Cisco Knowledge Suite Cisco SystemsCisco Press
   

   
Home
MyCKS
Cutting Edge
Certification
Core Reference
Guided Learning
   
Networking Architecture
LAN
WAN
Switching
Internet Protocols (IP)
Network Protocols
Transport and Application Protocols
Desktop Protocols
Security and Troubleshooting
Network Resources and Management
Integrated Services
 

Optimizing Routing Update Operation

   

< Back Contents Next >

Optimizing Routing Update Operation

  

 

Controlling Routing Update Traffic

  

 

Using Multiple Routing Protocols

  

 

Using and Configuring Redistribution

  

 

Verifying Redistribution Operation

  

 

Summary

Save to MyCKS

 
Advanced Cisco Router Configuration

From: Advanced Cisco Router Configuration
Author: Systems Cisco; Laura Chappell
Publisher: Cisco Press (53)
More Information

11. Optimizing Routing Update Operation

This chapter discusses some of the more commonly used capabilities for controlling when and how routers receive and send routing updates. The capabilities covered include passive interfaces, default routes, static routes, route filtering, and redistributing routes between different routing protocols.

Understanding these capabilities and configurations enables you to design and configure the most efficient routing environment.

Controlling Routing Update Traffic

Thus far, you have learned a variety of routing protocols and how they propagate routing information throughout an internetwork. There are times, however, when you do not want routing information propagated, as in the following examples:

  • When using an on-demand WAN link—You may want to minimize, or stop entirely, the exchange of routing update information across this type of link; otherwise, the link will remain up constantly.

  • When you want to prevent routing loops—Many companies have large enough networks where redundant paths are prominent. In some cases, for example, when a path is learned to the same destination by two different routing protocols, you may want to filter the propagation of one of the paths.

  • When you want to preserve bandwidth—Available bandwidth becomes a precious resource as networks grow. You can ensure maximum bandwidth availability for data traffic by reducing unnecessary routing update traffic.

This section discusses the following ways you can control or prevent routing update exchange and propagation:

  • Passive interface—Prevents all routing updates from being sent through an interface. For EIGRP and OSPF, this method includes hello protocol packets.

  • Default routes—Instructs the router that if it does not have a route for a given destination, to send the packet to the default route.

  • Static routes—A route to a destination that you configured in the router. In contrast, dynamic routes are those learned via routing protocol such as RIP or EIGRP.

  • Route update filtering—Use access lists to filter route update traffic about specific networks.

In order to make these capabilities effective, you must know your network traffic patterns and know what the intended goal is when applying the capabilities. If you do not know the problem you want to resolve when using these capabilities, you will not know how to verify that they are being effective.

TIP

You can use an analyzer, such as Network Associates' Sniffer, to monitor and document network traffic patterns.

Using and Configuring the passive-interface Command

The passive-interface command prevents all routing updates for a given routing protocol from being sent to or received from a network via a specific interface.

Remember that when using the passive-interface command in a network using a link-state routing protocol, the command prevents the router from establishing a neighbor adjacency with other routers connected to the same link as the one specified in the command. An adjacency cannot be established because the hello protocol is used to verify bidirectional communication between routers. If a router is configured to not send updates, then it cannot participate in bidirectional communication.

To configure a passive interface, regardless of the routing protocol, perform the following steps:

  1. Select the router that requires the passive interface.

  2. Determine which interface(s) you do not want routing update traffic to be sent through.

  3. Configure the passive interface as follows:

    router(config-router)#passive-interface type number

    • type refers to the type of interface, such as serial or Ethernet.

    • number refers to the interface number.

The passive-interface command is typically used in conjunction with other capabilities, as you will see in this chapter.

Use this command to prevent all route updates from being sent. If you want to be selective about which route updates not to send, you must use one of the other capabilities, such as route filtering and distribution filtering.

Using and Configuring Default Routes

Cisco enables you to configure default routes for other protocols. When you create a default route on a router, the router advertises an address of 0.0.0.0, in addition to the default network, unless it is IGRP or EIGRP. IGRP and EIGRP advertise only the network you configured. When a router receives the default route, it will forward any packets destined to a destination that does not appear in its routing table to the default route you configured. To specify a default route, perform the following steps:

  1. Determine which network(s) you want as the default network.

  2. Select the router(s) that need to have a default route defined. That is, identify the router(s) directly connected to the network for which a default route must be propagated, as shown in Figure11-1.

    Figure 11-1. Router p2r2 advertises the default route.

  3. Configure the selected network as default.

    • For IP, use the following command:

      router(config)#ip default-network network-number

      network-number is the number of the destination network.

    • For IPX, use the following command:

      router(config)#ipx advertised-default-route-only network

  4. For IGRP or RIP, enable these protocols to allow classless forwarding behavior.

    router(config)#ip classless

    This command allows the default path to be used for non-connected subnets of the same major classful network. If not enabled, the router drops the packet.

Using and Configuring Static Routes

Static routes are routes that you can manually configure on the router. Static routes are used most often to perform the following tasks:

  • Define specific routes to use when two autonomous systems must exchange routing information, rather than having entire routing tables exchanged.

  • Define static routes in stub environments where there is and will always be only one point of connection between two remote locations.

  • Define routes to destinations over a WAN link to eliminate the need for a dynamic routing protocol. That is, when you do not want routing updates to enable or cross the link (this situation is discussed in Chapter 14, “Configuring Dial-on-Demand Routing”).

When configuring static routes, keep in mind the following considerations:

  • When using static routes, all participating routers must have static routes defined so that they can advertise the remote networks. This requirement is necessary because static routes replace routing updates.

    If you want a router to advertise a static route, you must use the redistribution capability. This capability is discussed later in this chapter.

  • Static route entries must be defined for all routes that a router is responsible for. To reduce the number of static route entries, you can define a default static route. Default static routes are advertised (redistributed) automatically.

In Figure11-2, for example, router p1r2 has a static route entry for network 172.68.0.0.

Figure 11-2. Router p1r2 has one static route entry.

TIP

Only a default static route is automatically advertised.

To configure a static route, the commands vary by protocol. Although the following tasks apply to all protocols that support static routes, only the commands for IP and IPX are shown. For more command information, refer to www.cisco.com.

  1. Determine which networks you want defined as static. For example, if you are configuring static routes on a WAN router that is connecting to a branch office, you probably want to select the networks at the branch office.

  2. Determine the next-hop router to the destination networks or local router's interface that will call the remote router.

  3. Configure the static route on each router.

    • For IP, use the following command:

      router(config)#ip route prefix mask {address | interface}
            [distance] [permanent]

      prefix is the network address (in classful format) for the destination.

      mask is the prefix mask for the network address.

      address is the IP address of the next-hop router that can be used to reach that network.

      interface is the network interface to use to get to the destination network.

      distance is the administrative distance to assign to this route. (Administrative distance refers to how believable the routing protocol is. It is discussed later in this chapter.)

      permanent ensures that if the interface associated with the route goes down, the route will still remain in the routing table.

    • Static routes for IPX use the following commands:

      router(config)#ipx route {network | default}
            {network.node | interface} [floating-static]

      network is the network address of the destination.

      default specifies the destination as a static entry for the default route.

      network.node is the network address and node ID of the next-hop router.

      interface is the network interface to use to get to the destination network. Typical use is when using IPXWAN unnumbered interfaces.

      floating static enables you to make IPX static routes floating (they are permanent by default). A floating static route is a static route that can be overridden by a dynamically learned route.

    • If you configure IPX static routes, you must also consider configuring static SAPs. A static SAP is a server entry that is manually entered into the router's server information tables instead of being dynamically learned through a SAP broadcast.

Using and Configuring Route Filters

In the traffic management module, you learned how to filter different types of data and protocol traffic. This section discusses how access lists can be used to filter route (and IPX SAP) updates.

The Cisco IOS software can filter incoming and outgoing routing updates. In general, the process the router uses, shown in Figure11-3, is as follows:

  1. The router receives a routing update or is getting ready to send an update about one or more networks.

  2. The router looks at the interface involved with the action.

    If it is an incoming update, for example, then the interface on which it arrived is checked. If it is an update that must be advertised, the interface out of which it should be advertised is checked.

  3. The router determines whether a filter is associated with the interface or with the routing protocol. Filters can be associated with all interfaces running a specific routing protocol.

    If a filter is associated with the interface, the router views the access list to learn if there is a match for the given routing update.

    If a filter is not associated with the interface, the packet is processed as normal.

  4. If there is a match, the route entry is processed as configured.

    If no match is found, the update is dropped due to the implicit deny all at the end of the access list.

Use route filters when you want to advertise selected routes only, such as only a default route and no other routes that are usually advertised by the router.

If you want to filter all updates advertised by an interface, use the passive-interface command instead.

Figure 11-3. Filtering decision process.

You can filter routing update traffic for any protocol by defining an access list and applying it to specific routing protocol. To configure a filter, perform the following steps:

  1. Identify the network addresses you want to filter and create an access list.

  2. Determine whether you want to filter them on an incoming or outgoing interface.

  3. Assign the access list to filter outgoing routing updates:

    router(config-router)#distribute-list access-list-number | name out
          [interface-name | routing-process | autonomous-system-number]

distribute-list Command

Description

access-list-number

Standard access list number.

out

Define the filtering on outgoing routing updates.

interface-name

(Optional) Interface name.

routing-process

(Optional) Name of the routing process or the keyword static or connected.

Or, to assign the access list to filter incoming routing updates, use the following:

router(config-router)# distribute-list {access-list-number | namein
      [type number]

type indicates the interface type.

number indicates the interface number of where the access list should be applied on incoming updates.

Note that the distribute-list syntax is different, depending on whether it is an inbound or outbound filter.

distribute-list Command

Description

access-list-number

Standard access list number

in

Defines the filtering on incoming routing updates

interface-name

(Optional) Interface name

The distribute-list command uses standard access lists and can be applied to inbound or outbound routing updates. It can be used with IP and IPX.

IPX Route Filtering Configuration Example

In Figure11-4, networks 4a and 9e have been filtered from interface s0.

The distribute-list command applies access list 800 to outbound packets. The access list does not allow routing information from networks 4a and 9e to be distributed out the s0 interface. As a result, networks 4a and 9e are hidden.

When you filter RIP information, consider filtering SAPs as well.

Figure 11-4. Only network 6c is advertised in routing updates.

Command

Description

distribute-list 800 out s0

Applies access list 800 as a route redistribution filter on routing updates sent on serial 0

Access-list 800 permit 6c

800

Access list number

Permit

Traffic matching the parameters can be forwarded

6c

Network number

   

< Back Contents Next >

Save to MyCKS

 

Breaking News

One of the primary architects of OpenCable, Michael Adams, explains the key concepts of this initiative in his book OpenCable Architecture.

Expert Advice

Ralph Droms, Ph.D., author of The DHCP Handbook and chair of the IETF Dynamic Host Configuration Working Group, guides you to his top picks for reliable DHCP-related information.

Just Published

Residential Broadband, Second Edition
by George Abe

Introduces the topics surrounding high-speed networks to the home. It is written for anyone seeking a broad-based familiarity with the issues of residential broadband (RBB) including product developers, engineers, network designers, business people, professionals in legal and regulatory positions, and industry analysts.

             
     

From the Brains at InformIT

|

Contact Us

|

Copyright, Terms & Conditions

|

Privacy Policy

 

© Copyright 2000 InformIT. All rights reserved.