1.102.1. You see output similar to the following:

Notice that EIGRP load-balances between Serial0/0/0 (s=10.1 .103.3) and Serial0/0/1 (s=10.1.203.3). This behavior is part of EIGRP. It can help utilize underused links in a network, especially during periods of congestion.

Step 5: Analyze alternate EIGRP paths not in the topology table.
a. Perhaps you expected to see more paths to the R1 and R2 loopback networks in the R3 topology table.Why are these routes not shown in the topology table?
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________

Take the example of the R1 loopback interfaces:
R3 receives updates about the R1 loopback interfaces from both R1 and R2. Although both routes have the same minimum bandwidth, the path through R1 has a lower delay than that through R2. Therefore, the computed distance/advertised distance pair is 40640000/128256 through R1 and 41152000/40640000 through R2. The computed distance through R1 (40640000) becomes the feasible distance. R3 receives the path through R2, but it does not enter the path into the EIGRP topology table because the route information must meet the feasibility condition to be inserted into the topology table. Because the AD from R2 is not strictly less than the FD, the route is not inserted in the EIGRP topology table.

b. Issue the show ip eigrp topology all-links command to see all routes that R3 has learned through EIGRP. This command shows all entries that EIGRP holds on this router for networks in the topology, including the exit serial interface and IP address of the next hop to each destination network, and the serial number (serno) that uniquely identifies a destination network in EIGRP.

What is the advertised distance of the R1 loopback network routes from R1 and R2?
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________

The advertised distance of the loopback interfaces on R1 from R1 is 128256. The advertised distance to the same loopback interfaces advertised to R3 from R2 is 40640000.

c. Use the show ip eigrp topology 10.1.2.0/30 command to see the granular view of the alternate paths to 10.1.2.0, including ones with a higher reported distance than the feasible distance.

When using the show ip eigrp topology command, why is the route to 10.1.2.1 through R1 not listed in the topology table?
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________

The route does not meet the feasibility condition. The feasibility condition states that the AD must be strictly less than the FD for the route to be entered into the topology table. ADR1 (40640000) < FD (40640000) is false. Therefore, the route is not entered into the topology table.

What is its advertised distance from R1 ?
_______________________________________________________________________________
ADR1 = 40640000
What is its feasible distance?
_______________________________________________________________________________
FD = 40640000
If the R2 Serial0/0/1 interface were shut down, would EIGRP route through R1 to get to 10.1.2.0/30? Would the switch be immediate?
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________
_______________________________________________________________________________

Yes, EIGRP would begin routing through R1 to get to 10.1.2.0/30 after it detects that the link protocol is down, goes active on the route, and receives replies informing it of the route through R1 .

The switch over to the path through R1 would not be immediate. However, as soon as R3 realizes that the link is down, it will begin to recalculate. The switch will be very quick.

Record your answer, and then experiment by shutting down the R1 s0/0/01 interface while an extended ping is running as described below.

d. Start a ping with a high repeat count on R3 to the R1 Serial0/0/0 interface 10.1.102.1.

e. Enter interface configuration mode on R1 and shut down port Serial0/0/1, which is the direct link from R1 to R3.

f. When the adjacency between R1 and R3 goes down, some pings will be lost. After pings are again being successfully received, stop the ping using Ctrl+Shift+^.

How many packets were dropped?
_______________________________________________________________________________

Two packets were dropped during the cutover.

Note: When examining the EIGRP reconvergence speed after deactivating the serial link between R1 and R3, the focus should not be on the count of lost ping packets but rather on the duration of connectivity loss or how long it took to perform a successful cutover. The router waits for up to two seconds for each sent ICMP ECHO request to receive a reply and only then does it send another ECHO request. If the router did not wait for the reply, the count of lost packets would be much higher. Because two packets were lost, the cutover took approximately 4 seconds.

Another factor to consider is that an interface deliberately delays the information about loss of connectivity for 2 seconds to prevent transient link flaps (link going up and down) from introducing instability into the network. If the real speed of EIGRP is to be observed, this delay can be made as short as possible using the command carrier-delay msec 0 on all serial interfaces.

g. Issue the no shutdown command on the R1 Serial0/0/1 interface before continuing to the next step.

Step 6: Observe unequal-cost load balancing.
a. Review the composite metrics advertised by EIGRP using the show ip eigrp topology 10.1.2.0/30 command,.

The reported distance for a loopback network is higher than the feasible distance, so DUAL does not consider it a feasible successor route.

b. To demonstrate unequal-cost load balancing in your internetwork, upgrade the path to the destination network through R1 with a higher bandwidth. Change the clock rate and bandwidth on the R1, R2, and

c. Issue the show ip eigrp topology 10.1.2.0/30 command again on R3 to see what has changed.

After manipulating the bandwidth parameter, the preferred path for R3 to the loopback interfaces of R2 is now through R1. Even though the hop count is two and the delay through R1 is nearly twice that of the R2 path, the higher bandwidth and lower FD results in this being the preferred route.

d. Issue the show ip route command to verify that the preferred route to network 10.1.2.0 is through R1 via Serial0/0/0 to next hop 10.1.103.1. There is only one route to this network due to the difference in bandwidth.

e. Issue the debug ip eigrp 100 command on R3 to show route events changing in real time. Then, under the EIGRP router configuration on R3, issue the variance 2 command, which allows unequal-cost load balancing bounded by a maximum distance of (2) × (FD), where FD represents the feasible distance for each route in the routing table.

f. Issue the show ip route command again to verify that there are now two routes to network 10.1.2.0.

g. These unequal-cost routes also show up in the EIGRP topology table, even though they are not considered feasible successor routes. Use the show ip eigrp topology command to verify this.

h. Load balancing over serial links occurs in blocks of packets, the number of which are recorded in the routing table’s detailed routing information. Use the show ip route 10.1.2.0 command to get a detailed view of how traffic is shared between the two links.

i. Check the actual load balancing using the debug ip packet command. Ping from R3 to 10.1.2.1 with a high enough repeat count to view the load balancing over both paths. In the case above, the traffic share is 25 packets routed to R2 to every 48 packets routed to R1.

j. To filter the debug output to make it more useful, use the following extended access list.

Note: If a deliberate metric manipulation is necessary on a router to force it to prefer one interface over another for EIGRP-discovered routes, it is recommended to use the interface-level command “delay” for these purposes. While the “bandwidth” command can also be used to influence the metrics of EIGRP-discovered routes through a particular interface, it is discouraged because the “bandwidth” will also influence the amount of bandwidth reserved for EIGRP packets and other IOS subsystems as well. The “delay” parameter specifies the value of the interface delay that is used exclusively by EIGRP to perform metric calculations and does not influence any other area of IOS operation.

Router Interface Summary Table

Router Interface Summary
Router Model Ethernet Interface
#1
Ethernet Interface
#2
Serial Interface
#1
Serial Interface
#2
1700 Fast Ethernet 0
(Fa0)
Fast Ethernet 1
(Fa1)
Serial 0 (S0) Serial 0/0/1
(S0/0/1)
1800 Fast Ethernet 0/0
(Fa0/0)
Fast Ethernet 0/1
(Fa0/1)
Serial 0/0/0
(S0/0/0)
Serial 0/0/1
(S0/0/1)
2600 Fast Ethernet 0/0
(Fa0/0)
Fast Ethernet 0/1
(Fa0/1)
Serial 0/0 (S0/0) Serial 0/1 (S0/1)
2800 Fast Ethernet 0/0
(Fa0/0)
Fast Ethernet 0/1
(Fa0/1)
Serial 0/0/0
(S0/0/0)
Serial 0/0/1
(S0/0/1)
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.

Device Configurations (Instructor version)

Initial Configurations
Router R1

Router R2
Router R3
Final Configurations
Router R1
Router R2
Router R3

About the author

Scott

Leave a Comment