Configuring OSPF over an NBMA Topology

10 Mar

Configuring OSPF over an NBMA Topology
The design considerations of running OSPF over an NBMA topology require thought and research because the configuration choices that are made will dramatically affect your network.

This section covers one of the common choices on a partially meshed network, illustrates the kind of decisions that need to be made, and shows the configuration that would be implemented.

If the network is partially meshed, then the choice to use only point-to-point subnets can waste addresses. If you choose a point-to-multipoint configuration, the network uses one subnet, and there is no DR/BDR negotiation. This configuration has the advantage of saving addresses but behaves as if it is a series of point-to-point links. If private addressing is used, this might not be a consideration.

When the decision is made as to which technology is to be implemented, whether you are using point-to-point or point-to-multipoint, the configuration is straightforward. The choice is defined on the interface as an ip ospf network command. The network command syntax is as follows:

Router(config-if)#ip ospf network {broadcast | non-broadcast | {point-to-multipoint
[non-broadcast]}}

Configuring OSPF over an NBMA Topology 233
Figure 7-1 Diagram for the Configuring OSPF Example

Configuring OSPF Example

234 Chapter 7: Configuring OSPF in a Single Area

Configuring OSPF in NBMA Mode
In NBMA mode, the design considerations are imperative, because the selection of the DR and BDR must ensure physical connectivity to all routers in the NBMA cloud. Remember that this is a nonbroadcast environment, so you must configure the DR and BDR with a static list of the other routers attached to the cloud so that they can become neighbors and create adjacencies. This is achieved with the use of the neighbor command.

NOTE New technologies introduced to OSPF have removed the need to configure the
neighbors manually. These technologies are explained later in the section “Configuring OSPF in
Point-to-Multipoint Mode.”

The syntax of the command is as follows:

Router(config-if)#neighbor ip-address [priority number ] [poll--intervall sec]
[cost number]

You use the neighbor command to configure OSPF routers interconnecting to nonbroadcast
networks. The different options used with the neighbor command are explained in Table 7-4.

Table 7-4 The OSPF neighbor Command

OSPF neighbor Command

Configuring OSPF over an NBMA Topology 235
Example 7-2 shows how the command is used.
Example 7-2 The OSPF neighbor Command

OSPF neighbor Command

NOTE NBMA mode is used by default in a nonbroadcast multiaccess environment, so there is no need for the ip ospf network non-broadcast command. However, neighbor statements are necessary.

Configuring OSPF in Point-to-Multipoint Mode An OSPF point-to-multipoint interface is seen as a numbered point-to-point interface with one or
more neighbors. The cloud is configured as one subnet with a host route for each router involved in the OSPF cloud.

NOTE The ip ospf network point-to-multipoint non-broadcast command option is a feature related to point-to-multipoint networks with Cisco IOS Software Release 11.3a. You can find more information on the subject by searching Cisco.com with the keywords “OSPF point-tomultipoint network with separate costs per neighbor” or click the link to the Technical Support OSPF page.

By default, the network is considered to be a series of point-to-point interfaces. There is no need to specify neighbors, because the neighbors will see each other and simply become adjacent, with no need for the election of a DR or a BDR. Point-to-multipoint does not try to reduce adjacencies using a DR. Instead, it accepts the extra overhead of having a full set of adjacencies for the sake of stability. Point-to-multipoint forms an adjacency automatically along any PVC, which causes more overhead but is more resilient than NBMA.

You can specify neighbors with the neighbor command, in which case you should specify a cost to each neighbor. You are not required to have a fully meshed topology, which reduces the number of PVCs needed and the number of neighbor entries in the neighbor table.

It is possible to change the default of NBMA to point-to-multipoint with the command ip ospf network point-to-multipoint and to nonbroadcast network with the command ip ospf network point-to-multipoint non-broadcast. The point-to-multipoint network is then considered a nonbroadcast network, and the mode is a Cisco extension. The neighbor command is required to identify neighbors in a nonbroadcast network. Assigning a cost to a neighbor is optional.

Example 7-3 shows the necessary configuration for OSPF in point-to-multipoint mode (point-tomultipoint broadcast mode, compliant with the RFC 2328, because the keyword non-broadcast is not specified). There is no need to configure neighbors, although you can do so if desired.

Example 7-3 Configuring Point-to-Multipoint Networks

Router(config)#interface Serial0
Router(config-if)#ip address 10.1.1.1 255..255..255.0
Router(config-if)#encapsulation frame-relay
Router(config-if)#ip ospf network point-to-multipoint
Router(config)#router ospf 1
Router(config-router)#network 10.1.1.0 0.0.0.255 area 0

Configuring OSPF in Broadcast Mode
You use the broadcast mode to avoid using the neighbor command and all the attendant configurations.

This broadcast mode works best with a fully meshed network. Example 7-4 shows a typical
configuration of OSPF in broadcast mode.

Example 7-4 Configuring a Broadcast Network

Router(config)#iinntteerrffaaccee SSeerriiaall00
Router(config-if)#iipp aaddddrreessss 1100..11..11..11 225555..225555..225555..00
Router(config-if)#eennccaappssuullaattiioonn ffrraammee–rreellaayy
Router(config-if)#iipp oossppff nneettwwoorrkk bbrrooaaddccaasstt
Router(config)#rroouutteerr oossppff 11
Router(config-router)#nneettwwoorrkk 1100..11..11..00 00..00..00..225555 aarreeaa 00

Configuring OSPF in Point-to-Point Mode on a Frame Relay Subinterface
In the point-to-point mode, the adjacency created between the routers is automatic because each subinterface behaves as a physical point-to-point network. Therefore, the communication is direct and automatic.

The following steps explain how to configure OSPF point-to-point mode on subinterfaces:

Step 1 Configure Frame Relay encapsulation on the interface.
Step 2 At the interface level, create a subinterface. It is recommended that you remove any network layer address assigned to the physical interface using the no ip address command. The Layer 3 address should be assigned to the subinterface.
Step 3 Configure the Layer 3 and Layer 2 (DLCI) addresses on the subinterface.
Step 4 Point-to-point mode is the default OSPF mode for point-to-point subinterfaces, so no further configuration is required.

Example 7-5 shows the necessary configuration required for a point-to-point Frame Relay subinterface.
Example 7-5 Configuring a Point-to-Point Frame Relay Subinterface

Router(config)#iinntteerrffaaccee SSeerriiaall00
Router(config-if)#nnoo iipp aaddddrreessss
Router(config-if)#eennccaappssuullaattiioonn ffrraammee–rreellaayy
Router(config)#iinntteerrffaaccee SSeerriiaall00..11 ppooiinntt–ttoo–ppooiinntt
Router(config-subif)#iipp aaddddrreessss 1100..11..11..11 225555..225555..225555..00
Router(config-subif)#ffrraammee–rreellaayy iinntteerrffaaccee–ddllccii 5511
Router(config)#iinntteerrffaaccee SSeerriiaall00..22 ppooiinntt–ttoo–ppooiinntt
Router(config-subif)#iipp aaddddrreessss 1100..11..22..11 225555..225555..225555..00
Router(config-subif)#ffrraammee–rreellaayy iinntteerrffaaccee–ddllccii 5522
Router(config)#rroouutteerr oossppff 11
Router(config-router)#nneettwwoorrkk 1100..11..00..00 00..00..225555..225555 aarreeaa 00

The shading in the previous example shows the configuration required to create a subinterface running point-to-point Frame Relay with IP.

Of course, it is imperative to check any configuration on a network device, because any errors could potentially bring down the entire network. To verify the configuration, there is a wealth of Cisco commands. They are covered in the following section.

Checking the Configuration of OSPF on a Single Router
The set of commands shown in Table 7-5 is invaluable in both configuration and maintenance of a live network. These commands are particularly useful in troubleshooting the network. As such, these commands are a necessary set of tools for use on a daily basis, for the CCNP/CCDP/CCIP BSCI exam, and for the CCIE lab exam.

238 Chapter 7: Configuring OSPF in a Single Area
Table 7-5 The show Command Options for OSPF

Command Options for OSPF
NOTE An excellent resource is the Cisco web site, which takes you through a series of
flowcharts to identify a particular problem and how to solve it. These flowcharts are nested,
so expect to work through several levels. This is an excellent resource for understanding
troubleshooting techniques that are beyond the scope of this book or the BSCI exam. You can
find the flowcharts at this URL:
_http://www.cisco.com/warp/public/104/trouble_main.

Understanding the output of these commands is important. This is not just because the output might constitute questions on the exam, but because the capability to analyze what is happening on the network demands a thorough understanding of the concepts explained in this chapter. You need to understand the concepts in this chapter to interpret the output of a show command.

The OSPF show commands are highly detailed and give a comprehensive understanding of the state of the network.

No comments yet

Leave a Reply

You must be logged in to post a comment.