Routing and Switching

James Jefferson S. De Vera

 Routing and Switching:
Dynamic Routing (OSPF &
EIGRP)
James Jefferson S. De Vera
Configure the OSPF routing protocol.
Configure the EIGRP routing protocol.

James Jefferson S. De Vera

 Open Shortest Path First (OSPF)

•Link-State Protocol: Maintains a complete map of the

network's topology.
•Areas: Divides large networks into smaller areas for
easier management and faster convergence.
•Metric: Cost is the primary metric used by OSPF to
calculate the best path.
•Open Standard: OSPF is an open standard protocol,
making it widely adopted in the industry.
 What is OSPF?
Link-State Protocol: OSPF is a link-
state routing protocol, meaning it
maintains a complete and detailed
map of the network's topology.
Each router has a database that
represents the entire network's
structure, including the status of
links and the cost associated with
each link. This allows for efficient
and intelligent routing decisions.
Example

Imagine each router in the network as a person holding a

map of the entire city. When someone wants to go from
point A to point B, each person can consult their map and
easily find the most efficient route based on the current
conditions.
 Areas
OSPF enables the division of large
networks into smaller, manageable
areas. This hierarchical structure
simplifies routing calculations and
reduces the amount of routing
information that needs to be
exchanged between routers, leading
to faster convergence and improved
network stability.
Example

A university campus network might be divided into

areas based on different buildings or departments.
Each area would have its own set of routers and
links, making it easier to manage and
troubleshoot.
 Metric: Cost
OSPF uses cost as its primary metric to
determine the best path to a destination.
The cost of a link is typically inversely
proportional to its bandwidth, so faster
links have lower costs. Routers calculate
the cumulative cost of all links along a
path to determine the total cost of that
path. The path with the lowest total cost
is considered the best.
Example

If there are two paths from Router A to Router B,

one with a cost of 10 and the other with a cost of
20, OSPF will choose the path with the lower cost
of 10.
 Key Points
•OSPF is a robust and scalable dynamic routing protocol ideal for
large, complex networks.
•It maintains a complete map of the network topology for efficient
routing decisions.
•The use of areas helps to simplify management and improve
convergence.
•The cost metric ensures that OSPF chooses the most efficient
paths based on available bandwidth.
•Its open standard nature makes it a widely adopted and versatile
protocol in the networking industry.
 Configuring OSPF 1
OSPF basic configuration is very simple. Like with other routing protocols
covered so far (RIP, EIGRP) first you need to enable OSPF on a router. This
is done by using the router ospf PROCESS-ID global configuration
command. Next, you need to define on which interfaces OSPF will run and
what networks will be advertised. This is done by using the network
IP_ADDRESS WILDCARD_MASK AREA_ID command from the ospf
configuration mode.

NOTE the OSPF process number doesn't have to be the same on all routers
in order to establish a neighbor relationship, but the Area ID has to be the
same on all neighboring routers in order for routers to become neighbors.
 Let's get started with the basic OSPF
configuration.

First, we need to enable OSPF on both routers. Then we need to define

what network will be advertised into OSPF.
 This can be done by using the following
sequence of commands on both routers:
The network commands entered on both routers include subnets
directly connected to both routers. We can verify that!! the routers
have become neighbors by typing the show ip ospf neighbors
command on either router:
To verify if the routing updated were exchanged, we can use the show
ip route command. All routes marked with the character "O" are OSPF
routes. For example, here is the output of the command on R1:

You can see that R1 has learned about the network 192.168.0.0/24 through OSPF.
 Configuring OSPF
Although basic OSPF configuration can be very simple, OSPF provides
many extra features that can get really complex. In this example, we
will configure multiarea OSPF network and some other OSPF features.
Consider the following multiarea OSPF network.
In this example we have two OSPF areas, area 0 and area 1. As you can see
from the network topology depicted above, Routers R1 and R3 are in the area
0 and area 1, respectively. Router 2 connects to both areas, which makes him
an ABR (Area Border Router). Our goal is to advertise the subnets directly
connected to R1 and R3. To do that, the following configuration on R1 will be
used:

NOTE we have used the router-id 1.1.1.1 command to manually specify the
router ID of this router. OSPF process will use that RID (router-id) when
communicating with other OSPF neighbors.
Because R1 connects only to R2, we only need to establish a neighbor
relationship with R2 and advertise directly connected subnet into OSPF.
Configuration of R3 looks similar, but with one difference, namely area
number. R3 is in the area 1.
What about R2? Well, because R2 is an ABR, we need to establish
neighbor relationship with both R1 and R3. To do that, we need to
specify different area ID for each neighbor relationship, O for R1 and 1
for R2. We can do that using the following sequence of commands:
Now R2 should have neighbor relationship
with both R1 and R3. We can verify that by
using the show ip ospf neighbor command:
To verify if directly connected subnets are really advertised into the
different area, we can use the show ip route ospf command on both R1
and R3:

Characters IA in front of the routes indicate that these routes reside in

different areas.
NOTE - since they reside in different areas, R1 and R3 will never establish a
neighbor relationship.
 Enhanced Interior Gateway Routing Protocol
(EIGRP)
•Hybrid Protocol: Combines features of both distance-vector and
link-state protocols.
•Diffusing Update Algorithm (DUAL): Ensures loop-free paths
and fast convergence.
•Composite Metric: Considers bandwidth, delay, reliability, and
load to calculate the best path.
•Cisco Proprietary: EIGRP was developed by Cisco but is now
an open standard
 Hybrid Protocol

•EIGRP is unique in that it combines

characteristics of both distance-vector and link-
state routing protocols.
•Distance-Vector: Like RIP, EIGRP exchanges
routing tables with neighbors. However, it goes
beyond simple hop counts.
•Link-State: Similar to OSPF, EIGRP maintains
a topology table, giving it better awareness of the
network's structure. This combination results in a
protocol that is both efficient and fast-converging.
 Diffusing Update Algorithm (DUAL)

•DUAL is EIGRP's secret sauce. This algorithm

ensures that EIGRP routes are loop-free,
meaning data packets won't get stuck going in
circles. It also allows for very fast convergence
when changes occur in the network topology.
•Example: If a link fails, EIGRP can quickly
recalculate alternate paths without creating
routing loops, minimizing downtime.
 Composite Metric
EIGRP uses a more sophisticated metric than RIP.
It considers multiple factors to calculate the best
path:

● Bandwidth: The capacity of the link.

● Delay: The time it takes for a packet to travel
across the link.
● Reliability: The likelihood of a link remaining
up and operational.
● Load: The current traffic level on the link.
Example

EIGRP might prefer a path with slightly higher

bandwidth and lower delay over a path with lower
bandwidth but no delay, as it aims to balance
speed and reliability.
 Cisco Proprietary (Now Open Standard)
EIGRP was initially developed by Cisco as a
proprietary protocol. However, in
response to demand and to promote
interoperability, Cisco has made EIGRP an
open standard.

Example: This means that other vendors

can now implement EIGRP in their
networking equipment, making it a more
versatile choice for diverse network
environments.
● EIGRP configuration closely resembles RIP configuration. Only two
steps are required:
○ enabling EIGRP by using the router eigrp

ASN_NUMBER command
○ telling EIGRP which networks to advertise by using one or more

network statements
● The first command, router eigrp ASN_NUMBER, enables EIGRP on a
router. ASN_NUMBER represents an autonomous system number
and has to be the same on all routers running EIGRP, otherwise
routers won't become neighbors. The second command, network
SUBNET, enables EIGRP on selected interfaces and specifies which
networks will be advertised. By default, the network command
takes a classful network number as the parameter.
To illustrate a configuration of EIGRP, we will use the
following figure:

The network depicted above consists of only two routers.

Each router has a directly connected subnet that needs to be advertised
through EIGRP.
The following figure show the EIGRP
configuration on R1 and R2:
You can verify that routers have become neighbors by using the show
ip eigrp neighbors command on either router:

The command above lists all EIGRP neighbors. The address field lists the neighboring
router RID (router ID). The interface field shows on which local interface the neighbor
relationship has been formed.
You can verify that routes are indeed being exchanged by
using the show ip route command on both routers:
NOTE - the "D" character at the beginning of a line in a routing table indicates
that the route has been learned via EIGRP.
 Configuring EIGRP

● By default, the network command uses a classful

network as the parameter.
● All interfaces inside that classful network will participate
in an EIGRP process.
● To enable EIGRP only on specific interfaces, a wildcard
mask can be used.
● The syntax of the command is:
(router-eigrp) network WILDCARD_MASK
 Configuring EIGRP

● By default, the network command uses a classful

network as the parameter.
● All interfaces inside that classful network will participate
in an EIGRP process.
● To enable EIGRP only on specific interfaces, a wildcard
mask can be used.
● The syntax of the command is:
(router-eigrp) network WILDCARD_MASK
 Consider the following example.

Router R1 has two directly connected subnets, 10.0.0.0/24 and 10.0.1.0/24.

We want to enable EIGRP only on the subnet connected to the interface Fa0/0.
If we enter the network 10.0.0.0 command under the EIGRP configuration
mode, both subnets will be included in EIGRP process because we've used a
classful network number in the network command.
 Consider the following example.

Router R1 has two directly connected subnets, 10.0.0.0/24 and 10.0.1.0/24.

We want to enable EIGRP only on the subnet connected to the interface Fa0/0.
If we enter the network 10.0.0.0 command under the EIGRP configuration
mode, both subnets will be included in EIGRP process because we've used a
classful network number in the network command.
To configure EIGRP only on interface Fa0/0, the network 10.0.0.0
0.0.0.255 command can be used. This will enable EIGRP only on
interfaces starting with 10.0.0.X.
By using the command show ip protocols, you can verify
that only network 10.0.0.0/24 is included in EIGRP: