Implementing EIGRP

Basic EIGRP configuration is not that different from configuring RIP. The primary

difference for basic configuration is that you must specify an autonomous

system number that defines your routing domain. The autonomous system

number is assigned globally for the routing process and can be any number you

want, but that same number must be used on all routers. Routing updates will

not be exchanged between routers with different autonomous numbers. Because

the exam focuses heavily on troubleshooting, make sure you always check that

the autonomous numbers match in the exam scenarios.

The following example shows how to configure EIGRP for a router connected

to networks 192.168.10.0/24 and 192.168.20.0/24. The autonomous system

number is 1 and is specified when entering the routing process.

Similar to RIP version 2 and OSPF, EIGRP can be a classless routing protocol.

By default, it is classful. To enable classless routing, type the following command

under the routing process:

Verifying and Troubleshooting EIGRP

A good engineer does not just configure routing but knows to verify the configuration

again, so it is not discussed here. Keep in mind, though, that this is best

command to use to see whether your routing table is being populated.

number and the networks you are advertising.

Routing Protocol is “eigrp 1”

Outgoing update filter list for all interfaces is not set

Incoming update filter list for all interfaces is not set

Redistributing: eigrp 1

Automatic network summarization is in effect

Routing for Networks:

192.168.0.0

Routing Information Sources:

Gateway Distance Last Update

192.168.1.0 90 0:02:36

192.168.2.0 90 0:03:04

192.168.3.0 90 0:03:04

Distance: internal 90 external 170

Table 14.1 summarizes the important lines of this command.

Output Description

Outgoing/incoming filters Used to filter routing updates between routers.

Redistributing Covered in the Cisco Certified Network Professional (CCNP)

exam. This pertains to redistributing information between

routing protocols and is outside the scope of this exam.

summarization is in effect In this example, the command has not been applied, and

EIGRP is doing classful routing.

Routing for networks Which networks your router is advertising to other routers.

Routing information sources This defines which routers are sending your EIGRP routes,

the administrative distance for those routes, and the last time

your router received an update from other routers.

Distance The administrative distance for internal and external routes.

this outputs your topology table. Your topology table contains all the routes

your router knows about. Here is where you will also see your successor (best

routes) and your feasible successor (backup routes):

IP-EIGRP Topology Table for process 77

Codes: P - Passive, A - Active, U - Update, Q - Query, R - Reply,

r - Reply status

P 172.16.0.0 255.255.0.0, 2 successors, FD is 36251776

via 172.16.17.1 (36251776/36226176), Ethernet0

via 172.16.18.1 (36251776/36226176), Ethernet1

P 172.20.0.0 255.255.0.0, 1 successors, FD is 307200

via 172.16.81.28 (307200/281600), Ethernet1

via 172.16.19.5 (702311/295210), Ethernet2

From this output you can begin to get an idea of the topology of your network.

Notice that for the 172.16.0.0/16 network you have two successors. This is

because the metric is the same for both networks and, subsequently, you will

load balance across two networks. The metric that is put in the routing table is

the first number in parenthesis (36251776 in this example) and is called the feasible

distance (FD).

The 172.20.0.0 network has only one successor route out Ethernet1 that is

learned from a router with the IP address 172.16.81.28. You also have a backup

route (feasible successor) out Ethernet2 that is learned from a router at

172.16.19.5.

For the exam, make sure you are comfortable analyzing the output of these show

commands.

Active Versus Passive Routes

begin with a P for passive. According to the legend at the beginning of this output, a route

can also be A for active. A passive route is when your routing table has fully converged.

An active route is when a route has changed and your routers are querying other routers

to discover the change in the topology. Ideally, your routes should be in passive mode.

OSPF was developed by the Internet Engineering Task Force (IETF) in

1988 as a more scalable solution than RIP. Unlike EIGRP, OSPF is an open standard

and is not Cisco proprietary. It uses the Shortest Path First (SPF) algorithm

developed by Edgar Dijkstra. It is a link state routing protocol, which means that

it sends updates only when there is a change in the network, and instead of sending

routing updates, it sends link state advertisements (LSAs) instead.

Characteristics

OSPF is a polite protocol. Unlike chatty RIP, which broadcasts out its entire

routing table every 30 seconds regardless of whether other routers want to hear

it, OSPF takes a more gentlemanlike approach to routing. First, OSPF sends

out hello messages to neighboring routers to announce itself as an OSPF router

and discover who its neighbor routers are. Routers have to agree on certain

parameters (such as timers and being on a common subnet) before they can

become neighbors. After its neighbor routers are discovered, they begin to

exchange information about networks (links) it knows about, using messages

called link state advertisements (LSAs). After exchanging all routes, the routers

send out updates only when there is a change, and they send information only

for that affected route, not the entire routing table. Routers take the link state

advertisements heard from other routers and place those routes in its link state

database (similar to the topology database in EIGRP). Routers then run the SPF

algorithm to determine the best route to a destination and place that route in

the routing table.

To determine the best path, OSPF uses a metric called cost, which Cisco defines

as 108/bandwidth. If you had a 100Mbps link, the cost would be 1 because

100,000,000/100,000,000. Here are some other common costs:

The SPF algorithm places each router as the “root” of a tree and calculates the

shortest path from itself to each destination. The shortest path then gets put

into the routing table and is used to route packets to their destination.

An important concept to grasp with OSPF is that it is a hierarchical protocol.

Hierarchical routing protocols break up your autonomous system into multiple

areas and summarize routes between areas. If summarized wisely, you can cut

down a significant portion of routing updates by advertising only the summarized

route.

 As the number of networks increases in your domain, the amount of processing

required on each router increases. To lower the amount of processing required,

you can use route summarization. Route summarization looks for the same

sequence of bits used in subnetworks and creates a less-explicit summary route.

For example, Figure 14.2 shows four networks in area 2:

The first octet, 172, is the same for all four routes, but the second octet differs.

By looking for similar bits, we can create a single summary route:

16 0 0 0 1 0 0 0 0

17 0 0 0 1 0 0 0 1

18 0 0 0 1 0 0 1 0

19 0 0 0 1 0 0 1 1

The bits are the same up to the 4-bit position. Only the 16-bit position is set to

1, so by ignoring the last two bits (because they change), we are left with

172.16.0.0. The subnet mask has changed, however, because we are no longer

working with a /24. Instead, our subnet mask has moved two places to the left

because the last two bit positions vary for the four networks. Our resulting summarized

route is 172.16.0.0/22 (255.255.252.0). This will be the route that gets

injected into area 0 from area 2.

The routers in area 0 and area 1 have to process only the one summarized route

instead of four individual routes. Being able to summarize your routes between

areas provides several benefits:

statement (in contrast to four), but also because of the lack of recalculation

should a more specific network (that is, a /24) go down.

in area 2, it will not affect the routers in area 0 and area 1.

in areas 0 and 1 can converge faster.

route is sent, so the advertisement is smaller.

routing updates, you can control what routes get sent from one area to

another.

You might have noticed that both area 2 and area 1 are connected via area 0.

Area 0 is the “backbone” area in OSPF, and all other areas must be connected to

it. Routes are then summarized into your backbone area.

Summarizing is an excellent way to conserve your precious bandwidth. On networks

that contain more than two routers, OSPF can also conserve bandwidth

by electing a designated router for that network that all routers communicate

with. Routers exchange information with a designated router instead of each

other. This cuts down significantly on the number of advertisements.

The process of using a designated router is somewhat complex, so let’s go

through it one step at a time. First, the designated router (DR) is elected on only

two types of networks:

On a point-to-point network with only two routers, there is no need for this

type of election. Remember that on a point-to-point network, there is no point

(of having a DR).

Second, the DR is not the only type of router elected on these types of networks.

A backup designated router (BDR) is used in the event that a DR should fail.

The DR and BDR election is as follows:

the second-highest priority becomes the BDR. Priority is a number

between 0 and 255 and is configured on an interface with the command

router is set to priority 0, it will never become a DR or BDR.

default of 1, the tie breaker is the router with the highest router ID.

Every router has an identifier called a router ID (RID) that is used to identify

itself in its messages. The router ID is an IP address and is assigned as follows:

the OSPF routing process. You can choose a valid IP address that you

are using on the router or make up a new one.

address on any loopback interface is chosen as the router ID. A loopback

interface is a virtual, software-only interface that never goes down.

address on any active physical interface is chosen as the router ID.

See if you can spot the router ID given the following IP addresses on a router:

Serial 0/0: 192.168.100.19

FastEthernet 0/0: 10.0.0.1

Loopback 0: 172.16.201.200

Although the highest IP address is the one configured on the serial interface, a

loopback interface takes precedence over any physical interfaces. Therefore, the

router ID would be 172.16.201.200.

Let’s review. On broadcast and nonbroadcast multi-access networks, a designated

router and backup designated router are elected. The election is done by first

choosing the routers with the highest priority value or, if the priorities are same,

choosing the routers with the highest router ID. The router ID is chosen by the

highest IP address on any loopback interface or, if no loopback interfaces are

configured, the highest IP address on any active physical interface. Whew!

That’s a lot of work, but in the end it will conserve a significant amount of bandwidth

by minimizing the number of link state messages.

Now that we have elected a DR and BDR, the next phase is ready to begin. In

Figure 14.3, you see five routers. The Mocha router is the DR, and the Latte

router is the BDR. Instead of all routers sending link state advertisements to

each other, they send out messages only to the DR and BDR. Messages are sent

to the multicast address of 224.0.0.6; both the DR and BDR belong to this multicast

group address.

Next, the Mocha router, which is the DR, takes the information it learned from

the other routers and sends it back out to all routers, as shown in Figure 14.4.

Messages are sent to the All SPF Router multicast address of 224.0.0.5; all routers

running OSPF are members of this multicast group address.

 

 

 

 

 

 

 

 

 

 

 

 

 

Understanding the complexities involved in OSPF is the difficult part; configuring

it is fairly straightforward. The process is the same as with the other protocols.

First, we enable the routing protocol. This is done with the command

router ospf <process-id>. The process ID can be any number you prefer

Note that you specify a wildcard mask in the configuration.   Here, wildcard masks are used to

match the IP address that is being used on an interface.

Take a look at next figure, where we come across our three friends again: Moe,

Larry, and Curly. Given this example, the configuration for Moe would be

Moe(config-router)#network 192.168.20.0 0.0.0.255 area 0

Cappuccino Latte Chai

Decaf

DR

BDR

Mocha

Larry’s configuration would be

Larry(config)#router ospf 1

Larry(config-router)#network 192.168.20.0 0.0.0.255 area 0

Larry(config-router)#network 192.168.40.0 0.0.0.255 area 1

Finally, Curly’s configuration would be

Curly(config)#router ospf 1

Curly(config-router)#network 192.168.40.0 0.0.0.255 area 1

Curly(config-router)#network 192.168.50.0 0.0.0.255 area 1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The wildcard mask used in t

0.0.0.255 area 1 tells the router to match all addresses that begin with