Bab 3

INTRODUCTION
This chapter examines advanced routing protocols. Configuring the

link state protocols Open Shortest Path First (OSPF) and Intermediate
System-to-Intermediate System (IS-IS) are examined in Sections 3-1 and 3-2.
In link state protocols, updates are only sent when there is a change in the
network. There are many similarities between IS-IS and OSPF; however, there
is a difference in the way in which the areas are defined for each protocol.
Another big difference is that the IS-IS backbone area can be segmented, un-
like the backbone area in OSPF.
The steps for configuring the Enhanced Interior Gateway Routing Protocol
(EIGRP) are examined in Section 3-3. EIGRP is a Cisco proprietary protocol
and is often called a hybrid routing protocol that incorporates the best of the
distance vector and link-state algorithms.
Section 3-4 takes a look at advanced route redistribution. This section ex-
plores different routing redistribution techniques used to redistribute routes
into dynamic routing protocols, such as Routing Information Protocol (RIP),
OSPF, IS-IS, and EIGRP. This chapter concludes with a look at the OSPF
“Hello” packets that are used by routers to initiate and maintain communica-
tion with neighbor routers. The parameters of the “Hello” packet are exam-
ined using a network protocol analyzer.
Each routing protocol section in this chapter contains a networking challenge
that is included with the Net-Challenge Software. These challenges enable
you to configure each routing protocol on a virtual router.
3-1 CONFIGURING LINK STATE PROTOCOLS—OSPF

Open Shortest Path First (OSPF) is a dynamic routing protocol, classified specifi-
cally as a link state protocol. It was developed by the Interior Gateway Protocol OSPF
(IGP) working group for the Internet Engineering Task Force (IETF) specifically Open Shortest Path First
routing protocol.
for use in TCP/IP networks. OSPF is an open, not proprietary, protocol and is sup-
ported by many vendors. The main advantages of OSPF are rapid convergence and IETF
the consumption of very little bandwidth. When a network is completely converged, Internet Engineering
all the routers in the network agree on the best routes. After the initial flooding of Task Force.
routes in the form of link state advertisements (LSA), OSPF sends route updates
Link State
only when there is a change in the network. Every time LSAs are sent, each router Advertisement (LSA)
must recalculate the routing table.
The exchange of updated
This is a distinct advantage over RIP. Recall that RIP exchanges the entire routing link state information
table at fixed time intervals and RIP updates every 30 seconds. Also, in RIP, the when routes change.
routing table update is propagated through the network at regular timer intervals;
therefore, the convergence to final routes is slow. In OSPF, an LSA is sent only
when a loss of a route has been detected or when a network comes back online.
The loss is immediately reported to neighbor routers, and new routes are calculated
much faster than with RIP.
3-1: CONFIGURING LINK STATE PROTOCOLS—OSPF 101

Link State Protocols
Link state protocols establish a relationship with a neighboring router. The rout-
Link State Protocol ers exchange LSAs to update neighbors regarding route status. The LSAs are sent
Establishes a relationship only if there is a change or loss in the network routes and the link state protocols
with a neighboring
converge to route selection quickly. This is a distinct advantage over distance vec-
router and uses route
tor protocols that exchange updated routing tables at fixed time intervals and are
advertisements to build
routing tables. slow to converge. In fact, link state routing protocols are replacing distance vector
protocols in most modern networks. Link state protocols are also called shortest-
path first protocols, based on the algorithm developed by E. W. Dijkstra. Link state
protocols use “Hello” packets to verify that communication is still established with
neighbor routers. The key issues of link state protocols are summarized as follows:
• Finds neighbors/adjacencies
• Uses route advertisements to build routing table
• Sends “Hello” packets
• Sends updates when routing changes
OSPF sends small “Hello” packets at regular time intervals to adjacent routers to
Hello Packets
verify that the link between two routers is active and the routers are communicat-
Used in the OSPF ing. If a router fails to respond to a Hello it is assumed that the link or possibly the
protocol to verify
router is down. The OSPF Hello packet captured with a network protocol analyzer
that the links are still
communicating.
is discussed in Section 3-5.
Areas OSPF uses the concept of areas to partition a large network into smaller networks.
The partition of a large
The advantage of this is that the routers have to calculate routes only for their area.
OSPF network into If a route goes down in a given area, only the routers in that area have to calculate
smaller OSPF networks. new routes. Any number between 0 and 4,294,967,295 (232 – 1) can be used; how-
ever, area 0 is reserved for the root area, which is the backbone for the network.
Backbone The backbone is the primary path for data traffic to and from destinations and
The primary path for sources in the campus network. All areas must connect to area 0, and area 0 cannot
data traffic to and from be split. The area numbers can also be expressed in IP notation—for example, area
destinations and sources
0 could be 0.0.0.0—or you can specify an area as 192.168.25.0 or in subnet nota-
in the campus network.
tion. Hence, the need for the large upper-area number (232 – 1) = 255.255.255.255
when converted to a decimal number.
OSPF allows the use of variable length subnet masks (VLSM), which enable dif-
Variable Length ferent size subnets in the network to better meet the needs of the network and more
Subnet Masks (VLSM) efficiently use the network’s limited IP address space. For example, point-to-point
Enables the use of inter-router links don’t need a large block of addresses assigned to them. Figure 3-1
subnet masks to better
illustrates an example of an inter-router link.
fit the needs of the
network, thereby
minimizing the waste FA0/0 FA0/1
of IP addresses when 10.10.250.1 10.10.250.2
interconnecting subnets.
10.10.250.0 Network address
10.10.250.3 Broadcast address
10.10.25.0 Subnet
FIGURE 3-1 An inter-router link subnetted to provide for two host IP addresses, a network
address, and a broadcast address
102 CHAPTER 3: ADVANCED ROUTER CONFIGURATION II

A subnet of size 4 is sufficient for the inter-router link that includes the IP ad-
dresses for the router interfaces, the network address, and the broadcast address.
A subnet mask of 255.255.255.252 meets this requirement of a subnet size 4 and
is permissible in OSPF. This subnet mask provides for the addressing of the two
host addresses (the router interfaces on each end), and the network and broadcast
addresses, which provides the total subnet size of 4. This is an important advantage
of OSPF because using variable length subnet masks minimizes the waste of IP ad-
dresses when interconnecting subnets. Table 3-1 summarizes the advantages and
disadvantages of OSPF.
TABLE 3-1 Summary of Advantages and Disadvantages of OSPF
Advantages Disadvantages
Not proprietary—available for use by Can be complicated to implement.
all vendors.
Link state changes are immediately Is process intensive due to routing table
reported, which enables rapid calculations.
convergence.
Consumes little network bandwidth. Intermittent routes that are going up and Route Flapping
down will create excessive LSA updates—this
Intermittent routes
is called route flapping. going up and down
Uses VLSM creating excessive LSA
updates.
Uses areas to partition the network
into smaller networks, minimizing
the number of route calculations.
Configuring Routes with OSPF

This section describes a procedure for configuring OSPF on a router. The first ex-
ample is for configuring the three routers in the campus LAN shown in Figure 3-2.
The routers will be configured to run OSPF on each of the router’s three Ethernet
interfaces. The example begins with configuring Router A. Router A must first be
placed in the router’s configuration mode [Router(config)#], as shown:
RouterA#conf t
Enter configuration commands, one per line. End with CNTL/Z.
RouterA(config)#
The next step is to enter the information about the IP address for each of the Ether-
net interfaces. The IP addresses for Router A are as follows:
FastEthernet0/0: 10.10.20.250

10.10.100.0 NET
segment
RouterA E2 RouterB RouterC E2

10.10.200.0 NET 10.10.150.0 NET
E1 E2 E3 E1
E0 segment E0 E1 segment E0
10.10.20.0 NET 10.10.10.0 NET 10.10.1.0 NET 10.10.5.0 NET

LAN A LAN B LAN C LAN D
Hub Switch Switch Hub
A1 A2 B1 B2 C1 C2 D1 D2
RouterA Port Addresses (LAN A) RouterB Port Addresses (LAN B) RouterB Port Addresses (LAN C) RouterC Port Addresses (LAN D)
Ethernet 0 10.10.20.250 Ethernet 0 10.10.10.250 Ethernet 1 10.10.1.250 Ethernet 0 10.10.5.250

(Gateway Address) (Gateway Address) (Gateway Address) (Gateway Address)
LAN A Computers LAN B Computers LAN C Computers LAN D Computers
Computer A1 10.10.20.1 Computer B1 10.10.10.1 Computer C1 10.10.1.1 Computer D1 10.10.5.1

Subnet Mask 255.255.255.0 Subnet Mask 255.255.255.0 Subnet Mask 255.255.255.0 Subnet Mask 255.255.255.0
FIGURE 3-2 The three-router campus LAN
A subnet mask of 255.255.255.0 is assigned to each of the FastEthernet interfaces.

After the FastEthernet interfaces are configured, verify the configuration settings
using the sh ip int brief command, as shown. Make sure that the status for the
FastEthernet interfaces are up. This indicates that the interfaces are turned on and
an Ethernet networking device is connected. The protocol will show down until the
Ethernet cable is connected and the connecting interface is enabled. In this case, the
connecting interfaces to FA0/1 and FA0/2 are not enabled and, therefore, show a
status of down.
RouterA#sh ip int brief
Interface IP-Address OK? Method Status Protocol
FastEthernet0/0 10.10.20.250 YES manual up up
FastEthernet0/1 10.10.200.1 YES manual up down
Next, the command router ospf process id is used to enable OSPF routing. In this
router ospf [process case, the command router ospf 100 is entered. The 100 is the process id number.
id]
This number must be the same on each router for OSPF to exchange routes. The
The command used to
process ID number is selected by the network administrator and is not used for rout-
enable OSPF routing.
ing outside the network. It is customary to use the same process ID throughout the
network for ease of management, but it is not required. Entering the router ospf
100 command places the router in the RouterA(config-router)# prompt:
RouterA(config)#router ospf 100
RouterA(config-router)#

The next step is to define the network running OSPF by entering the network com-
mand followed by the IP address of the interface, the OSPF wild card bits, and then
an area number. The following text shows each step and lists the results of entering
a question mark as the command is entered. When entering the command network
?, the router prompts you to enter the IP address of the interface:
RouterA(config-router)#network ?
A.B.C.D Network number
A.B.C.D is the IP address or network number for the Ethernet interface. Next, en-
tering network 10.10.20.250 ? prompts you to enter the OSPF wild card bits in the Network Number
form of A.B.C.D. The wild card bits, also called the inverse mask bits, are used to Another name for the IP
match the network IP address (A.B.C.D format) to interface IPs. If there is a match, subnet.
the subnet on the interface is advertised out OSPF, and OSPF packets are sent out Wild Card Bits
the interface. A 0 wild card bit is used to indicate a “must” match. A 255 is a “don’t Used to match network
care,” hence the name inverse mask. IP addresses to interface
IPs.
The last entry when defining an OSPF route is for the area. Remember, areas are
used to partition a large network into smaller networks. Area 0 is the root area and Area 0
is the backbone for the network. All other areas must connect to area 0. Area 0 can- In OSPF, this is the root
not be split. Other area numbers are specified by the network administrator. area and is the backbone
for the network.
Assume that the router command network 10.10.20.250 0.0.0.0 area 0 is entered.
The wild card bits indicate that any interface with an address of 10.10.20.250 must
run OSPF on the interface and will be assigned to area 0 (the network backbone).
Assume that the router command network 10.10.20.250 0.255.255.255 area 0 is
entered. The wild card bits indicate that any interface with an IP address of 10.x.x.x
must run OSPF on the interface and will be assigned to area 0:
RouterA(config-router)#network 10.10.20.250 ?
A.B.C.D OSPF wild card bits
RouterA(config-router)#network 10.10.20.250 0.0.0.0 ?
area Set the OSPF area ID
RouterA(config-router)#network 10.10.20.250 0.0.0.0 area 0
The following command sequence details the three OSPF network entries needed to
configure OSPF routing for Router A:
Note that the Router A interface to LAN A (10.10.20.250 NET) is listed when con-
figuring OSPF. This is used in OSPF to advertise the LAN to the other routers. Also
note that the network has been assigned to area 0 (the backbone). The command sh
ip int brief is used to check the status of the interfaces. The output Protocol down
indicates that the cable to the interface is either unplugged or the interface is shut
down:
RouterA#show int brief
FastEthernet0/0 10.10.20.250 YES NVRAM up up

This problem with the Protocol down is fixed by reestablishing the physical con-
nection between the routers.
The next step is to configure RouterB. First, configure the four FastEthernet in-
terfaces on RouterB. Next, the OSPF routing protocol for RouterB is set. In this
example, one command-line instruction is used to configure RouterB to run OSPF
on all four of its interfaces. This is done with a subnet mask or wild card in OSPF.
First, enter RouterB’s configuration mode using the conf t command. The com-
mand router ospf 100 is entered. Note that the same process ID number of 100 is
being used. The next step is to enter network 10.0.0.0 0.255.255.255 area 0. This
command tells the router that any address that starts with a “10” belongs to area 0
on RouterB:
RouterB#conf t
RouterB(config)#router ospf 100
RouterB(config-router)#network 10.0.0.0 0.255.255.255 area 0
Verify that the interfaces are properly configured using the sh ip int brief com-
mand, as shown:
RouterB#sh ip int brief
The FastEthernet0/3 interface shows the protocol is down because the connecting
interface is shut down on RouterC.
The next step is to configure RouterC. The OSPF routing protocol for RouterC
is set using the command router OSPF 100 followed by network 10.0.0.0
0.255.255.255 area 0, as shown:
RouterC(config)#router ospf 100
RouterC(config-router)#network 10.0.0.0 0.255.255.255 area 0
The interfaces on RouterC are checked using the sh ip int brief command, as
shown:
RouterC#sh ip int brief
Notice that the Protocol column shows up for all interfaces. This is because all in-
terfaces are connected and the interfaces are enabled.

The following is a partial listing of the running-configuration file on Router A that
shows the router OSPF network configuration. Similar information will appear on
Routers B and C:
.
.
router ospf 100
network 10.10.200.1 0.0.0.0 area 0
network 10.10.20.250 0.0.0.0 area 0
network 10.10.100.1 0.0.0.0 area 0
.
.
The routing table for Router A can be checked using the command sh ip route, as
shown. The routing table indicates there are seven subnets in the campus network
shown in Figure 3-2. The Os indicate the subnets running OSPF, and C indicates
the subnets directly connected to the router:
RouterA#sh ip route
Codes: C connected, S static, I IGRP, R RIP, M mobile, B BGP D EIGRP,
EX EIGRP external, O OSPF, IA OSPF inter area
N1 OSPF NSSA external type 1, N2 OSPF NSSA external type 2
E1 OSPF external type 1, E2 OSPF external type 2, E EGP
i IS-IS, L1 IS-IS level-1, L2 IS-IS level-2, * candidate default
U per-user static route, o ODR T traffic engineered route
Gateway of last resort is not set
10.0.0.0/24 is subnetted, 7 subnets
O 10.10.5.0 [110/74] via 10.10.100.2, 00:03:28, FastEthernet0/2
C 10.10.20.0 is directly connected, FastEthernet0/0
[110/128] via 10.10.100.2, 00:03:28,
FastEthernet0/2
To display only the OSPF routes, you can use the sh ip route ospf command. The
following are the results for this command from Router A:
RouterA#sh ip route ospf
[110/128] via 10.10.100.2, 00:10:03,
FastEthernet0/2

Another command used for displaying protocol information for the router is sh ip
protocol. The following are the results for entering this command for Router A:
RouterA#sh ip protocol
Routing Protocol is “ospf 100” Sending updates every 0 seconds
Invalid after 0 seconds, hold down 0, flushed after 0
Outgoing update filter list for all interfaces is Incoming update
filter list for all interfaces is Redistributing: ospf 100
Routing for Networks:
10.10.20.250/32
10.10.100.1/32
10.10.200.1/32
Routing Information Sources:
Gateway Distance Last Update
10.10.100.1 110 00:06:01
10.10.200.2 110 00:06:01
Distance: (default is 110)
To verify the adjacent OSPF neighbors of the router and observe the neighbor state,
you can use the sh ip ospf neighbor command. This command displays the directly
connected neighbors, since the hello packets are exchanged among them. The fol-
lowing are the results for this command from Router A:
RouterA#sh ip ospf neighbor
Neighbor ID Neighbor ID Pri State Dead Time Address
Interface
The highest IP address
defined by the loopback 10.10.150.2 1 FULL/BDR 00:00:39 10.10.100.2
FastEthernet0/2
address of the neighbor
router ID. 10.10.200.2 1 FULL/DR 00:00:31 10.10.200.2
FastEthernet0/1
Designated Router
(DR) The show ip ospf neighbor command shows there are two adjacent neighbors:
The router with the 10.10.150.2 and 10.10.200.2. The Neighbor ID is the highest IP address defined
highest priority. by the loopback address of the neighbor router ID. In this case, 10.10.150.2 is the
Backup Designated highest IP address for Router C, and 10.10.200.2 is the highest IP address of Router
Router (BDR) B. Refer to Figure 3-2 for the IP assignments for each router. The Pri column
The router or routers stands for Priority, which indicates the priority of the neighbor routers. The router
with lower priority. with the highest priority becomes the Designated Router (DR), and the router or
State of FULL routers with lower priority will become a Backup Designated Router (BDR). In this
case, the priority is the same, so the router with the highest router ID becomes the
Indicates that the
routers are fully adjacent
DR, which is Router B, and Router C becomes the BDR. This process is called
to each other. a DR Election. It happens during the exchange of Hello packets. The main pur-
pose of the DR router is to maintain the OSPF topology table of the network. This
Dead Time way, there is only one source that can send the OSPF routing updates to the rout-
The Hello time interval ers within the area, thus minimizing the network traffic. Every router also sends
an OSPF router will its update to the DR and BDR via the multicast address. If the DR should fail, the
wait before terminating BDR will assume the role of the designated router. The state of FULL is a good and
adjacency with a
desired state, which indicates that the routers are fully adjacent to each other. The
neighbor.
dead time is the Hello time interval an OSPF router will wait before terminating

adjacency with a neighbor. The address is the IP address of its directly connected
neighbor and the interface indicates the interface on which the neighbor is con-
nected.
Load Balancing and Redundancy with OSPF

As long as there is an equal cost path, traffic load balancing on a per-packet basis
can be configured with OSPF. For example, if there is an additional link of the
same type interface and speed between Router A and Router B (refer to Figure 3-2),
OSPF will automatically load balance the network traffic to the network 10.10.10.0
NET between the two links because they both have the same OSPF cost. To find out
the OSPF cost of an interface, use the sh ip ospf interface command. The follow-
ing example shows the OSPF interface information on FastEthernet0/2, including
its interface cost of 1.
RouterA#sh ip ospf interface
FastEthernet0/2 is up, line protocol is up
Internet Address 10.10.100.1/24, Area 0
Process ID 100, Router ID 10.10.200.1, Network Type BROADCAST, Cost:
1
Transmit Delay is 1 sec, State DR, Priority 1
Designated Router (ID) 10.10.200.1, Interface address 10.10.100.1
Backup Designated router (ID) 10.10.150.2, Interface address
10.10.100.2
:
:
With the Shortest Path First protocol, OSPF uses the Dijkstra algorithm to calculate
the shortest path. The algorithm takes into consideration the topology path and the
associated cost and calculates the cumulative cost to reach a destination.
With destinations with multiple paths, OSPF will calculate the best shortest path
and install the route into the routing table. For the example network (refer to Figure
3-2), Router A has two paths to reach the 10.10.10.0 network on Router B. One
route is via FastEthernet0/1 to Router B and another route is via FastEthernet0/2 to
Router C and then to Router B. The output of show ip route shows that the Router
A has selected the path via its FastEthernet0/1 as the best path and this is confirmed
when issuing the command sh ip route 10.10.10.0.
RouterA#sh ip route 10.10.10.0
Routing entry for 10.10.10.0/24
Known via “ospf 100”, distance 110, metric 2, type intra area
Last update from 10.10.200.2 on FastEthernet0/1, 00:44:13 ago
Routing Descriptor Blocks:
* 10.10.200.2, from 10.10.200.2, 00:44:13 ago, via FastEthernet0/1
Route metric is 2, traffic share count is 1

When the path via the interface FastEthernet0/1 becomes unavailable, the OSPF
database gets updated, and another available route via RouterC is automatically se-
lected, as shown here:
*Mar 1 01:31:40.399: %LINEPROTO-5-UPDOWN: Line protocol on Interface
FastEthernet0/1, changed state to down
*Mar 1 01:31:41.411: %OSPF-5-ADJCHG: Process 100, Nbr 10.10.200.2 on
FastEthernet0/1 from FULL to DOWN, Neighbor Down: Interface down or
detached
Because OSPF uses the interface cost as one of the factors in calculating the best
Interface Cost path, one of the easiest ways to influence the OSPF path selection is by manipulat-
One of the factors used ing the interface cost. In our example, there are two equal routes to the network
in calculating the best 10.10.150.0 Net, as shown.
path.
10.10.200.2, from 10.10.200.2, 00:00:05 ago, via FastEthernet0/1
If the preferred route to reach the 10.10.150.0 network is via Router B, then the
OSPF cost of the interface connected to Router C can be increased. The default
OSPF cost of the interface is 1.To change the cost of this interface, issue the ip ospf
cost command at the interface, as shown:
RouterA(config)#int fastEthernet 0/2
RouterA(config-if)#ip ospf cost 5
By increasing the OSPF cost of interface0/2, the route to the network 10.10.150.0 is
being preferred via the interface FastEthernet0/1 off Router B. However, changing
the OSPF cost could have side effects if you are not careful. In our case, because
the interface to Router C now has a higher OSPF cost, what will happen to the LAN

D or 10.10.5.0 network? The following output shows the current routing table after
the OSPF cost has changed:
As you can see, the route to LAN D is now preferred through Router B. This means
one extra router hop that it must take from Router A to reach LAN D. By increas-
ing the interface cost, it affects everything along its path. This is definitely not the
desirable side effect that you want. The network engineer must understand the net-
work topology before adjusting the OSPF cost, Administrative Distance (AD), or
any routing protocol metrics. This has to be carefully done and thoroughly verified;
otherwise, it can result in a routing abnormality such as this.
Networking Challenge—OSPF
Use the Net-Challenge Simulator Software included with the text’s companion CD-ROM to demon-
strate that you can configure OSPF for Router A in the campus LAN (the campus LAN is shown in
Figure 3-2 and is displayed by clicking the View Topology button when the software is started). Place
the Net-Challenge CD-ROM in your computer’s drive. Open the Net-Challenge folder and click
NetChallenge V3-2.exe. When the software is running, click the Select Router Challenge button to
open a Select Router Challenge drop-down menu. Select Chapter 3—OSPF. This opens a checkbox
that can be used to verify that you have completed all the tasks:
1. Enter the privileged EXEC mode on the router.
2. Enter the router’s terminal configuration mode: Router(config).
3. Set the hostname to Router A.
4. Configure the FastEthernet0/0 interface with the following:
IP address: 10.10.20.250
Subnet mask: 255.255.255.0
5. Enable the FA0/0 interface.
IP address: 10.10.200.1
Subnet mask: 255.255.255.0
IP address: 10.10.100.1
Subnet mask: 255.255.255.0

10. Enable OSPF with a network number of 100.
11. Use a single command-line instruction to configure Router A to run OSPF on all three of the
FastEthernet interfaces (use area 100).
12. Use the sh ip int brief command to check the interface status.
13. Use the sh ip protocol command to see whether OSPF is running on Router A.
14. Use the sh ip route command to verify that the three FastEthernet ports are connected to Router A.
15. Use the sh run command to view the running-configuration file on Router A. Verify that OSPF is
enabled and the proper network address is specified.
16. The preferred route to reach the 10.10.150.0 network is via Router B. Issue the command to increase
the OSPF cost of the interface connected to Router C to 5 using the ip ospf cost command.
L1 Routers 3-2 CONFIGURING LINK STATE PROTOCOLS—IS-IS

Analogous to OSPF
nonbackbone routers. There are many similarities between IS-IS and OSPF. Both protocols use the link
state protocol with the Dijkstra algorithm. Both are classless protocols, which en-
L2 Routers able the support of VLSM. Both protocols use Hello packets to form and maintain
Analogous to OSPF adjacencies and both use the area concept. However, there is a difference in the way
backbone routers. in which the areas are defined for each protocol. In IS-IS, there are two hierarchical
L1/L2 Routers topology areas: level 1 (Intra-area) and level 2 (Inter-area). A router can either be
Analogous to OSPF area a level 1 (L1) router, a level 2 (L2) router, or both (L1/L2) routers. L1 routers are
border routers. analogous to OSPF nonbackbone routers, L2 routers are analogous to OSPF back-
bone routers, and L1/L2 routers are analogous to OSPF area border routers (ABR).
ABR Unlike OSPF ABRs, L1/L2 routers do not advertise routes from L2 routers to L1
Area border routers. routers. The packets from different areas can only be routed through the L1/L2 rout-
Connectionless ers. Essentially, L1/L2 routers are default gateways to L1 routers. Another big dif-
Network Service ference is that the IS-IS backbone area can be segmented. Unlike the backbone area
(CLNS) in OSPF, all routers in area 0 must be connected; the IS-IS L2 routers do not need to
IS-IS is designed to be connected directly together.
work on the same
network layer just like IS-IS was originally designed as part of the Open System Interconnection (OSI)
IP; therefore, it does not network layer service called Connectionless Network Service (CLNS). This means
require an IP protocol that IS-IS is designed to work on the same network layer just like IP; therefore, it
for it to function. does not require IP protocol for it to function. Later, it was adapted to work with IP.
Hence, it is sometimes referred to as integrated IS-IS. In IS-IS, every router uses the
Integrated IS-IS
Network Entity Title (NET) to define its process. The NET address is unique to each
Term indicating IS-IS
router; it is comprised of the following components in hexadecimal format:
was adapted to work
with IP. • The area ID in IS-IS is analogous to OSPF area number, and it is used by L2
NET routers.
Network Entity Title.
Area ID
Analogous to OSPF area
number, and it is used
by L2 routers.

• The system ID is analogous to the OSPF router ID, and it is used by L1 routers.
System ID
• The Network Service Access Point Selector (NSEL) identifies the network ser- Analogous to the OSPF
vice type. router ID, and it is used
by L1 routers.
The NET address can look intimidating because of its long hexadecimal format, but Network Service
it is not as bad as it seems. The way to work with a NET address is to start from right Access Point Selector
and work left. For example, given a NET address of 49.0001.0014.a909.5201.00, the (NSEL)
last 1 byte from the right is NSEL, which is always set to 00 on a router. The next Identifies the network
six bytes from the right are separated into three groups of 2 bytes and define the sys- service type.
tem ID. In this example, the system ID is 0014.a909.5201. The system ID is always
unique and is typically represented as the MAC address of the router. The remaining IS-IS
numbers to the left of the System ID define the Area ID, which is 49.0001. The area Interior Gateway Routing
ID has a variable length, but its first number must be at least 1 byte long. Protocol.
Configuring Routes with IS-IS

This section demonstrates how to configure the routers to use the IS-IS routing pro-
tocol for the three-router campus network shown in Figure 3-3. For this exercise,
RouterB has been turned off. This is reflected in the output of the command show
ip int brief. Notice that the Router A interface FastEtherent0/1, which connects
RouterB, is showing Status down and Protocol down.
RouterA#show ip int brief
FastEthernet0/1 10.10.200.1 YES manual down down
10.10.100.0 NET
segment

10.10.200.0 NET 10.10.150.0 NET
E1 E2 E3 E1
segment segment
E0 E0 E1 E0
10.10.20.0 NET 10.10.10.0 NET 10.10.1.0 NET 10.10.5.0 NET

A1 A2 B1 B2 C1 C2 D1 D2


FIGURE 3-3 Routers A, B, and C in the three-router campus network
3-2: CONFIGURING LINK STATE PROTOCOLS—IS-IS 113

The first step in configuring Router A to use IS-IS is to enable the protocol globally.
cnls routing As mentioned earlier, IS-IS is a part of CNLS as its network layer routing protocol.
The global command for When configuring most routing protocols, the global command of ip routing must
IS-IS. be issued. With IS-IS, the global command of cnls routing must be issued. Then,
router isis at the RouterA(config)# prompt, enter the command router isis to start using the
Starts the IS-IS routing IS-IS routing protocol. Next, the NET address will need to be entered to specify the
protocol. area the router belongs to as well as defining the System ID and the NSEL of 00.
Unlike other routing protocols, to specify the network that will be using IS-IS for
ip router isis
routing, this will need to be configured at each individual interface with the com-
Specifies the network
mand ip router isis. These steps are as follows:
that will be using IS-IS
for routing. RouterA(config)#cnls routing
RouterA(config)#router isis
RouterA(config-router)# net 49.0001.c202.00e8.0000.00
RouterA(config-router)#interface fastethernet0/0
RouterA(config-router)#ip router isis
RouterA(config-router)#interface fastethernet0/2
RouterA(config-router)#ip router isis
On Router A, the interface FastEthernet 0/0 and the interface Fastethernet 0/2 are
configured to use IS-IS. The same set of commands is issued on Router B and
Router C to route with IS-IS. You can verify that IS-IS is running on Router A by
entering the command sh ip protocol, as shown at the enable prompt (RouterA#):
RouterA# sh ip protocol
Routing Protocol is “isis”
Outgoing update filter list for all interfaces is not set
Incoming update filter list for all interfaces is not set
Redistributing: isis
Address Summarization:
None
Maximum path: 4
FastEthernet0/0
FastEthernet0/1
FastEthernet0/2
10.10.100.2 115 00:02:44
10.10.150.1 115 00:02:44
Distance: (default is 115)
The text shows that the routing protocol is “isis.” The networks it is routing are on
the interface FastEthernet0/0, FastEthernet0/1, and FastEthernet0/2. It also shows
that it is receiving routing information from the source 10.10.100.2, which is the
Router C interface FastEthernet0/2 and 10.10.150.1, which is the Router B interface
FastEthernet0/3. The routing information was updated 2 minutes 44 seconds ago.
Next, the command show ip route (sh ip route) can be used to examine the con-
tents of Router A’s routing table to see if any IS-IS routes have been received:

RouterA# sh ip route
Codes: C - connected, S - static, R - RIP, M - mobile, B - BGP
D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area
N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2
E1 - OSPF external type 1, E2 - OSPF external type 2
i - IS-IS, su - IS-IS summary, L1 - IS-IS level-1, L2 - IS-IS
level-2
ia - IS-IS inter area, * - candidate default, U - per-user
static route
o - ODR, P - periodic downloaded static route

i L1 10.10.1.0 [115/20] via 10.10.200.2, FastEthernet0/1
[115/20] via 10.10.200.2, FastEthernet0/1
The command sh ip route indicates that the network 10.10.5.0 has been received
via the IP 10.10.100.2, which is directly connected to the interface FastEthernet0/2. show ip route isis
(sh ip route isis)
The routes for network 10.10.1.0 and 10.10.10.0 are learned via the IP 10.10.200.2
on interface FastEthernet0/1. The network 10.10.150.0 is learned via both FastEth- Displays only the IS-IS
routes.
ernet0/1 and FastEthernet0/2. As a matter of fact, the routing table looks similar to
the one of OSPF. The “i” before the route indicates IS-IS routes, and “L1” stands
for IS-IS level 1. The command show ip route isis can be used to view the routing
table just for IS-IS, as shown:
RouterA#sh ip route isis
[115/20] via 10.10.200.2, FastEthernet0/1
Similar to an OSPF router, IS-IS establishes its adjacencies with its neighbors and
the neighbor information can be found with the command show clns is-neighbors. show clns
is-neighbors
RouterA#sh clns is-neighbors
Command used in IS-IS
to find adjacencies with
System Id Interface State Type Priority Circuit Id
neighbors.
Format
RouterB Fa0/1 Up L1L2 64/64 RouterA.02
Phase V
RouterC Fa1/0 Up L1L2 64/64 RouterA.03
Phase V

IS-IS is identified by its NET address not the IP address. Therefore, the network
administrator will need to remember the NET address. This can be a challenge,
especially in a big network but IS-IS does provide dynamic hostname exchange.
This allows the ASCII router name to be exchanged among the routers; therefore,
it can be used to associate with the system instead of the NET address. An example
of the IS-IS dynamic hostname exchange is next shown resulting from the sh clns
is-neighbor command. Notice that the router names are displayed in the System ID
and Circuit ID fields.
One confusing issue regarding IS-IS is when to use the command isis or clns. The
clns commands are used when dealing with the protocol at the Layer 2 level. The
isis commands are used when dealing with IP information. For example, the similar
neighbor information can be displayed with command show isis neighbor. The
differences are the IP address and IS-IS level type information are displayed using
show isis neighbor.
RouterA#sh isis neighbors
System Id Type Interface IP Address State Holdtime Circuit Id
RouterB L1 Fa0/1 10.10.200.2 UP 28 RouterA.02
RouterB L2 Fa0/1 10.10.200.2 UP 20 RouterA.02
RouterC L1 Fa1/0 10.10.100.2 UP 20 RouterA.03
RouterC L2 Fa1/0 10.10.100.2 UP 22 RouterA.03
The final step is to view the running-configuration in each router using the sh run
command. The running-configuration for Router A is provided. Notice the entries
for router isis and net 49.0001.c202.00e8.000.00, and ip router isis in the
running-configuration file.
RouterA#sh run
Building configuration...
Current configuration:
!
version 12.0
service timestamps debug uptime service timestamps log uptime no
service password-encryption
!
hostname RouterA
!
enable secret 5 $1$6M4r$dleo7h1WP0AYu0K/cM6M91
enable password Salsa
!
ip subnet-zero
!
interface FastEthernet0/0
ip address 10.10.20.250 255.255.255.0
ip router isis
no ip directed-broadcast
!
ip address 10.10.200.1 255.255.255.0

!
ip address 10.10.100.1 255.255.255.0
ip router isis
!
router isis
net 49.0001.c202.00e8.0000.00
!
ip classless
!
line con 0
transport input none line aux 0
line vty 0 4 password chile login
!
end
Remember to use the copy run start command to save the changes made to
NVRAM.
Load Balancing and Redundancy with IS-IS

Because IS-IS is another SPF protocol, its load balancing and redundancy capabili-
ties are almost identical to those used by OSPF. IS-IS can load balance the net-
work traffic over equal cost paths. On the unequal cost paths, only the best path is
installed into the routing table and that path becomes the primary route where the
network traffic will flow. When the primary link fails, the routes will converge to
the secondary link until the primary link is back in operation. All of these examples
have been shown within the OSPF section.
These same concepts explored in the OSPF section, as well as the cautions, still
apply to IS-IS. The only big differences between the two protocols usually are the
command syntax used in configuring some of the functions. In OSPF, an OSPF
cost is the value assigned to the interface, and its default value is 1. In IS-IS, it is a
metric that is assigned to the interface and its default metric value is 10. The IS-IS
metric can be adjusted at the interface level with the following command sequence:
RouterA(config-if)#isis metric 30
By increasing the IS-IS metric on the FastEthernet0/2 on Router A, the same results
happen as when increasing the OSPF cost on this very same interface. That is all
routes destined to any networks on Router C now prefer the interface FastEther-
net0/1 via Router B as their primary route or best path. The result of adjusting the
IS-IS metric is shown via command show ip route isis:

To verify the IS-IS interface metric, issue the show clns interface command. The
show clns interface output displays the interface FastEthernet0/2, which now has the level1-metric and
The command used to level2-metric of 30:
verify the IS-IS interface
metric. RouterA#sh clns interface
:
FastEthernet0/2 is up, line protocol is up
Checksums enabled, MTU 1497, Encapsulation SAP
ERPDUs enabled, min. interval 10 msec.
RDPDUs enabled, min. interval 100 msec., Addr Mask enabled
Congestion Experienced bit set at 4 packets
CLNS fast switching enabled
CLNS SSE switching disabled
DEC compatibility mode OFF for this interface
Next ESH/ISH in 19 seconds
Routing Protocol: IS-IS
Circuit Type: level-1-2
Interface number 0x2, local circuit ID 0x3
Level-1 Metric: 30, Priority: 64, Circuit ID: RouterA.03
Level-1 IPv6 Metric: 10
Number of active level-1 adjacencies: 1
Level-2 Metric: 30, Priority: 64, Circuit ID: RouterA.03
Level-2 IPv6 Metric: 10
Number of active level-2 adjacencies: 1
Networking Challenge: IS-IS

Use the Net-Challenge Simulator Software included with the text’s companion CD-ROM to demonstrate
that you can configure IS-IS for Router A in the campus LAN. (The campus LAN is shown in Figure
3-3 and is displayed on the computer screen after the software is started.) Place the Net-Challenge CD-
ROM in your computer’s drive. Open the Net-Challenge folder, and click NetChallenge V3-2.exe.
When the software is running, click the Select Router Challenge button to open a Select Router Chal-
lenge drop-down menu. Select Chapter 3—IS-IS. This opens a checkbox that can be used to verify that
you have completed all the tasks. A check next to a task indicates that the task has been successfully
completed:
1. Click the Router A button.
3. Enter the router configuration mode, Router(config).
4. Set the hostname to RouterA.
IP address: 10.10.20.2
Subnet mask: 255.255.255.0

IP address: 10.10.200.1
Subnet mask: 255.255.255.0
IP address: 10.10.100.1
Subnet mask: 255.255.255.0
11. Enable CNLS routing.
12. Enable IS-IS on Router A.
13. Use the net command to specify the NET address to be used by IS-IS.
14. Enable IS-IS on FA0/0 and FA0/2.
16. Use the sh ip protocol to see whether IS-IS is running.
17. Use the sh ip route command to verify that the four FastEthernet ports are connected to the router.
18. Use the sh run command to view the running-configuration file. Verify that IGRP is enabled and
the proper network address is specified.
19. Copy the running-configuration to the startup-configuration.
20. Display the contents of the startup configuration.
3-3 CONFIGURING HYBRID ROUTING

PROTOCOLS—EIGRP
This section introduces techniques for configuring a router’s interface to run EIGRP,
EIGRP
the Enhanced Interior Gateway Routing Protocol. EIGRP is an enhanced version of
the Interior Gateway Routing Protocol (IGRP). EIGRP is a Cisco proprietary pro- Enhanced Interior
Gateway Routing
tocol and is often called a hybrid routing protocol that incorporates the best of the
Protocol.
distance vector and link-state algorithms.
Hybrid Routing
EIGRP allows the use of variable length subnet masks, which is beneficial when Protocol
you’re trying to conserve the use of IP addresses. EIGRP also uses Hello packets Protocol that
to verify that a link from one router to another is still active. This is similar to the incorporates the best of
OSPF Hello packet described in Section 3-5. The routing table updates are ex- the distance vector and
changed when there is a change in the network. In other words, the routers don’t link-state algorithms.
exchange unnecessary information unless a route changes. This helps conserve the
limited bandwidth of the network data link. When route information is exchanged,
EIGRP quickly converges to the new route selection.
3-3: CONFIGURING HYBRID ROUTING PROTOCOLS—EIGRP 119

The four components of EIGRP are as follows:
• Neighbor Discovery Recovery: Used to learn about other routers on di-
rectly attached networks. This is also used to discover whether neighbor rout-
ers are unreachable. This discovery is accomplished by periodically sending
Hello packets. The Hello packets are used to verify that a neighbor router is
functioning.
• Reliable Transport Protocol: Used to guarantee delivery of EIGRP packets to
neighbor routers. Both unicast and multicast packet transmission are supported.
• DUAL Finite State Machine: Used to track all routes advertised by its neigh-
bors and is used for route computation to obtain loop-free routing.
• Protocol Dependent Modules: Responsible for handling network layer,
protocol-specific requirements. For example, the IP-EIGRP module is re-
sponsible for extracting information from the EIGRP packets and passing this
information to DUAL. DUAL uses this information to make routing deci-
sions, and IP-EIGRP then redistributes the learned routes.
Configuring Routes with EIGRP

This section describes a procedure for configuring EIGRP on a router. The first ex-
ample is for configuring Router A in the campus LAN shown in Figure 3-4.
10.10.100.0 NET
segment
10.10.200.0 NET 10.10.150.0 NET
E1 E2 E3 E1
E0 E0 E1 E0
segment segment
10.10.20.0 NET 10.10.10.0 NET 10.10.1.0 NET 10.10.5.0 NET

A1 A2 B1 B2 C1 C2 D1 D2


FIGURE 3-4 The three-router campus LAN

The first step is to configure the interfaces on each of the three routers. Table 3-2
lists the IP addresses and the subnet masks for the router interfaces.
TABLE 3-2 Router Interface IP Addresses and Subnet Masks
Interface IP Address Subnet Mask

Router A
FA0/0 10.10.20.250 255.255.255.0
FA0/1 10.10.200.1 255.255.255.0
FA0/2 10.10.100.1 255.255.255.0
Router B
FA0/0 10.10.10.250 255.255.255.0
FA0/1 10.10.1.250 255.255.255.0
FA0/2 10.10.200.2 255.255.255.0
FA0/3 10.10.150.1 255.255.255.0
RouterC
FA0/0 10.10.5.250 255.255.255.0
FA0/1 10.10.150.2 255.255.255.0
FA0/2 10.10.100.2 255.255.255.0
After configuring the router interfaces, the EIGRP routing protocol for Router A
will be configured. Use the conf t command to enter the router’s configuration
mode. Next, enter the command router eigrp [AS-number]. Any AS number can
be used. The router uses the AS numbers to determine which routers share routing
tables. Only routers with the same AS number will share routing updates. The com-
mand router eigrp 150 is entered as shown. The prompt changes to (config-router)
and the next command entered sets the network to run EIGRP. In this example, the
command network 10.0.0.0 is entered. This instructs the router to run EIGRP on
any of the router’s interfaces that have an IP address that begins with 10. A different
network command will be used on Router B to show how the command can be used
to specify a limited IP address range.
RouterA(config)#router eigrp 150
RouterA(config-router)#network 10.0.0.0
Now, the 10.x.x.x interfaces on Router A are configured to run EIGRP. The com-
mand sh ip protocol is entered to verify that the EIGRP routing protocol is enabled
on Router A:
RouterA# sh ip protocol
Routing Protocol is “eigrp 150”
filter list for all interfaces is Default networks flagged in outgoing
updates Default networks accepted from incoming updates EIGRP metric
weight K1 1, K2 0, K3 1, K4 0, K5 0

EIGRP maximum hopcount 100
EIGRP maximum metric variance 1
Redistributing: eigrp 150
Automatic network summarization is in effect
10.0.0.0
10.10.200.2 90 00:00:09
Distance: internal 90 external 170
The top line states that the routing protocol is “eigrp 150” and indicates that it has
been 9 seconds since the last update to the routing table. In EIGRP, updates to the
routing table are made when there are changes in the network.
Another useful command is sh ip route. The following are the results of entering
the command. In this case, the router does not show any EIGRP routes to the sub-
nets in the network, because EIGRP has not been configured on Router B or
Router C:
RouterA#sh ip route
The command sh ip int brief is entered and the status and protocols for the Ether-
net interfaces are up. These will show up as long as there is a network connection
to the interfaces:
RouterA#sh ip int brief
The sh run command is used to view the contents of the router’s running-
configuration file. The following shows the part of the configuration file that shows
the entries for EIGRP. Notice that these entries are the same as the commands en-
tered earlier when configuring EIGRP:
!
router eigrp 150
network 10.0.0.0
!

The next step is to configure Router B. The configuration mode (config)# for
Router B is entered and the command router eigrp 150 is entered. Remember, 150
is the AS number, which is the same number used when configuring Router A. The
next command is used to set the network that is running EIGRP. In this case, the
command network 10.10.0.0 is entered. This means that interfaces on Router B
with a 10.10.x.x address will run the EIGRP protocol. In this case, all interfaces on
Router B have a 10.10.x.x address and will run EIGRP:
RouterB#conf t
RouterB(config)#router eigrp 150
RouterB(config-router)#network 10.10.0.0
The command sh ip protocol is used to verify that EIGRP is running on Router B.

The text shows that eigrp 150 is running on Router B, and it has been 27 seconds
since the last update to the routing table:
RouterB#sh ip protocol
Routing Protocol is “eigrp 150”
filter list for all interfaces is Default networks flagged in outgoing
updates Default networks accepted from incoming updates EIGRP metric
weight K1 1, K2 0, K3 1, K4 0, K5 0
EIGRP maximum hopcount 100
EIGRP maximum metric variance 1
Redistributing: eigrp 150
Automatic network summarization is in effect
10.0.0.0
10.10.200.1 90 00:00:27
Distance: internal 90 external 170
The sh ip route command for Router B shows that six routes are on Router B, and
there are EIGRP routes to the 10.10.20.0 and 10.10.100.0 subnets. The code for the
EIGRP routes is D. Remember, the C code is for the subnets directly connected to
the router:
RouterB#sh ip route
D 10.10.20.0 [90/2195456] via 10.10.200.1, 00:00:09, FastEthernet0/2

A check of the IP routes on Router A also shows that Router A and Router B are ex-
changing routes with each other. Router A now shows six subnets. Once again, the
D indicates the EIGRP routes, and the C indicates the directly connected subnets:
RouterA#sh ip route
The last step is to configure EIGRP for Router C using the router eigrp 150 com-
mand. The command network 10.10.0.0 is used to instruct the router to assign
EIGRP to all interfaces that are part of the 10.10.0.0 network. In this case, all inter-
faces on Router C have a 10.10.x.x address and therefore will run EIGRP:
RouterC(config)#router eigrp 150
RouterC(config-router)#network 10.10.0.0
The sh ip route command is used to display the IP routes for Router C. Router C
shows seven subnets. In fact, there are seven subnets in the campus LAN shown in
Figure 3-4. This completes the setup for running EIGRP on the campus LAN:
RouterC#sh ip route
10.0.0.0 255.255.255.0 is subnetted, 7 subnets
[90/2681856] via 10.10.100.1, 00:00:01,
FastEthernet0/2

Just like other dynamic routing protocols, one useful command for troubleshooting
routing problems is to look at the router’s neighbors, because that is how routes are show ip eigrp
neighbors
exchanged. The command show ip eigrp neighbors will display the EIGRP adja-
cency neighbor. The command used
to display the EIGRP
RouterA#sh ip eigrp neighbors adjacency neighbor.
IP-EIGRP neighbors for process 150
H Address Interface Hold Uptime SRTT RTO Q Seq
(sec) (ms) Cnt Num
1 10.10.200.2 Fa0/1 10 00:31:06 152 912 0 30
0 10.10.100.2 Fa0/2 11 00:31:51 86 516 0 37
The output result of the show ip eigrp neighbors command is simpler comparing
to other routing protocol neighbor commands. There is no state that one has to de- Uptime
cipher. If the neighbor adjacency is established it will be displayed. The important Indicates the time
fields that one should pay attention to when troubleshooting EIGRP routing are Up- the neighbor has
last established its
time and Q Cnt. The uptime indicates the time the neighbor has last established its
adjacency.
adjacency. If the number is much lower than the others, there could be a neighbor
or connection issue. The Q Cnt is the number of EIGRP packets being queued to its Q Cnt
neighbor. The expected number is 0. If the number is consistently higher, this could The number of EIGRP
mean that there is a link issue or link congestion. packets being queued to
its neighbor.
The output of the show ip eigrp neighbors command consists of the following fields:
• H: The order in which the adjacency is established with the specific neighbor.
• Address: The IP address of the neighbor router interface.
• Interface: The router interface connected to the neighbor.
• Hold: Amount of time EIGRP will wait to hear from its neighbor before declar-
ing it down.
• Uptime: The uptime of the adjacent neighbor.
• SRTT: Smooth Round Trip Time for sending and receiving EIGRP packet from
its neighbor.
• RTO: Retransmission Timeout to resend EIGRP packet as calculated by SRTT.
• Q Cnt: Queue count is the number of EIGRP packets in the queue waiting to be
sent.
• Seq Num: Sequence number of the last update sent by its neighbor.
Load Balancing and Redundancy

EIGRP is a popular routing protocol to use on Cisco routers, not because it is Cisco
proprietary, but because it is easy to configure and manage. Like other dynamic
protocols, EIGRP is capable of load balancing over equal cost paths. By design,
EIGRP can use bandwidth, delay, reliability, load, and MTU as values to calculate
its composite metric to determine the best path to a destination.
By default, only bandwidth and delay are used by Cisco to calculate the composite
metric. Bandwidth is the minimum bandwidth of the entire path. Delay is the total

delay of the entire path. To view the composite metric of the current EIGRP topol-
show ip eigrp ogy, issue the show ip eigrp topology command. The highlighted numbers are the
topology
EIGRP composite metric of each path:
Command used to view
the composite metric RouterA#sh ip eigrp topology
of the current EIGRP IP-EIGRP Topology Table for AS(150)/ID(10.10.200.1)
topology, the command.
Codes: P - Passive, A - Active, U - Update, Q - Query, R - Reply,
r - reply Status, s - sia Status
P 10.10.1.0/24, 1 successors, FD is 30720

via 10.10.200.2 (30720/28160), FastEthernet0/1
via Connected, FastEthernet0/0
It is typically not recommended to manually adjust the EIGRP composite metric.

bandwidth Cisco’s recommendation is to leave the EIGRP as it is; however, if this must be
Command used to adjust done, then the two less destructive ways for setting an EIGRP preferred path is by
the bandwidth value. adjusting the bandwidth value or by adjusting the delay value. To adjust the band-
width value, this can be done via the bandwidth command at the router interface:
RouterA(config-if)#bandwidth 1000
The bandwidth command is in kilobits, so bandwidth 1000 would mean 1,000

Kbits or 1 Mbits as its value. The interface FastEthernet0/2 on Router A is con-
necting Router C. By reducing the bandwidth value, the composite metric of any
route via the interface FastEthernet0/2 has changed as reflected in the recent show
ip eigrp topology command. The composite metric that used to be 30,720 is now
2,565,120:
RouterA#sh ip eigrp topology
IP-EIGRP Topology Table for AS(150)/ID(10.10.200.1)



The show ip route eigrp also confirms that the path via FastEthernet0/1 to Router
B is now the preferred path for all network routes:
RouterA#sh ip route eigrp
D 10.10.1.0 [90/30720] via 10.10.200.2, 00:11:42,
FastEthernet0/1
D 10.10.5.0 [90/33280] via 10.10.200.2, 00:11:42,
FastEthernet0/1
D 10.10.10.0 [90/30720] via 10.10.200.2, 00:11:42,
FastEthernet0/1
D 10.10.150.0 [90/30720] via 10.10.200.2, 00:11:42,
FastEthernet0/1
Changing the bandwidth value could have more impact beyond EIGRP routing;
delay
other protocols, like QoS, depend on the bandwidth value to make their decisions.
Between the bandwidth value and the delay value, the delay value command is the This command sets the
delay on the interface.
one with the least impact. Therefore, to manipulate the EIGRP composite metric,
The delay value is
adjusting the delay value is preferred over adjusting the bandwidth value. To adjust measured in tens of
the bandwidth value, this can be done by adjusting the delay command at the router microseconds.
interface:
RouterA(config-if)#delay 100
The delay value is measured in tens of microseconds. The command delay 100
changes the default delay value from 100 microseconds to 1,000 microseconds.
This, in turn, increases the composite metric of the interface FastEthernet0/2 on
Router A connecting to Router C. The composite metric of any route via the inter-
face FastEthernet0/2 has changed as reflected in the recent show ip eigrp topology
command. The path composite metric that used to be 30,720 is now 53,760. Again,
the path via FastEthernet0/1 to Router B is now a more preferred path for all net-
work routes.
RouterA#sh ip eigrp topology
IP-EIGRP Topology Table for AS(150)/ID(10.10.200.1)


Networking Challenge: EIGRP

Use the Net-Challenge Simulator Software included with the text’s companion CD-ROM to demon-
strate that you can configure EIGRP for Router A in the campus LAN (the campus LAN is shown in
Figure 3-2 and is displayed by clicking the View Topology button when the software is started). Place
the Net-Challenge CD-ROM in your computer’s drive. Open the Net-Challenge folder and click Net
Challenge V3-2.exe. When the software is running, click the Select Router Challenge button to open
a Select Router Challenge drop-down menu. Select Chapter 3—EIGRP. This opens a checkbox that
can be used to verify that you have completed all the tasks:
2. Enter the router configuration mode: Router(config).
3. Set the hostname to RouterA.
IP address: 10.10.20.250
Subnet mask: 255.255.255.0
IP address: 10.10.200.1
Subnet mask: 255.255.255.0

IP address: 10.10.100.1
Subnet mask: 255.255.255.0
10. Enable EIGRP with an AS number of 200.
11. Enter the network command that enables EIGRP on the router.
13. Use the sh ip protocol command to see whether EIGRP is running on Router A.
14. Use the sh ip route command to verify that the three FastEthernet ports are connected to RouterA.
15. Use the sh run command to view the running-configuration file on Router A.
Verify that EIGRP is enabled and the proper network address is specified.
16. Use the ping command to verify connection to the following interfaces:
10.10.5.250
10.10.150.1
10.10.200.2
10.10.100.2
3-4 ADVANCED ROUTING REDISTRIBUTION

Routing redistribution is another important practice that a network engineer may
have to address when working in a big network environment. Routing redistribu- Routing
Redistribution
tion is the technique of injecting routes from one routing protocol into another rout-
ing protocol. In general, routes are only exchanged automatically with other routers The technique of
injecting routes from
that are running the same routing protocol (for example, OSPF-OSPF or EIGRP-
one routing protocol
EIGRP). Each routing protocol speaks its own language and does not automatically into another routing
communicate with other routing protocols. Additionally, each routing protocol has protocol.
its own metric or cost associated with it. For example, static routes use adminis-
trative distance, OSPF uses cost, IS-IS uses metric, and EIGRP uses a composite
metric. Therefore, routes from different routing protocols cannot be exchanged
automatically. As will be demonstrated, this process of exchanging routes has to be
configured manually.
There are several reasons why routing redistribution is necessary:
• One common reason is to redistribute static routes: These routes are not
learned, but are configured manually. They must be redistributed into a dy-
namic routing protocol so that they can be advertised.
• Another reason is when two or more dynamic routing protocols are used
on the network: This scenario happens often when connecting one network to
another either for peering purposes or for when companies merge their network
3-4: ADVANCED ROUTING REDISTRIBUTION 129

infrastructures due to a business acquisition. It is much easier to accept the new
routing protocol from another network and distribute the new routes into the
existing routing protocol rather than converting both routers to single routing
protocol. This saves time and money.
• Another reason is that certain routing protocols may not be supported
by certain network devices: A prime example of this is the EIGRP, which is
a Cisco proprietary protocol and is not supported by other network vendors.
If a non-Cisco router connects to a Cisco router that is running EIGRP, then
the Cisco router will need to redistribute routes learned from the other routing
protocol into EIGRP.
This section explores different routing redistribution techniques to redistribute

routes into dynamic routing protocols, such as RIP, OSPF, IS-IS, and EIGRP. Static
routing is not a dynamic routing protocol and is not applicable in this case, because
static routes must be configured manually. The configuration for this section is
based on the network diagram of Figure 3-5. Router A and Router B make up an
existing network and both routers are connected and are running the same routing
protocol. Router C brings in a new network with a different routing protocol that
will need to be redistributed into the existing network. Router B is where the redis-
tribution happens since this is the router that is directly connected to Router C.
RouterA RouterB
f0/1 f0/1
f0/0 f0/0 f0/1
19 istri
red
2.1 bu
68 ted
.30
f0/1 .0
RouterC
f0/0
LAN_A LAN_B LAN_C
10.10.10.0 Net 10.20.20.0 Net 192.168.30.0 Net
RouterA RouterB RouterC
Fa0/1: 10.1.1.1/30 Fa0/0: 10.1.1.2/30 Fa0/1: 172.16.20.2/30
Fa0/0: 10.10.10.1/24 Fa0/0: 10.20.20.1/24 Fa0/0: 192.168.30.1/24
Fa0/1: 172.16.20.1/30
FIGURE 3-5 Route redistribution from LAN C network to LAN A and LAN B
Route Redistribution into RIP

This section examines route redistribution into RIP. First, let’s explore the basic
RIPv2 configuration required for Router A and Router B. Both routers will have the
same routing configuration as shown and will be running a 10.0.0.0 classful network:
!
router rip
version 2
network 10.0.0.0
!

The output from the show ip route command displays the expected network routes
are being learned via RIPv2. The following is the result from Router A when the
command sh ip route rip is issued. This result shows that Router A has a RIP route
to the 10.20.20.0/24 network via the 10.1.1.2 interface on Router B:
RouterA#sh ip route rip
10.0.0.0/8 is variably subnetted, 3 subnets, 2 masks
R 10.20.20.0/24 [120/1] via 10.1.1.2, 00:00:02, FastEthernet0/1
However, Router B has an interface that is connected to Router C that is not part
of the network advertised via RIP. This interconnected interface on Router B is
FastEthernet0/1 and has an IP address of 172.16.20.1. This interface is not expected
to participate in RIP advertisement since Router C is running a different routing
protocol. This is why this route it is not included as a network statement in the RIP
configuration. Nonetheless, Router A should still learn about this interconnected
network, so Router A can reach it over the network.
In order to accomplish this, Router B can redistribute its connected network into
redistribute protocol
RIP. The command to do this is redistribute protocol. The following is a list of the
The command used to
options of what can be used to redistribute the networks into RIP:
inject routes from one
RouterB(config)#router rip routing protocol into
RouterB(config-router)#redistribute ? another routing protocol.
bgp Border Gateway Protocol (BGP)
connected Connected
eigrp Enhanced Interior Gateway Routing Protocol (EIGRP)
isis I SO IS-IS
iso-igrp IGRP for OSI networks
metric Metric for redistributed routes
mobile Mobile routes
odr On Demand stub Routes
ospf Open Shortest Path First (OSPF)
rip Routing Information Protocol (RIP)
route-map Route map reference
static Static routes
<CR>
RouterB(config-router)#redistribute connected
Therefore, the command redistribute connected can be issued to redistribute the

redistribute
connected networks, which, in this case, is the interface between Router B and connected
Router C.
The command issued
Once the command is issued, you can verify the results on Router A. The show ip to redistribute the
route rip command now shows that the network 172.16.20.0 is being advertised by connected networks.
RIP:
R 172.16.20.0 [120/1] via 10.1.1.2, 00:00:08, FastEthernet0/1

Now, assume there is a static route on Router B to the LAN C network of
redistribute static 192.168.30.0. This static route is only known by Router B; therefore, it will need
The command issued to to be redistributed into RIP, so that Router A can learn of it as well. Similar to the
redistribute the static redistribute connected command, you can use the redistribute static command
networks.
on Router B:
RouterB(config)#ip route 192.168.30.0 255.255.255.0 172.16.20.2
RouterB(config)#router rip
RouterB(config-router)#redistribute static
As a result, the redistributed static route to the 192.168.30.0 network entered in

Router B, is now known to Router A and as shown, Router A now learns of the
route to the LAN C network via RIP:
The other dynamic protocols are distributed into RIP in similar fashion. If Router C
runs OSPF as its routing protocol, then Router B will need to run the same in order
for the routers to communicate. To do this, both routers will have to agree on the
OSPF Process ID. In the following example, the OSPF Process ID will be 200. The
OSPF configuration for Router C and Router B are shown.
Router C’s OSPF configuration:
!
router ospf 200
log-adjacency-changes
network 192.168.30.0 0.255.255.255 area 0
network 172.16.20.0 0.0.0.3 area 0
!
Router B’s OSPF configuration:

!
router ospf 200
log-adjacency-changes
network 10.0.0.0 0.255.255.255 area 0
network 172.16.20.0 0.0.0.3 area 0
!
Now, Router B can receive an OSPF route from Router C as shown as the result of
issuing the command show ip route. Along with it, Router B receives a RIP route
from Router A. It is important to verify the routes of the protocol one is trying to
redistribute. Note: This important step is often neglected. Remember, the router
cannot redistribute what it doesn’t know and the route must be there for you to re-
distribute.

RouterB#sh ip route
level-2
static route
O 192.168.30.0/24 [110/2] via 172.16.20.2, 00:00:09,

FastEthernet0/1
C 10.20.20.0/24 is directly connected, FastEthernet0/0
Router A does not know of the new OSPF route yet, so the route has to be redistrib-
uted. On Cisco routers, there is a caveat for redistributing dynamic routing proto- redistribute ospf
process_id metric
cols into RIP. That caveat is the metric must be assigned in order for the routes to
0-16
be redistributed properly. This is not required for static and connected routes since
The command to
RIP assigns the metric of 1 to each of these by default. Also, recall that the metric
redistribute OSPF routes
value for RIP is hop count and each hop count along the path has the value of 1. into RIP.
The maximum hop count value is 16. The command to redistribute OSPF routes
into RIP is redistribute ospf process_id metric 0-16.
RouterB(config-router)#redistribute ospf 200 metric 1
Now, the RIP routing table on Router A shows the route 192.168.30.0 that is
learned from the LAN C network:
10.0.0.0/8 is variably subnetted, 3 subnets, 2 masks redistribute eigrp
R 10.20.20.0/24 [120/1] via 10.1.1.2, 00:00:08, FastEthernet0/1 AS_number metric
0-16
For EIGRP, the command to redistribute routes into RIP is almost the same as Command to redistribute
EIGRP routes into RIP.
OSPF. The command is redistribute eigrp AS_number metric 0-16. For IS-IS, the
command is redistribute isis IS-IS_Level metric 0-16. The configuration example redistribute isis
is as follows: IS-IS_Level metric
0-16
RouterB(config-router)#redistribute eigrp 200 metric 1
Command to redistribute
IS-IS routes into RIP.
RouterB(config-router)#redistribute isis level-1-2 metric 1

Route Redistribution into OSPF
Next, the steps for route redistribution for OSPF are examined for Router A and
Router B in the network shown in Figure 3-5. Both routers have been configured to
run OSPF using the following commands:
!
router ospf 200
network 10.0.0.0 0.255.255.255 area 0
!
Once the basic configuration for OSPF has been configured, the routers will have
the OSPF neighbor adjacency established and the OSPF routes will be exchanged.
The result of the show ip ospf neighbor and show ip route ospf on Router A are
shown in the output examples that follow. The results confirm the neighbor adja-
cency and that the expected network routes are being learned via OSPF:
RouterA#sh ip ospf neighbor
Neighbor ID Pri State Dead Time Address

Interface
172.16.20.1 1 FULL/BDR 00:00:34 10.1.1.2
FastEthernet0/1
O 10.20.20.0/24 [110/2] via 10.1.1.2, 00:15:38, FastEthernet0/1
Router B’s connected interface to Router C is not advertised as a network statement

in the OSPF configuration. Therefore, its connected network will not be shown in
Router A’s routing table. This is the same behavior demonstrated for RIP. Just like
RIP, connected networks can be redistributed in OSPF by using the redistribute
connected command. This command is next issued on Router B:
% Only classful networks will be redistributed
However, what we get is the message stating that only classful networks can be
redistributed. Because the connected network is 172.16.20.0/30, which is not a
classful network, this will not work. Cisco provides a solution to this problem with
a slightly different command:
RouterB(config-router)#redistribute connected subnets
redistribute
This command allows the classless network to be distributed. Once the command
connected subnets
is issued, we can then verify the result on Router A. The show ip route command
Allows the classless
now shows the network 172.16.20.0 is being advertised by OSPF. Not only that, the
network to be
distributed.
network route is flagged as an E2 or an OSPF external type 2 route, which is differ-
ent than other OSPF internal routes. By default, any routes that are being redistrib-
E2 uted from another routing protocol into OSPF will appear as OSPF E2 routes that
An OSPF external type 2 are the least preferred among the OSPF route types.
route.
RouterA#sh ip route

level-2
static route

O E2 172.16.20.0 [110/20] via 10.1.1.2, 00:01:59, FastEthernet0/1
Now, configure a static route on Router B to the LAN C network of 192.168.30.0

and redistribute it into OSPF. This enables Router A to learn the route to
192.168.30.0. To redistribute static routes, issue the command redistribute static
on Router B:
% Only classful networks will be redistributed
The same warning message is shown saying that, “only classful networks will be
redistributed.” This time, the network 192.168.30.0/24 is a classful network, so the redistribute static
subnets
network will be redistributed; otherwise, the command redistribute static subnets
Allows the static network
will need to be used to distribute classless networks similar to what was done for
to be distributed.
the connected network. As a result of the redistributed static route entered in Router
B, Router A now learns of the route to LAN C network via OSPF and the network
is shown as an OSPF external type 2 route:
O E2 192.168.30.0/24 [110/20] via 10.1.1.2, 00:03:37, FastEthernet0/1
The other dynamic protocols are distributed into OSPF in similar fashion. For ex-
ample, if Router C runs EIGRP as its routing protocol, then Router B will need to
run EIGRP in order for Router B and Router C to communicate. Both routers have
to agree on an EIGRP Autonomous system number. In this example, an EIGRP
AS number of 200 is selected. The EIGRP routing configuration for Router C and
Router B follow.

Router C’s EIGRP configuration:
!
router eigrp 200
network 192.168.30.0 0.
network 172.16.0.0
no auto-summary
!
Router B’s EIGRP configuration:

!
router eigrp 200
network 10.0.0.0
network 172.16.0.0
no auto-summary
!
Now, verify that Router B is able to receive an EIGRP route from Router C before
proceeding with the route redistribution. This can be done using the sh ip route
command. The routing table for Router B is shown. It shows a network route of
192.168.30.0/24 learned from Router C’s interface 172.16.20.2 via EIGRP:
RouterB#sh ip route
level-2
static route
D 192.168.30.0/24 [90/30720] via 172.16.20.2, 00:50:45,

FastEthernet0/1
The next step is to configure Router B to redistribute a learned EIGRP route into
OSPF. Recall that the OSPF cost or metric is derived from the bandwidth of the
interface. By default, Cisco routers will assign an OSPF cost/metric of 20 to any
redistributed routes with exception for redistributed routes from BGP. If the metric
does not need to be set, the command to distribute dynamic routing protocols is
all the same for OSPF. The command to redistribute EIGRP routes into OSPF is

redistribute eigrp AS_id [metric 0-16777214]. This time, the keyword subnets
will be used to avoid the warning message of only classful networks will be redis- redistribute eigrp
AS_id [metric
metric
tributed:
0-16777214]
RouterB(config)#router ospf 200 The command to
RouterB(config-router)#redistribute eigrp 200 subnets redistribute EIGRP routes
into RIP.
The command show ip route ospf on Router A would confirm the result that the
network 192.168.30.0/24 is advertised as an OSPF external type 2 route:
O E2 192.168.30.0/24 [110/20] via 10.1.1.2, 00:03:37, FastEthernet0/1
To redistribute RIP into OSPF, the command is redistribute rip

rip. For IS-IS, the
command is redistribute isis IS-IS_Level. The configuration example is as follows: redistribute rip
The command to
redistribute RIP into
RouterB(config-router)#redistribute rip OSPF.
RouterB(config-router)#redistribute isis level-1-2
redistribute isis
IS-IS_Level
The command to
Route Redistribution into EIGRP redistribute EIGRP routes
Next, the steps for route redistribution of routers into EIGRP is examined. First, into IS-IS.
EIGRP is configured for Router A and Router B. Both routers will have the same
EIGRP routing configuration, as shown:
!
router eigrp 200
network 10.0.0.0
no auto-summary
!
The EIGRP neighbor adjacency can be verified with the command show ip eigrp
neighbor, and the EIGRP route is verified with show ip route eigrp on Router A.
The results confirm that the neighbor adjacency is established and the correct net-
work routes are being exchanged via EIGRP:
RouterA#sh ip eigrp neighbors
IP-EIGRP neighbors for process 200
H Address Interface Hold Uptime SRTT RTO Q Seq
(sec) (ms) Cnt Num
0 10.1.1.2 Fa0/1 11 00:03:39 40 240 0 4
RouterA#show ip route eigrp

D 10.20.20.0/24 [90/30720] via 10.1.1.2, 00:03:48,
FastEthernet0/1

Router B’s connected interface to Router C will need to be redistributed, because it
is not included as a network statement in the EIGRP configuration. Router A’s rout-
ing table does not show a connected network for 172.16.20.0/30. This is expected,
because it has not yet been redistributed by Router B, as previously discussed in the
RIP and OSPF example. The command redistribute connected is issued on Router
B to redistribute the connected network:
Once the command redistribute connected is issued, the command sh ip route can
EX be issued on Router A to verify that the network 172.16.20.0 is being advertised via
External EIGRP type. EIGRP. The network route is also flagged as an EX or an external EIGRP type. The
external EIGRP route has an administrative distance (AD) value of 170, which is
higher than the internal EIGRP route’s AD value of 90:
RouterA#sh ip route
level-2
static route

D EX 172.16.20.0 [170/30720] via 10.1.1.2, 00:23:14,
FastEthernet0/1
D 10.20.20.0/24 [90/30720] via 10.1.1.2, 00:37:20,
FastEthernet0/1
Next, a static route on Router B needs to be configured to the LAN C network of

192.168.30.0. Additionally, the static route must be redistributed into EIGRP. This
enables Router A to learn the route as well. To redistribute static routes, the com-
mand redistribute static is issued on Router B.
As a result, Router A now learns of the route to LAN C network via EIGRP, and the
network 192.168.30.0 is displayed as an EIGRP external route, just like the distrib-
uted connected route:

D EX 192.168.30.0/24 [170/30720] via 10.1.1.2, 00:00:17,
FastEthernet0/1
D EX 172.16.20.0 [170/30720] via 10.1.1.2, 00:33:53,
FastEthernet0/1
D 10.20.20.0/24 [90/30720] via 10.1.1.2, 00:47:59,
FastEthernet0/1
As we have learned, EIGRP uses a composite metric, which consists of bandwidth,

delay, reliability, load, and MTU. To redistribute other dynamic routing protocols,
the metric value has to be converted into EIGRP’s metric values or new EIGRP
metric values must be assigned. Routes will not be redistributed properly without
these values. Cisco recommends that every redistributed dynamic routing protocol
be assigned with EIGRP metrics. As it turns out, even though only the two EIGRP
metric values of bandwidth and delay are used by default, Cisco enforces all the
metric values be defined. The parameters defining the EIGRP metric assignments
are as follows:
• Bandwidth metric is a value between 1–4294967295 in Kbps.
• Delay metric is a value between 0–4294967295 in ten microseconds.
• Reliability metric is a value between 1–255, where 255 is 100 percent reliable.
• Load metric is a value between 0–255, where 255 is a 100 percent load.
• MTU is a maximum transfer unit value between 1-65,535, where 1,500 is a
norm.
The following example demonstrates how to redistribute RIPv2 routes into EIGRP.
Router C and Router B are both configured to run RIPv2. The RIPv2 routing con-
figurations for Router C and Router B follow:
Router C’s RIPv2 configuration:
!
router rip
version 2
network 172.16.0.0
network 192.168.30.0
!
Router B’s RIPv2 configuration:

!
router rip
version 2
network 10.0.0.0
network 172.16.0.0
!
Now, you must verify that Router B is able to receive a RIP route from Router C
before proceeding with the route redistribution. The routing table of Router B is

shown in the output that follows that shows a network route of 192.168.30.0/24,
learned from Router C’s 172.16.20.2 interface via RIP.
RouterB#sh ip route
level-2
static route
R 192.168.30.0/24 [120/1] via 172.16.20.2, 00:00:20,

FastEthernet0/1
D 10.10.10.0/24 [90/30720] via 10.1.1.1, 00:48:39,
FastEthernet1/0
The next step is to configure Router B to redistribute a learned RIP route into
redistribute rip
EIGRP. As previously discussed, all the EIGRP metric values will need to be as-
metric bandwidth
delay reliability load signed to the redistributed routing protocol. The command to redistribute RIP
MTU routes into EIGRP is redistribute rip metric bandwidth delay reliability load MTU,
The command to
as demonstrated here:
redistribute RIP routes RouterB(config)#router eigrp 200
into EIGRP. RouterB(config-router)#redistribute rip metric 100000 10 255 1 1500
Instead of entering the long metric command every time, a default metric can be
default-metric defined. For every redistribute statement, it will use the default metric. If the redis-
bandwidth delay tribute statement has metric values assigned, it overrides the default metric. The
reliability load MTU
command for setting the default metric is default-metric bandwidth delay reliabil-
The command for setting
ity load MTU. The following is an example of how a default metric is configured:
the default metric.
RouterB(config-router)#default-metric 100000 10 255 1 1500
RouterB(config-router)#redistribute rip
This command sets the default metric to the following:

Bandwidth: 100000 kbps
Delay: 1000 microseconds
Reliability: 100 percent reliable
Load: minimal
MTU: norm

The command show ip route eigrp on Router A confirms the result that the net-
work 192.168.30.0/24 is advertised as an external EIGRP route:
D EX 192.168.30.0/24 [170/30720] via 10.1.1.2, 00:00:02,
FastEthernet0/1
D EX 172.16.20.0 [170/30720] via 10.1.1.2, 01:18:33,
FastEthernet0/1
D 10.20.20.0/24 [90/30720] via 10.1.1.2, 01:32:39,
FastEthernet0/1
Once the default metric is defined in EIGRP, the command redistribute ospf can
be used to redistribute OSPF into EIGRP. For IS-IS, the command is redistribute
isis IS-IS_Level. The configuration example is as follows:
RouterB(config-router)#redistribute rip
RouterB(config-router)#redistribute isis level-1-2
Route Redistribution into IS-IS

Surprisingly, the configuration for route redistribution into IS-IS is simple to apply.
First, let’s configure Router A and Router B with IS-IS routing and then enable the
interfaces that will participate in IS-IS routing. The running configuration files for
Router A and Router B are next shown for IS-IS.
Router A’s configuration
ip address 10.10.10.1 255.255.255.0
ip router isis
!
ip address 10.1.1.1 255.255.255.252
ip router isis
!
router isis
net 49.0001.c202.00e8.0001.00
!
Router B’s configuration:

ip address 10.20.20.1 255.255.255.0
ip router isis
!
ip address 10.1.1.2 255.255.255.252
ip router isis
!

router isis
net 49.0001.c202.00e8.0002.00
!
The IS-IS routing is verified with show ip route isis on Router A. The result con-
firms the correct network routes are being exchanged via IS-IS:
i L1 10.20.20.0/24 [115/20] via 10.1.1.2, FastEthernet0/1
One caveat to know about redistributing routes into IS-IS is that if no metric is as-
signed, all distributed routes will be automatically assigned a metric value of 0.
In IS-IS, the default metric value for each interface is 10. Therefore, redistributed
routes could end up with lower metric values than IS-IS learned routes. If there
happens to be multiple routes to the same network, one is learned via internal IS-IS
routes and another one is learned via redistributed routes. IS-IS would prefer the
redistributed routes. It is recommended that at least a metric value of 10 be assigned
to redistributed routes. To advertise a connected network of 172.16.20.0/30, the
command redistribute connected is issued on Router B with a metric of 10:
RouterB(config)#router isis
RouterB(config-router)#redistribute connected metric 10
Once the command is issued, you can then verify the results on Router A. The show
ip route command now shows the network 172.16.20.0 is being advertised via IS-
IS and the network route is also flagged as an L2 or an IS-IS level-2 type:
RouterA#sh ip route
level-2
static route

Now, configure a static route on Router B to the LAN C network of 192.168.30.0

and redistribute it into IS-IS. This enables Router A to learn the route as well. To
redistribute static routes into IS-IS, the command redistribute static is issued on
Router B.

RouterB(config-router)#redistribute static metric 10
As a result, Router A now learns of the route to LAN C network via IS-IS and the
network is displayed as an IS-IS level-2 route just like the distributed connected
route:
To redistribute dynamic routing protocols into IS-IS, the redistribute command is

configured in the same fashion. The following configurations show how to redis-
tribute RIP, OSPF, and EIGRP into IS-IS respectively:
RouterB(config-router)#redistribute rip metric 10
RouterB(config-router)#redistribute ospf 200 metric 10
RouterB(config-router)#redistribute eigrp 200 metric 10
As long as those dynamic routing protocols are configured on Router B and Router
C correctly and their routes are exchanged, the result of show ip route isis on
Router A should yield the same result as the one from the static route.
3-5 ANALYZING OSPF “HELLO” PACKETS Hello Packets

Periodically sent in
Hello packets are periodically sent in OSPF networks to initiate and maintain com- link state protocols to
munications with neighbor routers. The Hello packets contain parameters including initiate and maintain
specifications for the following: communications with
neighbor routers.
• Network Mask: The mask contains 32 bits in four octets (8-bit groups). The
masking is a logical AND operation, and the bits that are set high allow the ad- Network Mask
dress data to pass. For example, a network mask of 255.255.255.0 has the first A 32-bit value used to
24 bits set high in the first three octets. The hexadecimal equivalent (base 16) divide sections of IP
for this network mask is 0xffffff00. addresses.
• Hello Interval: The time between Hello packets. Hello Interval

The time between Hello
• Router Dead Interval: The length of time a router neighbor is quiet (no Hello
packets.
packets) before assuming the neighbor is dead.
Router Dead Interval
OSPF uses a Class D multicast IP address to send out the Hello packets to the The length of time a
neighbors. OSPF networks have been assigned the multicast IP address of 224.0.0.5 router neighbor is quiet
(no Hello packets) before
for sending Hello packets. Multicast IP addresses for use with multicast protocols
assuming the neighbor
are in the range of 224.0.0.0 to 239.255.255.255. This is called the Class D IP ad-
is dead.
dress range. The IP address range above Class D is called Class E, and addresses
3-5: ANALYZING OSPF “HELLO” PACKETS 143

range from 240.0.0.0 to 254.255.255.255. Class E is called the IP address experi-
mental range.
The test network shown in Figure 3-6 was set up so that the transmission of OSPF
RID Hello packets could be viewed. The Wireshark protocol analyzer was installed on
Router ID. computer D1 in LAN D (192.168.1.0 NET). LAN D has been configured to be an
OSPF advertised route; therefore, computers in LAN D will receive the OSPF mul-
ticasts used to capture the data packets. OSPF broadcasts the LAN address as part
of the routed addresses. The multicast is sent to all neighbor OSPF router connec-
tions. Figure 3-7 shows the captured data packets for the network. The highlighted
line, beginning with the ID of “000001” in the first column, is the first detected
occurrence of a data packet on the network. The elapsed time is 5.278.530.840
seconds (column 2). The size of the packet is 82 bytes (column 3). The destination
(specified in column 4) is an OSPFIGP_Router multicast. The source IP address is
192.168.1.250, which is the IP address for the router interface. The packet summary
(column 6) indicates this is an OSPF Hello from RID (router ID) 192.168.200.11.
192.168.200.0 NET
E1 (192.168.200.11)
RouterC
E0 (192.168.1.250)
192.168.1.0 NET
LAN D
Hub
D1 D2
Protocol Analyzer
FIGURE 3-6 The test network for capturing the OSPF Hello packets
One of the complex things about OSPF is the assignment of the router ID (RID).
The router ID is an IP address chosen from all interfaces on the router. Cisco IOS
first examines the loopback address for the router’s ID IP address. If a loopback
address is not being used, the highest IP address for a router interface is selected
as the router ID. In this case, the highest loopback IP address is 192.168.200.11,
the IP address for the E1 Ethernet interface. The middle panel in Figure 3-7 shows

that the destination address is 224.0.0.5 and defines this as OSPFIGP_Router. OS-
OSPFIGP
PFIGP stands for Open Shortest Path First Interior Gateway Protocol, which is a
concatenated form of OSPF and IGP. The OSPF protocol is classified as an Interior Open Shortest Path
First Interior Gateway
Gateway Protocol (IGP). IGP represents entities under the same autonomous do-
Protocol.
main (administrative and security policies). Remember, when you specify the OSPF
protocol, a process ID number is entered; for example, router ospf 100. IGP
Interior Gateway
Protocol.
FIGURE 3-7 The captured OSPF multicasts
The bottom panel in Figure 3-7 is the actual data packet displayed in hexadecimal
(base 16) code. The highlighted hex code is
E0000005
This is the hexadecimal value for 224.0.0.5, the destination IP address of the multi-
cast.
Hexadecimal Value: E 0 00 00 05
Decimal Equivalent: 224 0 0 5
Note
The decimal equivalent of E0 is 14, and (14 ✕ 16 + 0) = 224.
A detailed view of the OSPF packet information is provided in Figure 3-8. This in-
formation is available in the Open Shortest Path First (OSPF) window, as shown in
Figure 3-8.
3-5: ANALYZING OSPF “HELLO” PACKETS 145

FIGURE 3-8 A detailed view of the OSPF packet information
In addition to the version number, the router ID is listed (192.168.200.11), and the
Hello interval is specified to be 10 seconds. Looking again at Figure 3-7, the OSPF
Hello packets are approximately 10 seconds apart. The router dead interval is 40
seconds.
This section has demonstrated how to use the Surveyor Demo Protocol Analyzer to
capture OSPF “Hello” packets. Figures 3-7 and 3-8 and the accompanying text ex-
plained how to extract information from the captured packets.

SUMMARY
This chapter presented examples of configuring routing protocols. The network
challenge exercises provided the opportunity for the student to test her or his con-
figuration skill prior to actually configuring a real router. The student should be
able to configure and verify operation of the following protocols:
Static
RIP/RIPv2
OSPF
ISIS
EIGRP
Additionally, this chapter examined the steps for route redistribution. The last sec-
tion examined the OSPF Hello packets.
QUESTIONS AND PROBLEMS

Section 3-1
1. OSPF is (select all that apply)
a. Open Shortest Path First routing protocol
b. An open protocol
c. Developed specifically for TCP/IP networks
d. Developed specifically for IPX networks
e. A distance vector protocol
f. A dynamic routing protocol
g. A link state protocol
h. A high consumer of bandwidth
2. In OSPF, route updates are sent in the form of
a. Link state advertisements
b. Exchanging routing tables every 30 seconds
c. Exchanging routing tables every 90 seconds
d. IETF packets
3. The OSPF routing protocol uses these to verify that a link between two routers
is active and the routers are communicating
a. LSAs
b. Hello packets
c. ARP messages
d. Ping
QUESTIONS AND PROBLEMS 147

Bab 3

Uploaded by

Copyright:

Available Formats

Bab 3

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Bab 3

Uploaded by

Copyright:

Available Formats

INTRODUCTION

This chapter examines advanced routing protocols. Configuring the

3-1 CONFIGURING LINK STATE PROTOCOLS—OSPF

3-1: CONFIGURING LINK STATE PROTOCOLS—OSPF 101

102 CHAPTER 3: ADVANCED ROUTER CONFIGURATION II

TABLE 3-1 Summary of Advantages and Disadvantages of OSPF

Configuring Routes with OSPF

3-1: CONFIGURING LINK STATE PROTOCOLS—OSPF 103

RouterA E2 RouterB RouterC E2

10.10.20.0 NET 10.10.10.0 NET 10.10.1.0 NET 10.10.5.0 NET

Hub Switch Switch Hub

Ethernet 0 10.10.20.250 Ethernet 0 10.10.10.250 Ethernet 1 10.10.1.250 Ethernet 0 10.10.5.250

LAN A Computers LAN B Computers LAN C Computers LAN D Computers

Computer A1 10.10.20.1 Computer B1 10.10.10.1 Computer C1 10.10.1.1 Computer D1 10.10.5.1

FIGURE 3-2 The three-router campus LAN

A subnet mask of 255.255.255.0 is assigned to each of the FastEthernet interfaces.

104 CHAPTER 3: ADVANCED ROUTER CONFIGURATION II

3-1: CONFIGURING LINK STATE PROTOCOLS—OSPF 105

106 CHAPTER 3: ADVANCED ROUTER CONFIGURATION II

3-1: CONFIGURING LINK STATE PROTOCOLS—OSPF 107

108 CHAPTER 3: ADVANCED ROUTER CONFIGURATION II

Load Balancing and Redundancy with OSPF

3-1: CONFIGURING LINK STATE PROTOCOLS—OSPF 109

110 CHAPTER 3: ADVANCED ROUTER CONFIGURATION II

3-1: CONFIGURING LINK STATE PROTOCOLS—OSPF 111

L1 Routers 3-2 CONFIGURING LINK STATE PROTOCOLS—IS-IS

112 CHAPTER 3: ADVANCED ROUTER CONFIGURATION II

Configuring Routes with IS-IS

RouterA E2 RouterB RouterC E2

10.10.20.0 NET 10.10.10.0 NET 10.10.1.0 NET 10.10.5.0 NET

Hub Switch Switch Hub

Ethernet 0 10.10.20.250 Ethernet 0 10.10.10.250 Ethernet 1 10.10.1.250 Ethernet 0 10.10.5.250

LAN A Computers LAN B Computers LAN C Computers LAN D Computers

Computer A1 10.10.20.1 Computer B1 10.10.10.1 Computer C1 10.10.1.1 Computer D1 10.10.5.1

FIGURE 3-3 Routers A, B, and C in the three-router campus network

3-2: CONFIGURING LINK STATE PROTOCOLS—IS-IS 113

114 CHAPTER 3: ADVANCED ROUTER CONFIGURATION II

Gateway of last resort is not set

10.0.0.0/24 is subnetted, 7 subnets

3-2: CONFIGURING LINK STATE PROTOCOLS—IS-IS 115

116 CHAPTER 3: ADVANCED ROUTER CONFIGURATION II

Load Balancing and Redundancy with IS-IS

3-2: CONFIGURING LINK STATE PROTOCOLS—IS-IS 117

Networking Challenge: IS-IS

118 CHAPTER 3: ADVANCED ROUTER CONFIGURATION II

3-3 CONFIGURING HYBRID ROUTING

3-3: CONFIGURING HYBRID ROUTING PROTOCOLS—EIGRP 119

Configuring Routes with EIGRP

10.10.20.0 NET 10.10.10.0 NET 10.10.1.0 NET 10.10.5.0 NET

Hub Switch Switch Hub

Ethernet 0 10.10.20.250 Ethernet 0 10.10.10.250 Ethernet 1 10.10.1.250 Ethernet 0 10.10.5.250

LAN A Computers LAN B Computers LAN C Computers LAN D Computers

Computer A1 10.10.20.1 Computer B1 10.10.10.1 Computer C1 10.10.1.1 Computer D1 10.10.5.1

FIGURE 3-4 The three-router campus LAN

120 CHAPTER 3: ADVANCED ROUTER CONFIGURATION II

TABLE 3-2 Router Interface IP Addresses and Subnet Masks

Interface IP Address Subnet Mask

3-3: CONFIGURING HYBRID ROUTING PROTOCOLS—EIGRP 121

122 CHAPTER 3: ADVANCED ROUTER CONFIGURATION II