Avaya IP Exterior Gateway Protocols User Manual

BayRS Version 15.0 Part No. 308628-15.0 Rev 00

June 2001 600 Technology Park Drive

Billerica, MA 01821-4130

Configuring IP Exterior Gateway Protocols (BGP and EGP)

and recommendations in this document are believed to be accurate and reliable, but are presented without express or implied warranty. Users must t ak e full respo nsib ility fo r th e ir app lica tio ns o f any products specified in this document. The information in this document is proprietary to Nortel Networks Inc.

The software described in this docume nt is furnished under a license agreement and may only be used in accordance with the terms of that license. The software license agreement is included in this document.

Trademarks

Nortel Networks, the Nortel Networks logo, the Globemark, Unified Networks, AN, ARN, ASN, BayRS, BCC, and Passport are trademarks of Nortel Networks.

Adobe and Acrobat Reader are trademarks of Adobe Systems Incorporated. Ethernet is a tradema r k of X ero x C orp oration.

Restricted Rights Legend

Use, duplication, or disclosure by the United States Government is subject to restrictions as set forth in subparagraph (c)(1)(ii) of the Rights in Technical Data and Computer Software clause at DFARS 252.227-7013.

Notwithstanding any other license ag reement that may pertain to, or accompany the delivery of, this comput er software, the rights of the United States Government regarding its use, reproduction, and disclosure are as set forth in the Commercial Computer Software-Restricted Rights cl ause at FAR 52.227-19.

Statement of Conditions

In the interest of improving internal design, operational function, and/or reliability, Nortel Networks Inc. reserves the right to make changes to the pr oducts described in this document without notice.

Nortel Networks Inc. does not assume any liability that may occur due to the use or application of the product(s) or circuit layout(s) described herein.

Portions of the code in this software product may be Copyright © 1988, Regents of the University of California. All rights reserved. Redistribution and use in source and binary forms of such portions are permitted, provided that the above copyright notice and this paragraph are duplicated in all such forms and that any documentation, advertising materials, and other materials related to such distribution and use acknowledge that such portions of th e software were developed by the University of California, Berkeley. The name of the University may not be used to endorse or promote products derived from such portions of the software without specific prior written permission.

SUCH PORTIONS OF THE SOFTWARE ARE PROVIDED “AS IS” AND WITHOUT ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.

In addition, the program and information contained herei n are l icensed only pursuant to a license agreement that contains restrictions on use and disclosure (that may incorporate by reference certain limitations and notices imposed by third parties).

308628-15.0 Rev 00

Nortel Networks Inc. Software License Agreement

NOTICE: Please carefully read this license agreement before copying or usin g the accompanying software or installing the hardware unit with pre-enabled software (each of which is referred to as “Software” in this Agreement). BY COPYING OR USING THE SOFTWARE, YOU ACCEPT ALL OF THE TERMS AND CONDITIONS OF THIS LICENSE AGREEMENT. THE TERMS EXPRESSED IN THIS AGREEMENT ARE THE ONLY TERMS UNDER WHICH NORTEL NETWORKS WILL PERMIT YOU TO USE THE SOFTWARE. If you do not accept these terms and conditions, return the product, unused and in the original shipping container, within 30 days of purchase to obtain a credit for the full purchase price.

1. License grant. Nortel Networks Inc. (“Nortel Networks”) grants the end user of the Software (“Licensee”) a personal, nonexc lusi v e, no ntransferab le licen se: a) to u se the So ftw are eithe r on a sing le compu ter or, if applicable, on a single authorized device identified by host ID, for which it was originally acquired; b) to copy the Software solely for backup purposes i n support of authorized use of the Software; and c) to use and copy the associated user manual solely in support of authorized use of the Software by Licensee. This license applies to the Software only and does not extend to Nortel Networks Agent software or other Nortel Networks software products. Nortel Networks Agent software or other Nortel Networks software products are licensed for use under the terms of the applicable Nortel Networks Inc. Software License Agreement that accompanies such software and upon payment by the end user of the applicable license fees for such software.

2. Restrictions on use; reservation of rights. The Software and user manuals are protected under copyright laws. Nortel Networks and/or its licensors retain all title and ownership in both the Software and user manuals, including any revisions made by Nortel Networks or its licensors. The copyright notice must be reproduce d and included with any copy of any portion of the Software or user manuals. Licensee may not modify, translate, decompile, disassemble, use for any competitive analysis, reverse engineer, distribute, or create derivative works from the Software or user manuals or any copy, in whole or in part. Except as expressly provided in this Agreement, Licensee may not copy or transfer the Software or user manuals, in whole or in part. The Software and user manuals embody Nortel Networks’ and its licensors’ confidential and pro pri etary inte lle ctu al pro p erty. Licensee shall not sublicense, assign, or otherwise disclose to any third party the Software, or any information about the operation, design, performance, or implementation of the Software and user manuals that is confidential to Nortel Networks and its licensors; however, Licensee may grant permission to its consultants, subcontractors, and agents to use the Software at Licensee’s facility, provided they have agreed to use the Software only in accordance with the terms of this license.

3. Limited warranty . Nortel Networks warrants each item of Software, as delivered by Nortel Networks and properly installed and operated on Nortel Networks hardware or other equipment it is originally licensed for, to function substantially as described in its accompanying user manual duri ng its warranty period, which begins on the date Software is first shipped to Licensee. If an y item of Softwa re fails to so f unction during its wa rranty period, as the sole remedy Nortel Networks will at its discretion provide a suitable fix, patch, or workaround for the problem that may be included in a future Software release. Nortel Networks further warrants to Licensee that the media on which the Software is provided will be free from defec ts in materials and wo rkmanship under normal u se for a period of 90 days from the date Software is first shipped to Licensee. Nortel Networks will replace defective media at no charge if it is returned to Nortel Net w orks during the warrant y period along with proof of the date of shipment . This warranty does not apply if the media has been damaged as a result of accident, misuse, or abuse. The L i censee assumes all responsibility for selection of the Softw are to achieve Licensee’s intended results and for the installation, use, and results obtained from the Software. Nortel Networks does not warrant a) that the functions contained in the software will meet the Licensee’s requirements, b) that the Softw are will operate in the hardw are or software combina tions that the Licensee may select, c) that the operation of the Software will be uninterrupted or error free, or d) that all defects in the operation of the Softw are will be correcte d. Nortel Netw orks is not oblig ated to remedy any Sof tware defec t that cannot be reproduced with the latest Software release. These warranties do not apply to the Software if it has been (i) altered, except by Nortel Networks or in accordance with its instructions; (ii) used in conjunction with another vendor’s produc t, resulting in the defect; or (iii) damaged by improper environment, abuse, misuse, accident, or negligence. THE FOREGOING WARRANTIES AND LIMITATIONS ARE EXCLUSIVE REMEDIES AND ARE IN LIEU OF ALL OTHER WARRANTIES EXPRESS OR IMPLIED, INCLUDING WITHOUT LIMITA T ION ANY WARRANTY OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Licensee is responsible

308628-15.0 Rev 00

iii

for the security of its own data and information and for maintaining adequate procedures apart from the Software to reconstruct lost or altered files, data, or programs.

4. Limitation of liability. IN NO EVENT WILL NORTEL NETWORKS OR ITS LICENSORS BE LIABLE FOR ANY COST OF SUBSTITUTE PROCUREMENT; SPECIAL, INDIRECT, INCIDENTAL, OR CONSEQUENTIAL DAMAGES; OR ANY DAMAGES RESULTING FROM INACCURATE OR LOST DATA OR LOSS OF USE OR PROFITS ARISING OUT OF OR IN CONNECTION WITH THE PERFORMANCE OF THE SOFTWARE, EVEN IF NORTEL NETWORKS HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. IN NO EVENT SHALL THE LIABILITY OF NORTEL NETWORKS RELATING TO THE SOFTWARE OR THIS AGREEM ENT EXCEED THE PRICE PAID TO NORTEL NETWORKS FOR THE SOFTWARE LICENSE.

5. Government licensees. This provision applies to all Softw a re and do cum entation acquired directly o r in di rectly by or on behalf of the United States Government. The Software and documentation are commercial products, licensed on the open market at market prices, and were developed entirely at private expense and without th e use of any U.S. Government funds. The license to the U.S. Government is granted only with restricted rights, and use, duplication, or disclosure by the U.S. Government is subject to the restrictions set forth in subparagraph (c)(1) of the Commercial Computer Software––Restricted Rights clause of FAR 52.227-19 and the limitatio ns set o ut in this license for civilian agencies, and subparagraph (c)(1)(ii) of the Rights in Technical Data and Computer Software clause of DFARS

252.227-7013, for agencies of the Department of Defense or their successors, which ever is applicable.

6. Use of software in the European Community. This provision applies to all Software acquired for use within the European Community. If Licensee uses the Software within a country in the European Community, the Software Directive enacted by the Council of European Communities Directive dated 14 May, 1991, will apply to the examination of the Software to facilitate interoperability. Licensee agrees to notify Nortel Networks of any such intended examination of the Software and may procure support and assistance from Nortel Networks.

7. Term and termination. This license is effective until ter mina ted; however, all of the restrictions with respect to Nortel Networks’ copyright in the Software and user manuals will cease being effective at the date of expiration of the Nortel Networks copyright; t hose restrictions relating to use and disclosure of Nortel Networks’ confidential information shall continue in effect. Licensee may terminate this license at any time. The license will automatically terminate if Licensee fails to comply with any of the terms and conditions of the license. Upon termination for any reason, Licensee will immediately destroy or return to Nortel Networks the Software, user manuals, and all cop ies. Nortel Networks is not liable to Licensee for damages in any form solely by reason of the termination of this license.

8. Export and re-export. Licensee agrees not to export, directly or indirectly, the Software or related technical data or information without f irst obta ining a n y required e xport lice nses or oth er go v ernmenta l appro vals. Without limiting the foregoing, Licen see , on b e half of itself and its subsidiarie s an d affiliates, agrees that it will not , wi tho ut first obtaining all export licenses and approvals required by the U.S. Government: (i) export, r e-export, transfer, or divert any such Software or technical data, or any direct product thereof, to any country to which such exports or re-exports are restricted or embargoed under United States export control laws and regulations, or to any national or resident of such restricted or embargoed countries; or (ii) provide the Software or related technical data or information to any military end user or for any military end use, including the design, development, or production of any chemical, nuclear, or biological weapons.

9. General. If any provision of this Agreement is held to be invalid or unenforceable by a court of competent jurisdiction, the remainder of the provisions of this Agreement shall remain in full force and effect. This Agreement will be governed by the laws of the state of California.

Should you have any questions concerning this Agreement , contact Nortel Networks Inc. , 2375 N. Glenville Dr., Richardson, TX 75082.

LICENSEE ACKNOWLEDGES THAT LICENSEE HAS READ THIS AGREEMENT, UNDERSTANDS IT, AND AGREES TO BE BOUND BY ITS TERMS AND CONDITIONS. LICENSEE FURTHER AGREES THAT THIS AGREEMENT IS THE ENTIRE AND EXCLUSIVE AGREEMENT BETWEEN NORTEL NETWORKS AND LICENSEE, WHICH SUPERSEDES ALL PRIOR ORAL AND WRITTEN AGREEMENTS AND COMMUNICATIONS BETWEEN THE PARTIES PERTAINING TO THE SUBJECT MATTER OF THIS AGREEMENT. NO DIFFERENT OR ADDITIONAL TERMS WILL BE ENFORCEABLE AGAINST NORTEL NETWORKS UNLESS NORTEL NETWORKS GIVES ITS EXPRESS WRITTEN CONSENT, INCLUDING AN EXPRESS WAIVER OF THE TERMS OF THIS AGREEMENT.

308628-15.0 Rev 00

Preface

Before You Begin ............................................................................................................. xv

Text Conventions .............................................................................................................xvi

Acronyms ........................... .......................... .......................... ......................... ...............xviii

Related Publications ........................................................................................................ xx

How to Get Help ..............................................................................................................xxi

Chapter 1 Exterior Gateway Protocols (BGP and EGP)

Autonomous Systems and Gateway Protocols ...............................................................1-1

Border Gateway Protocol (BGP) ..............................................................................1-3

Exterior Gateway Protocol (EGP) .............................................................................1-3

Classless Interdomain Routing ................................................................................1-4

BGP Concepts and Terminology .....................................................................................1-4

Peer-to-Peer Sessions .............................................................................................1-5

Stub and Multihomed Autonomous Systems ...........................................................1-6

Interior BGP Routing ... ...... ....... ...... ............................................. ....... ...... ....... ...... ...1-6

IBGP Route Reflector ...............................................................................................1-7

Equal-Cost Multipath ................................................................................................1-8

BGP Updates ...........................................................................................................1-8

Path Attributes ..........................................................................................................1-9

BGP/OSPF Interaction ...........................................................................................1-10

BGP-4 Confederations ...........................................................................................1-11

BGP-4 TCP MD5 Message Authentication ........................................................... .1-11

BGP Implementation Notes ..........................................................................................1-12

308628-15.0 Rev 00

Chapter 2 Starting BGP and EGP Services

Starting BGP with the BCC ............................................................................................2-1

Step 1: Configuring Global BGP ..............................................................................2-1

Step 2: Defining a Peer-to-Peer Connection ............................................................2-2

Starting IP and BGP with Site Manager .........................................................................2-3

Deleting BGP with Site Manager ..............................................................................2-4

Deleting BGP-3 and BGP-4 with Site Manager ........................................................2-5

Starting IP and EGP with Site Manager .........................................................................2-6

Deleting EGP from the Router .................................................................................2-7

Chapter 3 Configuring Global BGP Parameters

Disabling and Reenabling BGP ................................ ...... ....... .........................................3-2

Supplying a BGP Router ID ............................................................................................3-4

Identifying the Local AS ..................................................................................................3-5

Disabling and Reenabling IBGP Support .............................. ...... ...................................3-6

Specifying Route Types for IBGP Advertisements ..........................................................3-7

Enabling BGP Interaction with OSPF and RIP ...............................................................3-9

Setting the Update Interval Timer .................................................................................3-10

Allowing Redundant Connections ................................... ....... ...... ...... ...........................3-11

Enabling Multihop Connections ....................................................................................3-13

Disabling and Reenabling Dynamic Policy Configuration ............ ...... ...........................3-15

Configuring the BGP Soloist Slot Mask ........................................................................3-16

Disabling and Reenabling Route Aggregation ............................. ...... ....... ...... ....... ...... .3-17

Enabling and Disabling Black Hole Punching ...............................................................3-18

Disabling and Reenabling the BGP-4 MED Attribute ....................................................3-20

Configuring BGP-4 Confederations ..............................................................................3-21

Disabling BayRS Local Preference Calculation and Route Selection ...........................3-25

Calculating BGP-4 Local Preference Values ..........................................................3-25

Best-Route Selection ............................................... ....... ...... ...... ...........................3-27

Configuring BGP Message Logging .............................................................................3-29

Configuring EBGP Route Flap Damping ......................................................................3-31

Assigning Weight Classes and Values to an AS ...........................................................3-37

308628-15.0 Rev 00

Chapter 4 Configuring BGP Peers

Defining a Peer-to-Peer Session ....................................................................................4-2

Initiating a Peer-to-Peer Session ....................................................................................4-4

Negotiating the BGP Version ..........................................................................................4-6

Keeping the Connection Alive .........................................................................................4-8

Setting the External Advertisement Timer ......................................................................4-9

Specifying a Holddown Time ........................................................................................4-11

Setting a Minimum AS Origination Interval ...................................................................4-12

Overriding the Local AS Number ..................................................................................4-14

Specifying a Maximum Update Size .............................................................................4-14

Specifying a Time-to-Live Value ...................................................................................4-16

Specifying the Next-Hop Router ...................................................................................4-17

Setting the Route Echo Switch .....................................................................................4-18

Disabling and Reenabling Loop Detection ...................... ....... ...... .................................4-20

Configuring Peers over an Unnumbered Point-to-Point Link ........................................4-21

Configuring and Enabling MD5 Authentication .............................................................4-22

Entering and Storing MD5 Authentication Keys .....................................................4-23

Initializing TCP with the MD5 Option ......................................................................4-24

Generating MD5 Signatures on Transmitted BGP TCP Packets ............................4-24

Verifying MD5 Signatures on Received BGP TCP Packets ...................................4-25

Configuring BGP-4 Authentication .........................................................................4-25

Chapter 5 Configuring BGP Accept and Announce Policies

Configuring a BGP Accept Policy ...................................................................................5-2

Specifying Match Criteria for a BGP Accept Policy ..................................................5-6

Supplying Modification Values for a BGP Accept Policy ........................................5-10

Configuring a BGP Announce Policy ............................................................................5-13

Announce Policy Guidelines ...................................................................................5-13

Specifying Match Criteria for a BGP Announce Policy ...........................................5-16

Supplying Modification Values for a BGP Announce Policy ...................................5-22

Configuring BGP-4 AS Path Pattern-Matching .............................................................5-27

308628-15.0 Rev 00

vii

Chapter 6 Configuring a Route Reflector

Configuring a Single Route Reflector in an AS ...............................................................6-3

Configuring a Route Reflector Cluster ............................................................................6-5

Configuring Multiple RR Clusters in an AS .....................................................................6-7

Configuring an RR Client ..............................................................................................6-10

Chapter 7 Configuring Route and Traffic Balancing

Configuring IBGP for Route and Traffic Balancing ..........................................................7-1

Configuring EBGP for Route and Traffic Balancing ........................................................7-5

Chapter 8 Customizing EGP Services

EGP Concepts and Terminology .....................................................................................8-2

EGP Implementation Notes ............................................................................................8-5

Customizing EGP on the Router ....................................................................................8-6

Enabling and Disabling EGP ....................................................................................8-6

Supplying a Local AS Number .................................................................................8-7

Configuring a Neighbor ...................................................................................................8-7

Specifying the Neighbor’s Address ..........................................................................8-8

Specifying the Gateway Mode ..................................................................................8-9

Enabling and Disabling the Neighbor Relationship ................................................8-10

Choosing the Acquisition Mode ..............................................................................8-11

Choosing the Poll Mode .........................................................................................8-12

Setting Neighbor Timers ........................................................................................8-13

Appendix A Site Manager Parameters

BGP Parameters ............................................................................................................ A-2

BGP Configuration Parameters ............................................................................... A-3

BGP Global Parameters .......................................................................................... A-4

BGP-3 Global Parameter ...................................................................................... A-11

BGP-4 Global Parameter ...................................................................................... A-11

BGP Peer Parameters ........................................................................................... A-11

BGP AS Weight and Weight Class Parameters .................................................... A-20

BGP Event Message Parameters .. ....... ...... ....... ...... ....... ...... ...... ....... ................... A-24

viii

308628-15.0 Rev 00

EGP Parameters .......................................................................................................... A-25

EGP Global Parameters ....................................................................................... A-25

EGP Neighbor Parameters .................................................................................... A-26

Routing Policy Par a meters .......................................................................................... A-29

Common Accept Policy Parameters ...................................................................... A-29

EGP-Specific Accept Policy Parameters ............................................................... A-34

BGP-3-Specific Accept Policy Parameters ........................................................... A-36

BGP-4-Specific Accept Policy Parameters ............................................................ A-40

Common Announce Policy Parameters ................................................................. A-50

EGP-Specific Announce Policy Parameters .......................................................... A-68

BGP-3-Specific Announce Policy Parameters ....................................................... A-70

BGP-4-Specific Announce Policy Parameters ....................................................... A-74

Appendix B Converting Cisco to Nortel Networks Equivalents

Configuration Command Equivalents ............................................................................ B-1

Interpreting the Configuration Command Equivalents Table ......................................... B-6

Comparing the Operational Commands ................... ...... ....... ...... ...... ....... ...... ....... ...... .. B-8

Interpreting the Operational Commands Table .............................................................. B-9

Comparing BGP Route Selection Processes .............................................................. B-11

Regular Expression Symbols ...................... ...... ....... ...... ....... ...... ................................ B-12

Nortel Networks AS Path Pattern-Matching Symbols .................................................. B-13

Index

308628-15.0 Rev 00

Figures

Figure 1-1. Internetwork Segmented into Three Autonomous Systems .....................1-2

Figure 1-2. BGP Connecting Two Autonomous Systems Running OSPF ..................1-4

Figure 1-3. Transit AS .................................................................................................1-7

Figure 3-1. BGP Confederation ................................................................................3-23

Figure 4-1. Establishing and Confirming a Connection Between BGP Peers ............4-4

Figure 4-2. BGP over an Unnumbered Point-to-Point Link .......................................4-21

Figure 6-1. IBGP Single Route Reflector Topology ....................................................6-2

Figure 7-1. BGP/OSPF Autonomous System ............................................................7-2

Figure 7-2. IBGP ECMP Route Balancing ..................................................................7-3

Figure 7-3. IBGP ECMP Traffic Balancing ..................................................................7-4

Figure 7-4. ECMP Static Routes ................... ....... ...... ....... .........................................7-6

Figure 8-1. EGP Connection Between Two Autonomous Systems Running RIP .......8-2

308628-15.0 Rev 00

Tables

Table 1-1. BGP-3 Path Attributes ..............................................................................1-9

Table 1-2. BGP-4 Mandatory Path Attributes ............................................................1-9

Table 1-3. BGP-4 Optional Path Attributes .............................................................1-10

Table 3-1. Route Types for BGP Advertisements ......................................................3-8

Table 3-2. Slot Mask Parameter Values ..................................................................3-16

Table 3-3. Black Hole Punching Parameter Settings ..............................................3-19

Table 3-4. Best-Route Selection Rules ........................................ ....... ...... .............. 3 -27

Table 3-5. Local Preference Calculation Method ....................................................3-28

Table 3-6. Route Flap Damping Template Parameters ...........................................3-33

Table 4-1. MD5 Signature Verification Rules on BGP TCP Packets .......................4-25

Table 5-1. BCC Definition Parameters for BGP Accept Policies ...............................5-4

Table 5-2. BCC Match Parameters for BGP Accept Policies ....................................5-7

Table 5-3. BCC Modification Parameters for BGP Accept Policies .........................5-10

Table 5-4. BCC Definition Parameters for BGP Announce Policies ........................5-14

Table 5-5. BCC Match Parameters for BGP Announce Policies .............................5-17

Table 5-6. BCC Modification Parameters for BGP Announce Policies ....................5-22

Table 5-7. Characters in AS Path Pattern-Matching ...............................................5-27

Table 7-1. IBGP ECMP Methods ..............................................................................7-4

Table 7-2. EBGP ECMP Methods .............................................................................7-8

Table 8-1. Router Mode Determinator ......................................................................8-3

Table B-1. Cisco to Nortel Networks BGP Translation ............................................. B-2

Table B-2. Cisco and Nortel Networks BGP Operational Commands ...................... B-8

Table B-3. Route Selection Process Comparison .................................................. B-11

Table B-4. Regular Expression Symbols ................................................................ B-12

Table B-5. Nortel Networks AS Path Pattern-Matching Symbols ........................... B-13

308628-15.0 Rev 00

xiii

Preface

Routers at the borders or edg es of a utonomous s ystems a re call ed gateways. These gateways use exterior gateway protocols to exchange rea chability information with each other and route packets between routing domains. This guide describes how to configure and use the IP Border Gateway Prot ocol (BGP) and the Exte rior Gateway Protocol (EGP).

You can use the Bay Command Console (BCC*) or Site Manager to configure BGP and EGP on a router. In this guide, you will find instructions for using both the BCC and Site Manager.

Before You Be gin

Before using this guide, you must complete the following procedures. For a new router:

• Install the router (see the installation guide that came with your router).

• Connect the router to the network and create a pilot configuration file (see

Quick-Starting Routers, Configuring Remote Access for AN and Passport ARN Routers

Make sure that you are running the latest version of Nortel Networks* BayRS* and Site Manager software. For information about upgrading BayRS and Site Manager, see the upgrading guide for your version of BayRS.

308628-15.0 Rev 00

, or

Connecting ASN Routers to a Network)

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Text Conventions

This guide uses the following text conventions:

angle brackets (< >) Indicate that you choose the text to enter based on the

description inside the brackets. Do not type the brackets when entering the command.

Example: If the command syntax is:

ping

<ip_address>

ping 192.32.10.12

, you enter:

bold text

Indicates command names and options and text that you need to enter.

Example: Enter Example: Use the

show ip {alerts | routes

dinfo

command.

braces ({}) Indicate required elements in syntax descriptions

where there is more than one option. You must choose only one of the options. Do not type the braces when entering the command.

Example: If the command syntax is:

show ip {alerts | routes show ip alerts or show ip routes

}

, you must enter either:

, but not both.

brackets ([ ]) Indicate optional elements in syntax descriptions. Do

not type the brackets when entering the command. Example: If the command syntax is:

show ip interfaces [-alerts show ip interfaces

]

, you can enter either:

show ip interfaces -alerts

ellipsis points (. . . ) Indicate that you repeat the last element of the

command as needed.

xvi

Example: If the command syntax is:

ethernet/2/1 ethernet/2/1

[<parameter> <value>]

and as many parameter-value pairs as

needed.

. . .

, you enter

308628-15.0 Rev 00

Preface

italic text Indicates new terms, book titles, and variables in

command syntax descri pti ons. Where a v a ri abl e is two or more words, the words are connected by an underscore.

Example: If the command syntax is:

show at <valid_route> valid_route

is one variable and you substitute one value

for it.

screen text Indicates system output, for example, prompts and

system messages. Example:

Set Trap Monitor Filters

separator ( > ) Shows menu paths.

Example: Protocols > IP identifi es the IP option on t h e Protocols menu.

vertical line (

) Separates choices for command keywords and

arguments. Enter only one of the choices. Do not type the vertical line when entering the command.

Example: If the command syntax is:

show ip {alerts | routes} show ip alerts

, you enter either:

show ip routes

, but not both.

308628-15.0 Rev 00

xvii

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Acronyms

This guide uses the following acronyms:

ARP Address Resolution Protocol AS autonomous system ATM asynchronous transfer mode BGP Border Gatew a y Protoc ol CIDR classless interdomain rout ing DES data encryption standard EBGP Exterior Border Gateway Protocol ECMP equal-cost multipath EGP Exterior Gateway Protocol FDDI Fiber Distributed Data Interface HSSI High Speed Serial Interface IBGP Interior Border Gateway Protocol

xviii

ICMP Internet Control Message Protocol IGP interior gateway protocol IP Internet Protocol ISP Internet service provider LAN local area network MD5 M essage Digest 5 MED multiexit discriminator MEK message encryption key MIB management information base MSS maximum segment size NLRI network layer reachability information NPK node protection key NVRAM nonvolatile random access memory OSPF Open Shortest Path First

308628-15.0 Rev 00

Preface

PPP Point-to-Point Protocol PVC permanent virtual circuit RARP Reverse Address Resolution Protocol RFC request for comments RIP Routing Information Protocol RR route reflector SMDS Switched Multimegabit Data Service TCP Transmission Control Protocol UDP User Datagram Protocol WAN wide area network

308628-15.0 Rev 00

xix

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Related Publications

For more information about IP services, refer to the following publications:

• Reference for BCC IP show Commands (part number 308603-14.20 Rev 00)

show

Provides descriptions of all commands that display BGP configuration and statistical data.

• Configuring IP, ARP, RARP, RIP, and OSPF Services (part number 308627-15.0 Rev 00)

Provides a description of IP, ARP, RARP, RIP, and OSPF services and instructions for configuring them.

• Configuring GRE, NAT, RIPSO, and BFE Services (part number 308625-14.20 Rev 00)

Provides a description of Generic Routing Encapsulation (GRE), Network Address Translation (NAT), Revised IP Security Option (RIPSO), and Blacker front-end services and instructions for configuring them.

You can print selected technical manuals and release notes free, directly from the Internet. Go to the www.nortelnetworks.com/documentation URL. Find the product for which you ne ed do cumen tat ion. Then locate the speci fic category and model or version for your hardware or software product. Use Adobe* Acrobat Reader* to open the manuals and release notes, search for the sections you need, and print them on most standard printers. Go to Adobe Systems at the

www.adobe.com URL to download a free copy of the Adobe Acrobat Reader.

commands for IP services, including the

You can purchase selected documentation sets, CDs, and technical publications through the Internet at the www1.fatbrain.com/documentation/nortel/ URL.

308628-15.0 Rev 00

How to Get Help

If you purchased a service contract for your Nortel Networks product from a distributor or authorized reseller, contact the technical support staff for that distributor or reseller for assistance.

If you purchased a Nort el Net w orks s ervic e progr am, cont act on e of t he fol lo win g Nortel Networks Technical Solutions Centers:

Technical Solutions Center Telephone

Europe, Middle East, and Africa (33) (4) 92-966-968 North America (800) 4NORTEL or (800) 466-7835 Asia Pacific (61) (2) 9927-8800 China (800) 810-5000

An Express Routing Code (ERC) is available for many Nortel Networks products and services. When you use an ERC, your call is routed to a technical support person who speciali zes in supp orting t hat product or servi ce. To locate an ERC for your product or service, go to the www12.nortelnetworks.com/ URL and click ERC at the bottom of the page.

Preface

308628-15.0 Rev 00

xxi

Chapter 1

Exterior Gateway Protocols (BGP and EGP)

This chapter introduces the concepts and terminology used in this guide.

Topic Page

Autonomous Systems and Gateway Protocols 1-1 BGP Concepts and Terminology 1-4 BGP Implementation Notes 1-12

Autonomous Systems a nd Gateway Protocols

LANs and WANs interconnected by IP routers for m a group of netw orks call ed an internetwork. For administrative purposes, an internetwork is divided into autonomous systems. An autonomous system (AS) is a group of routers (called gateways in IP terminolog y) and host s run by a singl e techni cal admi nistr ator th at has a single, clearly defined routing policy. Each autonomous system has its own unique AS number as signe d by the app ropri ate In ter net Regi stry entit y. Figure 1-1 shows a sample internetwork segmented into three autonomous systems.

308628-15.0 Rev 00

1-1

Configuring IP Exterior Gateway Protocols (BGP and EGP)

LAN

Router

Autonomous

system 1

LAN

Autonomous

system 2

LAN

Router

LAN

Router

Autonomous

system 3

Router

LAN

Router

Figure 1-1. Internetwork Segmented into Three Autonomous Systems

The routers at the edges (or borders) of autonomous systems are called gateways. These gate ways use exterior gatewa y protocols to exchange reachability information and to route packets between routing domains.

1-2

IP0006B

308628-15.0 Rev 00

Border Gateway Protocol (BGP)

The Border Gateway Protocol (BGP) is an exterior gateway protocol used by border routers to exchange network reachability information with other BGP systems. BGP routers form peer relationships with other BGP routers in other autonomous systems or within the same autonomous system. BGP peers transmit and receive current ro uting in form ation over a reli able tr anspor t la yer conne ction, making periodic updates unnecessary. BGP is designed for inter-AS exchanges, but can be used between multiple routing domains (for example, RIP to OSPF).

BGP peers exchange complete routing information only after the peer connection is established. Thereafter, BGP peers exchange routing updates. An update includes a network number, a list of autonomous systems that the routing information passed through (the AS path), and other path attributes that describe the route to a set of dest inati on net works. When mul tipl e path s are avail able, BGP compares the path attributes to choose the preferred path.

BGP exchanges information between ASs as well as between routers in the same AS. To differentiate between these uses, the latter is called interior BGP (IBGP).

Exterior Gateway Protocols (BGP and EGP)

Exterior Gateway Protocol (EGP)

You use the Exterior Gateway Protocol to exchange network reachability information between routers in different autonomous systems. An interior gateway protocol (IGP), such as RIP or OSPF, is used with in an AS to facilitate the communication of routing information within an autonomous system. The routers that serve as the end points of a connection between two autonomous systems also run an exterior gateway protocol, such as EGP-2.

Routers establish EGP neighbor relationships to periodically exchange reliable network reachability inf ormation. EGP neig hbors exchange complete reachability information, not just upda tes. T he rout er us es this infor mat ion to mainta in a li st of gateways, the networks the gateways can reach, and the corresponding distances.

Chapter 8, “

308628-15.0 Rev 00

Customizing EGP Services,” describes the use of EGP.

1-3

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Classless Interdomain Routing

Classless interdomain routing (CIDR) is an addressing scheme that uses supernet addresses to represent multiple IP destinations. Rather than advertise a separate route for each destination network in a supernet, a router uses a supernet address to advertise a single route (called an aggregate route) that represents all the destinations. CIDR reduces the size of the routing tables used to store advertised IP routes. BGP-4 supports classless interdomain routing.

BGP Concepts and Terminology

BGP is an exterior gateway protocol designed to exchange network reachability information with other BGP systems in other autonomous systems or within the same autonomous system.

Figure 1-2

shows two autonomous systems: AS1 and AS2. Networks within AS1 and AS2 are connected by routers running an interior gateway protocol—in this case, OSPF. The two ASs are connected by routers that run an exterior gateway protocol—BGP—in addition to OSPF.

OSPF

AS1

OSPF

OSPF/ BGP

Exterior

BGP

connection

OSPF/ BGP

AS2

OSPF

IP00025A

Figure 1-2. BGP Connecting Two Autonomous Systems Running OSPF

1-4

308628-15.0 Rev 00

Exterior Gateway Protocols (BGP and EGP)

Nortel Networks supports BGP-3 and BGP-4:

• BGP-3 assumes that eac h adver tised networ k is a n atura l clas s netwo rk (A , B,

or C), based on its high-order bits. BGP-3 cannot advertise subnets or supernets.

• BGP-4 has no concept of address classes. Each network listed in the network

layer reachability information (NLRI) portion of an update message contains a prefix length field, which describes the length of the mask associated with the network. The prefix length field allows for both supernet and subnet advertisement. The supernet advertisement is what makes classless interdomain routing (CIDR) possible. See “Class le ss Int erdomain Routing” on page 1-4.

In addition, BGP-4 supports BGP confederations and TCP MD5 message authentication.

This sect i on covers the following topics:

Topic Page

Peer-to-Peer Sessions 1-5 Stub and Multihomed Autonomous Systems 1-6 Interior BGP Routing 1-6 IBGP Route Reflector 1-7 Equal-Cost Multipath 1-8 BGP Updates 1-8 Path Attributes 1-9 BGP/OSPF Interaction 1-10 BGP-4 Confederations 1-11 BGP-4 TCP MD5 Message Authentication 1-11

Peer-to-Peer Sessions

A BGP router employs a BGP speaker, which is an entity within the router that transmits and receives BGP messages and acts upon them. A BGP speaker forms a neighbor relationship with another BGP speaker by establishing a peer-to-peer session. See Chapter 4, “

308628-15.0 Rev 00

Configuring BGP Peers.”

1-5

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Stub and Multihomed Autonomous Systems

An AS can include one or more BGP speakers t hat e st abl is h peer-to-peer ses si ons with BGP speakers in other autonomous systems to provide external route information for the networks within the A S. A multihomed AS has multiple BGP speakers. A stub AS has a single BGP speaker that establishes a peer-to-peer session with one external BGP speaker. The BGP speaker provides external route information only for the networks contained within its own AS.

Interior BGP Routing

Nortel Networks implements interior BGP (IBGP) intra-AS routing. With IBGP, each router in the AS runs an interior gateway protocol (IGP), such as OSPF, for internal routing updates and also maintains an IBGP connection to each BGP border router. The IBGP information, along with the IGP route to the originating BGP border router, determines the next hop to use for external networks.

Some IGPs carry no BGP information. However, an OSPF type 5 LSA can carry BGP-specific information in its tag field. Each router uses IBGP exclusively to determine reachability to external networks. When an IBGP update for a network is received, it is passed to IP for inclusio n in the routing table only if a viable IGP route to the correct border gateway is available.

1-6

An AS with more than one BGP spe aker c an use I BGP to p rovide a tra nsit s ervic e for networks outside the AS. An AS that provides this service is called a transit AS (Figure 1-3

308628-15.0 Rev 00

Exterior Gateway Protocols (BGP and EGP)

AS 10

AS 20

AS 50

Figure 1-3. Transit AS

BGP A

IGP

BGP C

AS 30

AS 11

AS 12

BGP B

IP0021A

In Figure 1-3, AS 20 is the transit AS. It provides information about its internal networks, as well as transit networks, to the remaining ASs. The IBGP connections between BGP routers A, B, and C provide consistent routing information to the ASs.

IBGP Route Reflector

A BGP router configured for IBGP establishes a peer-to-peer session with every other IBGP speaker in the AS. In an AS with a large number of IBGP speakers, this full-mesh topology can result in high bandwidth and maintenance costs. For example, a full-mesh topology for an AS with 50 IBGP speakers requires 1225 internal peer-to-peer connections.

308628-15.0 Rev 00

1-7

Configuring IP Exterior Gateway Protocols (BGP and EGP)

T o avoid the high cost s of a full-me sh topology to support IBGP speakers wit hin a large AS, you can configure a router to function as an IBGP route reflector (RR). An IBGP speaker that needs to communicate with other BGP speakers in the AS establishes a single peer-to-peer RR client session with the IBGP route reflector.

For information about the IBGP route reflector, see Chapter 6, “Configuring a

Route Reflector.”

Equal-Cost Multipath

Equal-cost multipath (ECMP) support allows a BGP speaker to perform route or traffic balancing within an AS by using multiple equal-cost routes submitted to the routing table by OSPF, RIP, or static routes. For instructions on configuring route and traffic balancing, see Chapter 7, “

Balancing.” For more information ab out equal- cost multi path, see Configuring IP,

ARP, RARP, RIP, and OSPF Services.

BGP Updates

BGP-3 and BGP-4 speakers exchange routing updates that include a network number and a list of autonomous systems that the routing information has passed through (the AS path) as well as a list of unreachable networks. In addition, an update includes the following:

Configuring Route and Traffic

1-8

• List of path attributes

• Local preference value—BGP-4 only. (See “

Preference Calculation and Route Selection” on page 3-25.)

Disabling BayRS Local

308628-15.0 Rev 00

Path Attributes

A BGP-3 update message has a variable-length sequence of path attributes. Each attribute inclu des an attr ibute valu e and an attri bute descr iption. Table 1-1 mandatory and optional BGP-3 path attributes.

Table 1-1. BGP-3 Path Attributes

Attribute Description

AS path Mandatory attribute containing a list of the ASs that must be traversed

Origin Mandatory attribute containing one of the following values:

Next hop Mandatory attribut e that speci fies the IP a ddre ss of the rout er to u se as

Inter-AS Optional attribute used to choose between paths to the destinations

Unreachable Discretionary attribute used to indicate destinations that have become

Exterior Gateway Protocols (BGP and EGP)

lists the

to reach th e given destinations

• IGP (the path is valid all the way to the IGP of the originating AS)

• EGP (the path was advertised using EGP by the last AS in the AS

path)

• Incomplete (the path is valid only to the last AS in the AS path)

a next hop for the advertised destinations

listed

unreachable

A BGP-4 update message has a variable-length sequence of path attributes. Each attribute include s an attri bute valu e and an att ribute d escripti on. Table 1-2 mandatory BGP-4 path attributes.

Table 1-2. BGP-4 Mandatory Path Attributes

308628-15.0 Rev 00

lists the

Attribute Description

AS path Contains a list of the ASs that must be traversed to reach the given

destinations

Origin Contains one of the following values:

• IGP (the path is valid all the way to the IGP of the originating AS)

• EGP (the path was advertised using an EGP by the last AS in the

AS path)

• Incomplete (the path is valid only to the last AS in the AS path)

Next hop Specifies the IP address of the router to use as a next hop for the

advertised destinations

1-9

Configuring IP Exterior Gateway Protocols (BGP and EGP)

In addition, the BGP-4 update message can include the optional path attributes listed in Table 1-3

Table 1-3. BGP-4 Optional Path Attributes

Attribute Description

Multiexit discriminator Chooses between paths to the destinations listed Local preference Allows AS border routers to indicate the preference

Atomic aggregate Ensures that certain network layer reachability

Aggregator Identifies which AS performed the most recent route

Cluster list Lists the members of a route reflector cluster Originator ID Identifies the originator of the route into a route

BGP community Identifies the communities to which the route

they assigned to a chosen route when advertising it to IBGP peers

information (NLRI) is not deaggregated

aggregation. This attribute contains the last AS number that formed the aggregate route followed by the IP address of the BGP speaker that for med the aggregate route.

reflector cluster

belongs. (A comm unity is a group of destinat ions that share some common property.)

BGP/OSPF Interaction

RFC 1745 defines the in ter ac ti on b et we en BGP and OSPF whe n OSPF is the IGP within an autonomous system. For routers running both protocols, the OSPF router ID and the BGP ID must be the same IP address. A BGP route policy must be configured to allow BGP advertisement of OSPF routes.

Interaction between BGP-4 and OSPF includes the ability to advertise supernets to support classless interdomain routing (CIDR). BGP-4 allows interdomain supernet advertisements; OSPF can carry supernet advertisements within a routing domain.

1-10

308628-15.0 Rev 00

BGP-4 Confederations

The BGP confederation feature can reduce the size and complexity of an IBGP mesh by breaking large autonomous systems into a confederation of smaller subautonomous systems. This division reduces the size of IBGP meshes and the complexity of the associated configuration management. Other autonomous systems view the c onfede rati on as a sing le AS wi th the co nfeder atio n ID as its AS number. BGP confederations are available only with BGP-4.

The BGP-4 confederation feature complies with RFC 1965 and provides the following functions:

• Lets you configure a confederation ID on the router

• Implements new AS_PATH segment types

• Lets you configure new AS_PATH variables, AS_CONFED_SET and

AS_CONFED_SEQUENCE, for specifying confederation para mete rs

• Implements correct AS_PATH setting and manipulation to neighboring

autonomous systems that are within and outside the confederation

Exterior Gateway Protocols (BGP and EGP)

Configuring BGP-4 Confederations” on page 3-21 for a detailed description

See “ of this feature and for configuration information.

BGP-4 TCP MD5 Message Authentication

BGP-4 lets you configure the authentication of BGP messages by TCP MD5 signatures, in compliance with RFC 2385, “Protection of BGP Sessions via the TCP MD5 Signature Option.” When BGP authentication is enabled, a BGP speaker can verify that the BGP messages it receives from its peers are actually from a peer and not from a third party masquerading as a peer.

See “

Configuring and Enabling MD5 Authentication” on page 4-22 for a detailed

description of this feature and for configuration information.

308628-15.0 Rev 00

1-11

Configuring IP Exterior Gateway Protocols (BGP and EGP)

BGP Implementation Notes

The following guidelines are crucial to successful BGP configuration.

Caution:

If you do not follow these guidelines, BGP either will not work efficiently or will become disabled on the interfaces involved.

• BGP will not operate with an IP router in nonforwarding (host-only) mode.

Make sure that the routers yo u want BGP to o perate with are in forwarding mode. For instructions on setting the forwarding mode, see Configuring IP, ARP, RARP, RIP, and OSPF Services.

• If you are using BGP for a multihomed AS (one that contains more than one

exit point), Nortel Networks strongly encourages you to use OSPF for your IGP and BGP for your sole exterior gateway protocol, or use intra-AS IBGP routing. For information about configuring OSPF, see Configuring IP, ARP, RARP, RIP, and OSPF Services.

• If OSPF is the IGP, you should use the default OSPF tag construction. Using

EGP or modifying the OSPF tags makes network administration and proper construction of BGP path attributes more difficult.

• For any router supporting both BGP and OSPF, the OSPF router ID and the

BGP identifier must be set to the same IP address.

• For BGP to run as a soloist, Internet service provider (ISP) mode must be

enabled. For instructions on enabling ISP mode, see Configuring IP, ARP, RARP, RIP, and OSPF Services.

1-12

• In configurations where BGP speakers reside on routers that have multiple

network connections over multiple IP interfaces (the typical case for IBGP speakers), consider using the address of the router’s circuitless (virtual) IP interface as the local peer address. In this way, you ensure that BGP is reachable as long as t here i s an activ e circ uit on t he rou ter. For instructions on configuring the circuitless (or virtual ) IP interface, s ee Configuring IP, ARP, RARP, RIP, and OSPF Services.

• By default, an external BGP speaker will neither advertise any routes to a

peer, nor inject any routes into its IGP. Configure route policies to enable any route advertisement. For instructions on configuring policies, see Chapter 5,

“Configuring BGP Accept and Announce Policies.”

308628-15.0 Rev 00

Exterior Gateway Protocols (BGP and EGP)

• Coordinate routing policies among all BGP speakers within an AS so that

every BGP border router within an AS constructs the same path attributes for an external path.

• Configure accept and announce policies on all IBGP connections to accept

and propagate all ro utes. Mak e c onsist ent r outin g poli cy decis ions o n e xt ernal BGP connections.

• To configure BGP and download full Internet routes on the Passport* 5430

Multiservice Access Switch, you must install the router with 64 MB of memory.

308628-15.0 Rev 00

1-13

Chapter 2

Starting BGP and EGP Services

This chapter describes how to use the BCC and Site Manager to start BGP services using default values and how to use Site Manager to start EGP services using default values. It also describes how to use Sit e Manager to delete BGP and EGP services.

Topic Page

Starting BGP with the BCC Starting IP and BGP with Site Manager 2-3 Starting IP and EGP with Site Manager 2-6

Starting BGP with the BCC

To start BGP using the BCC:

1. Configure BGP on the router.

2. Define a BGP peer-to-peer connection.

Note:

Before you configure BGP-4, see “BGP Implementation Notes” on

page 1-12 for information.

Step 1: Configuring Global BGP

To configure BGP on the r out er, go to the global IP pr ompt ( for e xample, and enter:

bgp router-id

<router_id>

2-1

box; ip

)

308628-15.0 Rev 00

2-1

Configuring IP Exterior Gateway Protocols (BGP and EGP)

router_id

is the BGP router ID e xpressed as an oct et stri ng. The ro uter ID typicall y is a circuitless IP interface used by BGP to communicate wi th other BGP routers. If the router is also running OSPF, the BGP ID must match the OSPF ID.

For example, the following command configures global BGP with a router ID (local IP address) of 2.2.2.2:

ip#

bgp router-id 2.2.2.2

bgp#

BGP is now running on t he router with de fault va lues for al l BGP parameters. You customize BGP by modifying BGP parameters as described in Chapter 3,

“Configuring Global BGP Parameters.”

Step 2: Defining a Peer-to-Peer Connection

For BGP to exchange routing information with BGP peers located in other autonomous systems or within the same AS, you must configure at least one peer connection.

To define a peer-to-p eer connection, go to the BGP p rompt (f or example,

) and enter:

bgp

<local_ip_address> remote <remote_ip_address>

peer local

<as_number>

box; ip;

2-2

local_ip_address

is the address, expressed in dotted-decimal format, of an IP

interface on the local router.

remote_ip_address as_number

is the number of the AS in which the remote peer is located.

is the address of an IP interface on the remote peer router.

For example, the following command defines a peer-to-peer connection between local IP interface 2.3.3.3 and remote interface 2.3.3.4. The remote BGP peer is located in AS 4.

bgp#

peer local 2.3.3.3 remote 2.3.3.4 as 4

peer/2.3.3.3/2.3.3.4#

The BGP pee r-t o-peer relationship is established with default values for all BGP peer parameters. You customize the peer-to-peer connection by modifying BGP peer parame ters as described in Chapter 4, “

Configuring B GP Peers.”

308628-15.0 Rev 00

Starting IP and BGP with Site Manager

Before you can select a protocol to run on the router, you must configure a circuit that the protocol can use as an interface to an attached network. For information and instructions, see Configuring WAN Line Services and Configuring Ethernet, FDDI, and Token Ring Services.

After you configure the circuit, you can access the Site Manager Select Protocols window.

Note:

Before you configure BGP-4, see “BGP Implementation Notes” on

page 1-12 for information.

To start IP and BGP from the Select Protocols window, complete the following steps:

Site Manager Procedure

Starting BGP and EGP Services

You do this System responds

1. In the Select Protocols window, select the following protocols:

•IP

•

Then click on OK.

2. Set the following parameters:

•

Click on descriptions in

RIP, and OSPF Services

3. Click on OK. The BGP Configuration window opens.

4. Set the following parameters:

•

Click on descriptions on page A-3.

5. Click on OK. The BGP Peer window opens.

BGP

IP Address Subnet Mask Transmit Bcast Addr UnNumbered Assoc Address

or see the parameter

Help

Configuring IP, ARP, RARP,

Identifier Local AS

or see the parameter

Help

The IP Configuration window opens.

(continued)

308628-15.0 Rev 00

2-3

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Site Manager Procedure

You do this System responds

6. Set the following parameters:

•

Pee r Address

•

Pee r AS

•

Local Address

•Peer Mode

Click on descriptions beginning on page A-11.

7. Click on OK. Site Manager enables default BGP

Help

or see the parameter

Deleting BGP with Site Manager

You can delete BGP from all router circuits on which it is currently enabled. To d elete BGP using Site Manager, comple te the following steps:

You do this System responds

(continued)

service and returns you to the Configuration Manager window.

Site Manager Procedure

2-4

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Click on OK. Site Manager removes BGP from all circuits

Protocols

. The IP menu opens.

BGP Delete BGP

. The BGP menu opens.

. Site Manager asks you to confirm the

The Protocols menu opens.

deletion of BGP.

on the router and returns you to the Configuration Manager window.

308628-15.0 Rev 00

Deleting BGP-3 and BGP-4 with Site Manager

You can delete BGP-3 and BGP-4 from all router circuits on which they are currently enabled.

To delete BGP-3 using Site Manager, complete the following steps:

Site Manager Procedure

You do this System responds

Starting BGP and EGP Services

1. In the Configuration Manager window , choose

2. Choose

3. Choose

4. Choose

5. Click on OK. Site Manager removes BGP-3 from all

Protocols

. The IP menu opens.

BGP Delete BGP-3

. The BGP menu opens.

. Site Manager asks you to confirm the

The Protocols menu opens.

deletion of BGP-3.

circuits on the router and returns you to the Configuration Manager window.

To delete BGP-4 using Site Manager, complete the following steps:

Site Manager Procedure

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Click on OK. Site Manager removes BGP-4 from all

Protocols

. The IP menu opens.

BGP Delete BGP-4

. The BGP menu opens.

. Site Manager asks you to confirm the

The Protocols menu opens.

deletion of BGP-4.

circuits on the router and returns you to the Configuration Manager window.

308628-15.0 Rev 00

2-5

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Starting IP an d EGP with Site Manager

After you configure the circuit, you can access the Site Manager Select Protocols window.

To start IP and EGP from the Select Protocols window, complete the following steps:

Site Manager Procedure

You do this System responds

1. In the Select Protocols window, select the following protocols:

•IP

•

Then click on OK.

2. Set the following parameters:

•

Click on description in

RIP, and OSPF Services

3. Click on OK. The EGP Configuration window opens.

4. Set the following parameters:

•

Click on descriptions beginning on page A-25.

5. Click on OK. Site Manager enables EGP service and

EGP

IP Address Subnet Mask Transmit Bcast Addr UnNumbered Assoc Address

or see the parameter

Help

Configuring IP, ARP, RARP,

Local Autonomous System ID (decimal) Remote Peer IP Address Gateway Mode

or see the parameter

Help

The IP Configuration window opens.

returns you to the Config ur ation Manag er window.

2-6

308628-15.0 Rev 00

The instructions in this chapter show you how to start EGP using default values. For information about modifying EGP default values, see Chapter 8,

“Customizing EGP Services.”

Deleting EGP from the Router

You can delete EGP from all router circuits on which it is currently enabled. To delete EGP, complete the following steps:

You do this System responds

Starting BGP and EGP Services

Site Manager Procedure

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Click on OK. Site Manager removes EGP from all circuits

Protocols

. The IP menu opens.

EGP Delete EGP

. The EGP menu opens.

. Si te Manager asks you to confir m the

The Protocols menu opens.

deletion of EGP.

on the router and returns you to the Configuration Manager window.

308628-15.0 Rev 00

2-7

Chapter 3

Configuring Global BGP Parameters

You customize global BGP parameters for your AS as described under the following topics:

Topic Page

Disabling and Reenabling BGP 3-2 Supplying a BGP Router ID 3-4 Identifying the Local AS 3-5 Disabling and Reenabling IBGP Support 3-6 Specifying Route Types for IBGP Advertisements 3-7 Enabling BGP Interaction with OSPF and RIP 3-9 Setting the Update Interval Timer 3-10 Allowing Redundant Conne cti ons 3-11 Enabling Multihop Connections 3-13 Disabling and Reenabling Dynamic Policy Configuration 3-15 Configuring the BGP Soloist Slot Mask 3-16 Disabling and Reenabling Route Aggregation 3-17 Enabling and Disabling Black Hole Punching 3-18 Disabling and Reenabling the BGP-4 MED Attribute 3-20 Configuring BGP-4 Confederations 3-21 Disabling BayRS Local Preference Calculation and Route Selection 3-25 Configuring BGP Message Logging 3-29 Configuring EBGP Route Flap Damping 3-31 Assigning Weight Classes and Values to an AS 3-37

308628-15.0 Rev 00

3-1

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Disabling and Reenabling BGP

When you start BGP on the router, BGP is automatically enabled for both BGP-3 and BGP-4 peer-to-peer connections.

You can use the BCC and Site Manager to disable and reenable BGP-4 on the router. You can also use Site Manager to disable and reenable BGP-3 and BGP-4.

BGP will not operate with an IP router in not-forwarding (host-only)

Note:

mode. Make sure that the routers you want BGP to operate with are in forwarding mode. mode, see Configuring IP, ARP, RARP, RIP, and OSPF Services.

Using the BCC

For instructions on configuring the router for forwarding

To disable and re enable BGP, go to the BGP prompt (for example, and enter:

state

is either

state

For example, the following command disables BGP:

bgp#

Using Site Manager

To disable and reenable BGP, complete the following steps:

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

<state>

enabled

state disabled

Protocols IP

. The IP menu opens.

BGP

. The BGP menu opens.

BGP Global

box; ip; bgp

(default) or

. The Edit BGP Global Parameters window

disabled

Site Manager Procedure

The Protocols menu opens.

opens.

(continued)

)

3-2

308628-15.0 Rev 00

Configuring Global BGP Parameters

Site Manager Procedure

You do this System responds

5. Set the

Help

page A-4

6. Click on OK. You return to the Configuration Manager

7. In the Configuration Manager window, choose

8. Choose

9. Choose

10.Choose

11. Set the or see the pa rameter description on page

A-11

12. Click on OK. You return to the Configuration Manager

13.In the Configuration Manager window, choose

14.Choose

15.Choose

16.Choose

17. Set the or see the pa rameter description on page

A-11

18. Click on OK. You return to the Configuration Manager

BGP Enable

or see the parameter description on

Protocols

. The IP menu opens.

. The BGP menu opens.

BGP BGP-3 Global

Enable

Protocols

. The IP menu opens.

. The BGP menu opens.

BGP BGP-4 Global

Enable

parameter . Clic k on

. The Edit BGP-3 Global Parameters m enu

parameter . Click on

. The Edit BGP-4 Global Parameters

parameter . Click on

Help

(continued)

window. The Protocols menu opens.

opens.

window. The Protocols menu opens.

window opens.

window.

308628-15.0 Rev 00

3-3

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Supplying a BGP Router ID

The BGP identifier is the IP address of an interface on the router. When you enable BGP on the router, you must specify a configured IP address as the router ID. You can change the router ID to the IP address of another router IP interface.

If both OSPF and BGP are running on the router, the OSPF router ID

Note:

and the BGP router ID must be identical . In addition, the ro uter ID must match one of the IP addresses configured on the router.

You can use the BCC or Site Manager to supply a BGP identifier for the router.

Using the BCC

To change the BGP router ID, go to the BGP prompt (for example, and enter:

router-id

ip_address

For example, the following command specifies IP address 2.2.2.2 for the BGP router ID:

bgp# bgp#

Using Site Manager

To change the BGP router ID, complete the following steps:

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

<ip_address>

is the address of an IP interface on the router.

router-id 2.2.2.2

Site Manager Procedure

The Protocols menu opens.

Protocols IP

. The IP menu opens.

BGP BGP Global

. The BGP menu opens.

. The Edit BGP Global Parameters window

opens.

box; ip; bgp

(continued)

)

3-4

308628-15.0 Rev 00

Configuring Global BGP Parameters

You do this System responds

5. Set the on on page A-4

6. Click on OK. You return to the Configuration Manager

BGP Identifier

Help

or see the parameter description

Identifying the Local AS

Each autonomous system in the Internet has a unique AS ID. You can use the BCC or Site Manager to supply the ID of the AS in which the BGP router is located.

Using the BCC

To specify the ID of the local AS, go to the BGP prompt (for example,

bgp

) and enter:

local-as

local_as

<local_as>

is the number of the AS (fr om 0 t h r ough 65535 ) where the router resi de s.

Site Manager Procedure

parameter. Click

window.

(continued)

box; ip;

For example, the following command specifies AS 5 as the local AS:

bgp# bgp#

Using Site Manager

To specify the ID of the local AS, complete the following steps:

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

308628-15.0 Rev 00

local-as 5

Protocols IP

. The IP menu opens.

BGP

Site Manager Procedure

The Protocols menu opens.

. The BGP menu opens.

(continued)

3-5

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Site Manager Procedure

You do this System responds

4. Choose

5. Set the on on page A-4

6. Click on OK. You return to the Configuration Manager

BGP Global

BGP Local AS

Help

or see the parameter description

. The Edit BGP Global Parameters window

parameter . Click

opens.

window.

Disabling and Reenabling IBGP Support

By default, BGP supports IBGP intra-AS sessions. (For information, see “BGP Concepts and Terminology” on page 1-4).

A BGP transit AS should use IBGP intra-AS routing. A stub or multihomed AS usually does not use IBGP routing.

You can use the BCC or Site Manager to disable and reenable IBGP support.

Using the BCC

(continued)

3-6

To disable and reenable IBGP support, go to the BGP prompt (for example,

ip; bgp intra-as-routing

state

) and enter:

is either

<state>

enabled

(default) or

disabled

For example, the following command disables IBGP:

intra-as-routing disabled

bgp# bgp#

308628-15.0 Rev 00

box;

Using Site Manager

To disable and reenable IBGP support, complete the following steps:

You do this System responds

Configuring Global BGP Parameters

Site Manager Procedure

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Set the

Help

page A-5

6. Click on OK. You return to the Configuration Manager

Protocols IP

. The IP menu opens.

BGP BGP Global

BGP Intra-AS

or see the parameter description on

. The BGP menu opens.

. The Edit BGP Global Parameters window

parameter . C lick o n

The Protocols menu opens.

opens.

window.

Specifying Route Types for IBGP Advertise m ents

If IBGP is enabled, you can specify the types of routes that BGP advertises in IBGP sessions. By default, IBGP propagates only routes learned from external BGP peers. You can use the BCC o r Sit e Ma nager to c onfigu re I BGP to prop agate routes learned from all route sources (excluding IBGP and OSPF interarea and intra-area routes, which IBGP never advertises).

308628-15.0 Rev 00

3-7

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Using the BCC

To specify the types of routes that IBGP advertises, go to the BGP prompt (for example,

box; ip; bgp

redistribute-protocols

) and enter:

protocols>

protocols

T a ble 3-1. Route Types for BGP Advertisements

Route Type Meaning

bgp (default) BGP propagates routes learned from external BGP peers. all BGP propagates routes from all route sources.

For example, the followi ng command c onfigu res BGP to adve rtis e rout es fr om all route sources:

bgp# bgp#

Using Site Manager

T o spec if y the types of rou tes that BGP advertis es in IBGP sessi ons, comp lete the following steps:

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Set the Click on description on page A-5

6. Click on OK. You return to the Configuration Manager

is one of the values listed in Table 3-1.

redistribute-protocols all

Site Manager Procedure

Protocols

. The IP menu opens.

BGP BGP Global

BGP From Protocols

Help

. The BGP menu opens.

. The Edit BGP Global Parameters window

parameter.

or see the parameter

The Protocols menu opens.

opens.

window.

3-8

308628-15.0 Rev 00

Configuring Global BGP Parameters

Enabling BGP Interaction with OSPF and RIP

By default, BGP does not exchange routes with OSPF or RIP. However, you can configure the r outer to allow BGP to advertise BGP-learned route s to OSPF, RIP, or both OSPF and RIP.

You can use the BCC or Site Manager to configure the router to advertise BGP-learned routes.

Using the BCC

To configure the router to advertise BGP-learned information to OSPF, RIP, or both, go to the BGP prompt (for example,

box; ip; bgp

) and enter:

igp-interaction

protocols

none

•

ospf

•

rip

•

For example, the following command configures BGP to advertise routes to both RIP and OSPF:

bgp# bgp#

Using Site Manager

To configure the router to advertise BGP-learned information to OSPF, RIP, or both,

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

is one of the following values:

(default): BGP adve rtises no routes to OSPF or RIP.

: BGP advertises BGP-learned routes to OSPF.

: BGP advertis es BGP-learned routes to RIP.

igp-interaction {ospf rip}

complete th e following st eps:

Site Manager Procedure

The Protocols menu opens.

Protocols

. The IP menu opens.

BGP

. The BGP menu opens.

(continued)

308628-15.0 Rev 00

3-9

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Site Manager Procedure

You do this System responds

4. Choose

5. Set the parameter. Click on parameter description on page A-10

6. Click on OK. You return to the Configuration Manager

BGP Global

BGP/IGP Interaction Control

. The Edit BGP Global Parameters window

Help

or see the

Setting the Update Interval Timer

BGP periodically injects external BGP routes into the routing table. The default minimum interval between route injections is 5 seconds.

You can use the BCC or Site Manager to change the minimum number of seco nds between route injections.

Using the BCC

To change the minimum number of seconds between route injections, go to the BGP prompt (for example,

box; ip; bgp

(continued)

opens.

window.

) and enter:

3-10

inject-time

seconds

seconds>

is the minimum interval (from 1 through 2,147,483,647) between route

injections. For example, the following command causes BGP to inject external BGP routes

into the routing table with a minimum interval of 10 seconds:

bgp#

inject-time 10

bgp#

308628-15.0 Rev 00

Using Site Manager

To change the minimum number of seconds between route injections, complete the following steps:

You do this System responds

Configuring Global BGP Parameters

Site Manager Procedure

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Set the Click on description on page A-6

6. Click on OK. You return to the Configuration Manager

Protocols IP

. The IP menu opens.

BGP BGP Global

BGP Interval Timer

Help

. The BGP menu opens.

. The Edit BGP Global Parameters window

parameter.

or see the parameter

Allowing Redundant Connections

By default, BGP performs redundancy checking on peer-to-peer TCP sessions. BGP can maintain only one TCP session with a remote BGP peer. If the remote peer attempts to establish another session on another physical connection, BGP rejects the session. BGP uses a collision-detection method based on the router ID to check for redundant sessions.

You can disable redundancy che cking to allo w TCP sessions with the same remote peer on multiple physical connections. configuration with multiple sessions on multiple physical connections is redundancy: if on e conne ction fail s, the peers can co mmunica te ove r anot her l ink. The disadvantage is that such a configuration results in multiple copies of each route.

The Protocols menu opens.

opens.

window.

The advantage of a peer-to-peer

You can use the BCC or Site Manager to disable redundancy checking and to specify the maximum number of redundant routes that BGP allows. By default, BGP allows up to 255 redundant routes.

308628-15.0 Rev 00

3-11

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Using the BCC

To disable or reenable redundancy checking, go to the BGP prompt (for example,

box; ip; bgp redundant-connection

state

is either

) and enter:

enabled

state>

(default) or

disabled

To specify the maximum number of redundant routes, go to the BGP prompt and enter:

max_routes>

max-redundant-routes

max_routes

For example, the following command sequence disables BGP redundancy checking, allowing BGP t o es ta bli sh mul ti pl e TCP s essions (on differe nt physical connections) with the same remot e pee r and configures BGP to maintain up to 50 redundant routes:

bgp# bgp# bgp#

The following command reenables redundancy checking to allow only one TCP session with the same remote peer:

bgp# bgp#

Using Site Manager

To d i sable or reenable redunda ncy checking, com plete the following steps:

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

is the maximum number of redundant routes (from 0 through 255).

redundant-connection disabled max-redundant-routes 50

redundant connection enabled

Site Manager Procedure

The Protocols menu opens.

Protocols

. The IP menu opens.

BGP

. The BGP menu opens.

(continued)

3-12

308628-15.0 Rev 00

Configuring Global BGP Parameters

Site Manager Procedure

You do this System responds

4. Choose

5. Set the parameter. Click on parameter description on page A-7

6. Click on OK. You return to the Configuration Manager

BGP Global

Detect Redundant Connections

. The Edit BGP Global Parameters window

Help

or see the

Enabling Multihop Connections

By default, BGP enforces the one-hop rule for BGP peers (the remote peer must be located on a directly attached network.)

You can use the BCC or Site Manager to override the restriction and allow multihop connections.

Caution:

establish a BGP connection tha t t rav erses a third-party AS, which may viola te policy considerations and may also introduce forwarding loops.

Enabling multihop BGP connections can cause EBGP speakers to

(continued)

opens.

window.

Using the BCC

To override the one-hop rule and allow multihop connections, go to the BGP prompt (for example,

multi-hop

state

For example, the following command enables BGP for multihop peer connections:

bgp# bgp#

308628-15.0 Rev 00

<state>

is either

multi-hop enabled

enabled or disabled

box; ip; bgp

) and enter:

(default).

3-13

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Using Site Manager

To override the one-hop rule and allow multihop connections, complete the following steps:

Site Manager Procedure

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Set the parameter. Click on parameter description on page A-7

6. Click on OK. You return to the Configuration Manager

Protocols

. The IP menu opens.

BGP BGP Global

Multi-hop Ebgp Connection

. The BGP menu opens.

. The Edit BGP Global Parameters window

or see the

Help

The Protocols menu opens.

opens.

window.

3-14

308628-15.0 Rev 00

Configuring Global BGP Parameters

Disabling and Reenabling Dynamic Policy Configuration

By default, BGP reconfigures IP policies dynamically. This means that if you modify a policy, BGP dynamically reevaluates all affected routes in light of the modified policy. BGP then sends the appropriate withdraw or update message to the affected peers. BGP keeps track of the routes sent to each peer, allowing for precise determination of which routes to send and which to withdraw.

If you modify an IP policy with this feature disabled, BGP restarts all BGP connections.

Note:

There is no advantage to disabling dynamic policy configuration. Disabling this parameter will adversely affect BGP protocol operation overhead and network stability.

You can use the following Site Manager procedure to disable and reenable dynamic policy configuration:

Site Manager Procedure

308628-15.0 Rev 00

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Set the

Support

the parameter description on page A-7

6. Click on OK. You return to the Configuration Manager

Protocols IP

. The IP menu opens.

BGP BGP Global

BGP Dynamic Policy Change

parameter. Click on

. The BGP menu opens.

. The Edit BGP Global Parameters window

Help

or see

The Protocols menu opens.

opens.

window.

3-15

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Configuring the BGP Soloist Slot Mask

By default, BGP runs as a soloist on a slot determined by the BGP soloist slot mask. Nortel Networks recommends that the slot mask include only nonforwarding slots, so that BGP operations (route calculation, for example) occur on one slot while the other slots maintain maximum forwarding capability.

If the slot on which the soloist is running fails, BGP runs on an eligible slot. By default, BGP considers all sl ots with IP inter faces to be eli gible slots . You can use the BCC or Site Manager to specify one or more slots for the BGP soloist.

Using the BCC

To configure the BGP slot mask, go to the BGP prompt (for example,

bgp

) and enter:

slot-mask

is one of the values listed in Table 3-2. To include mor e than one sl ot, encl ose

slot

slot>

box; ip;

the slot numbers in braces.

T able 3-2. Slot Mask Parameter Values

Value Meaning

all-slots (default) The BGP soloist can run on all slots. 1 to 14 The BGP soloist can run only on the specified slots.

For example, the following co mman d conf igu re s slo ts 2, 3, and 4 to be eligible to run the BGP soloist:

bgp#

slot-mask {2 3 4}

bgp#

3-16

308628-15.0 Rev 00

Using Site Manager

To configure the BGP slot mask, c omplete the following steps:

You do this System responds

Configuring Global BGP Parameters

Site Manager Procedure

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Set the Click on description on page A-7

6. Click on OK. You return to the Configuration Manager

Protocols IP

. The IP menu opens.

BGP BGP Global

BGP Soloist Slots

Help

. The BGP menu opens.

. The Edit BGP Global Parameters window

parameter.

or see the parameter

The Protocols menu opens.

opens.

window.

Disabling and Reenabling Rou te Aggregation

By default, BGP aggregates non-BGP-originated subnet routes to their corresponding natural network routes for advertisement to BGP peers. Disabling route aggregation causes BGP to advertise each subnet.

You can use the BCC or Site Manager to disable this feature. (This switch does not affect the advertisement of BGP-originated routes.)

Using the BCC

T o dis able or reenabl e route aggre gatio n for non-BGP-origin ated subnet routes, go to the BGP prompt (for example,

subnet-aggregation

state

For example, the following command disables subnet aggregation:

bgp# bgp#

308628-15.0 Rev 00

state>

is either

enabled

(default) or

subnet-aggregation disabled

box; ip; bgp

disabled

) and enter:

3-17

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Using Site Manager

To disable or reenable route aggregation, complete the following steps:

Site Manager Procedure

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Set the Click on description on page A-8

6. Click on OK. You return to the Configuration Manager

Protocols IP

. The IP menu opens.

BGP BGP Global

Aggregate Subnets

Help

. The BGP menu opens.

. The Edit BGP Global Parameters window

parameter.

or see the parameter

The Protocols menu opens.

opens.

window.

Enabling and Disabling Black Hole Punching

If BGP advertises aggregate routes, you can configure BGP to submit each aggregate route to the routing table as a to drop a packet whose lo ngest matchi ng destina tion pref ix is the black hole route. (For more information about black hole routes, see

RIP, and OSPF Services

By default, BGP does not submit a black hole route to the IP routing table for an aggregate route that it advertises to a BGP peer.

black hole. This setting forces the rou ter

Configuring IP, ARP, RARP,

3-18

You can use the BCC or Site Manager t o enable black h ole punch ing. You can also configure IP to return an ICMP dest ination unreac hable message to the se nder of a packet that best matches the black hole route.

308628-15.0 Rev 00

Using the BCC

Configuring Global BGP Parameters

To enable or disable black hole punching, go to the BGP prompt (for example,

box; ip; bgp black-hole-punching

action

Table 3-3. Black Hole Punchin g Parameter Settings

Value Meaning

disabled (default) Disables black hole punching drop Enables black hole punching. IP drops packets for the black

reject Enables black hole punching. IP drops packets for the black

) and enter:

action>

is one of the values described in Table 3-3

hole route destination without returning an ICMP message to the sender.

hole route destination and returns an ICMP destination unreachable message to the sender.

For example, the following command causes BGP to submit aggregate routes to the routing table as bla ck hole routes. IP dro ps packet s for the blac k hole route but does not return ICMP destination unreachable messages to the sender.

bgp# bgp#

308628-15.0 Rev 00

black-hole-punching drop

3-19

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Using Site Manager

To enable or disable black hole punching, complete the following steps:

Site Manager Procedure

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Set the Click on description on page A-8

6. Click on OK. You return to the Configuration Manager

Protocols IP

. The IP menu opens.

BGP BGP Global

Black Hole Routes

Help

. The BGP menu opens.

. The Edit BGP Global Parameters window

parameter.

or see the parameter

The Protocols menu opens.

opens.

window.

Disabling and Reenabling th e BGP-4 MED Attribute

By default, BGP-4 considers the multiex it discriminator (MED) path attribute in the route selection process (see Table 1-3 on page 1-10

You can use the BCC or Site Manager to configure BGP-4 so that it disregard s the MED attribute in the route selection process.

Using the BCC

3-20

To configure BGP-4 to disregard the MED attribute in t he r out e s el ect ion process, go to the BGP prompt (for example,

med-comparison

state

is either

state>

enabled

(default) or

disabled

box; ip; bgp

) and enter:

For example, the following command causes BGP-4 to disregard the MED attribute in an update when selecting a route:

bgp#

med-comparison disabled

bgp#

308628-15.0 Rev 00

Using Site Manager

To configure BGP-4 to disregard the MED attribute in t he r out e s el ect ion process, complete the following steps:

You do this System responds

Configuring Global BGP Parameters

Site Manager Procedure

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Set the on on page A-9

6. Click on OK. You return to the Configuration Manager

Protocols IP

. The IP menu opens.

BGP BGP Global

Route with MED

Help

or see the parameter description

. The BGP menu opens.

. The Edit BGP Global Parameters window

parameter. Click

Configuring BGP-4 Confede rations

If this router is a member of an autonomous system that is a confederation of sub-ASs, you can configure the confederation ID. You can also specify the list of this BGP speaker’s peers in other sub-ASs within the same confederation.

If the confederation ID is not configured (nil), this AS is not a member of any confederation. If the list of confederatio n peers is empty (nil), no peers to this speaker exist among the neighbor ASs that are members of this confederation.

The Protocols menu opens.

opens.

window.

You can use the BCC or Site Manager to configure BGP-4 confederations.

308628-15.0 Rev 00

3-21

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Using the BCC

Figure 3-1

shows an example of a BGP-4 confederation. In this example, the

following steps establish sub-ASs 65000 and 65001 as confederation peers.

Specify sub-AS 65002 as the local AS number and the external visible AS

number 20 as the BGP confederation identifier by entering:

bgp#

local-as 65002

bgp#

confederation-id 20

Specify the local AS numbers of the other sub-ASs within the same

confederation as confederation peers by entering:

bgp#

confederation-peers {65000 65001}

Specify BGP peers within the confederation by entering:

bgp#

peer 192.32.194.1/192.32.194.2 as 65000

peer/192.32.194.1/192.32.194.2# bgp#

peer 192.32.194.5/192.32.194.6 as 65001

peer/192.32.194.5/192.32.194.6# bgp#

peer 192.32.195.2/192.32.195.1 as 65002

peer/192.32.195.2/192.32.195.1# bgp#

peer 192.32.195.5/192.32.195.6 as 65002

peer/192.32.195.5/192.32.195.6# bgp#

Specify BGP peers in another AS or con federation wi th their local AS number

back back back back

(Router 3) (Router 6) (Router 8) (Router 9)

(if not a confederation) or their confederation ID (if a confederation) by entering:

3-22

bgp#

peer 192.32.195.1/192.32.195.2 as 55

peer/192.32.195.1/192.32.195.2# bgp#

A sub-AS number cannot be the same as any external BGP peer AS;

Note:

back

however, internal sub-AS numbers can be assigned to a sub-AS in another confederation. Suggestion: Use the reserved AS number range. (IANA reserved numbers are 64512 through 65535.)

308628-15.0 Rev 00

Configuring Global BGP Parameters

Sub-AS

65001

Router 2

Router 1

Sub-AS

65000

AS 20

Router 3

Router 4

Router 9

Router 6

Router 7

Sub-AS

65002

Router 5

To AS 55

Router 8

IP0107A

Figure 3-1. BGP Confederation

308628-15.0 Rev 00

3-23

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Using Site Manager

To configure a confederation, complete the following steps:

Site Manager Procedure

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Set the following parameters:

•

Click on descriptions beginning on page A-10

6. Click on OK. You return to the Configuration Manager

Protocols

. The IP menu opens.

BGP BGP Global

Confederation ID Confederation Peer

Help

. The BGP menu opens.

. The Edit BGP Global Parameters window

or see the parameter

The Protocols menu opens.

opens.

window.

3-24

308628-15.0 Rev 00

Configuring Global BGP Parameters

Disabling BayRS Local Preference Calculation and Route Selection

A BGP speaker calculates a local preference value for each route that it receives from an external peer and passes this value as the local_pref attribute in routing updates that it announces to its internal BGP (IBGP) peers. A BGP sp eaker that receives a routing update from an IBGP peer uses the local_pref attribute in its best-route selection process.

By default, BGP calculates a local-preference value by using the algorithm described in “ best-route selection process is described in “

Different implementations of BGP use different methods of arriving at a local preference value. However, within an AS, all BGP speakers should use the same method to determine local preference and to select the best routes.

Calculating BGP-4 Local Preference Values” on page 3-25. The

Best-Route Selection” on page 3-27.

Note:

For a comparison of

processes, see “Comparing BGP Route Selection Processes” on page B-11.

If you are configuring BGP in a network that also includes routers from vendors other than Nortel Networks, you can disable BayRS local preference calculation and route selection. Instead, BGP assigns a value of 100 to externally received routes and uses the rules listed in Table 3-4 on page 3-27

the Nortel Networks and Cisco route selection

Calculating BGP-4 Local Preference Values

BGP-4 update messages c ont ain a l ocal preference valu e t hat an AS border router can assign to a route when advertising it to IBGP peers. The calculation of the local preference attribute is specific to each implementation. A higher value indicates that the route is more preferred.

The router uses the following equation t o calcul ate a va lue for t he local p referenc e attribute:

local_preference origin_value

routes.

is 0 for routes with an origin path attribute of IGP or 4096 for other

= 8191 -

origin_value

AS_path_weight

for best-route selection.

308628-15.0 Rev 00

3-25

Configuring IP Exterior Gateway Protocols (BGP and EGP)

AS_path_weight

is a sum of weight values associated with AS numbers listed in the route’s AS Path attribute. These weight values can be configured and default to 8. For information about configuring AS path weights, contact the Nortel Networks Technical Solutions Center.

A steep penalty is applied to routes that are advertised with an origin attribute other than IG P or incomplete .

For an OSPF internal route or a d ir ect route, the loc al preference attri but e is set to:

local_preference

is the OSPF metric for an OSPF route or the configured cost for a direct

metric

= (8191 + 256 - (

metric

+ 255))

route. For a RIP route, an EGP route, an OSPF ASE route, or a static route, the local

preference attribute is set to:

local_preference

is the RIP metric for a RIP route, the EGP metric for an EGP route, the

metric

= (256 -

metric

)

OSPF metric for an OSPF ASE route, or the configured cost for a static route. Local preference valu es f or OSPF int ernal routes and dir ect rou tes ar e higher than

the local preference values calculated for BGP routes.

3-26

308628-15.0 Rev 00

Best-Route Selection

BGP uses the rules in Table 3-4 as tie-breakers to select the best BGP route. This table lists the rules in the order in which they are evaluated.

Table 3-4. Best-Route Selection Rules

Configuring Global BGP Parameters

Default Preference Tiebreaker Rules (local-pref-calculation enabled)

1. Choose the route with the lower route weight.

2. Choose the route with the hig her local preferen ce attribute.

3. Choose the route with the shortest AS path.

4. Choose the route with the smallest ORIGIN (IGP < EGP < INCOMPLETE).

5. Choose the route with the lower multiexit discriminator (MED) attribute if both routes include this optional attribute.

6. Choose the route with the lower interior cost to the next hop.

7. Choose external BGP over IBGP.

8. Choose the route with the lower BGP identifier.

9. Choose the route wit h the lowe r BGP connection remote address.

10.Choose the route with the lowe r BGP connection local address.

Calculated Preference Tiebreaker Rules (local-pref-calculation disabled)

1. If the next hop is inaccessible, do not consider it.

2. Consider larger BGP administrative weights first.

3. If the routes have the same weight, consider the route with the highest local preference .

4. If the routes have the same local preference, prefer the route that the local router originated.

5. If no route was locally originated, prefer the shorter AS path.

6. If the routes have the same AS path length, prefer the lowest origin code (IGP is preferred over EGP, which is preferred over incomplete).

7. If the origin codes are the same and all paths come from the AS, prefe r the path with the lowes t multiexit discriminator (MED) metric. Treat a missing MED metric as 0.

8. If the MED metrics are the same, prefer external paths over internal paths.

9. If IGP synchronization is disabled and only internal paths rema in, p refer the p ath th rough the closest neighbor.

10.Prefer the route with the lowest IP address value for the BGP router ID.

Using the BCC

To configure local preference calculation, go to the BGP prompt (for example,

box; ip; bgp local-pref-calculation

state

308628-15.0 Rev 00

) and enter:

<state>

is one of the values shown in Table 3-5.

3-27

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Table 3-5. Local Preference Calculation Method

State Meaning

enabled (default) BGP uses the decision rules in the first column of Table 3-4 for

local preference calculation and best-route selection.

disabled BGP does not perform local preference calculation. Instead, it

assigns a value of 100 to th e route and use s a modified form ula for best-route selection.

For example, the following command disables BayRS BGP local preference calculation and route selection:

local-pref-calculation disabled

bgp# bgp#

Using Site Manager

To disable or reenable local preference calculation, complete the following steps:

Site Manager Procedure

3-28

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Set the parameter. Click on parameter description on page A-9

6. Click on OK. You return to the Configuration Manager

Protocols IP

. The IP menu opens.

BGP BGP Global

Local Pref Calculation

. The BGP menu opens.

. The Edit BGP Global Parameters window

Help

or see the

The Protocols menu opens.

opens.

window.

308628-15.0 Rev 00

Configuring BGP Message Logging

By default, the router logs only a few event messages generated by BGP. To troubleshoot a problem, you can configure the router to log BGP event messages for the entire router or for a specific peer-to-peer session by specifying:

• Local and remote addresses of a peer-to-peer session or sessions

• Message severity level: fault, warning, information, trace, debug, or all levels

• BGP message type: open, keepalive, update, or notification

When you are not troubleshooting a problem, configure BGP message logging to limit the number of debug messages that BGP generates and logs. If you allow BGP to log all debug messages, the debug messages will overrun and overwrite the log file.

For more information about event messages, see the BayRS event

Note:

message database at http://www25.nortelnetworks.com/library/tpubs/events. (You also can access the event message database from the BayRS documentation CD.)

Configuring Global BGP Parameters

Using the BCC

To configure BGP event message logging, go to the debug-control prompt (for example,

debug-control log-message-type {debug

log-packet-type {disabled

To disable event mes sag e logging, set the (the default value).

308628-15.0 Rev 00

box; ip; bgp; debug-control

) and enter:

info

open

warning

update

box; ip; bgp; peer/2.2.2.2/2.2.2.4;

fault

notification

log-packet-type

trace

all-levels}

keepalive}

parameter to

disabled

3-29

Configuring IP Exterior Gateway Protocols (BGP and EGP)

For example, the following command sequence configures the router to log BGP warning, fault, and trace messages for open and keepalive packets for a peer connection, and then disables message logging:

peer/2.2.2.2/2.2.2.4# debug-control/2.2.2.2/2.2.2.4# debug-control/2.2.2.2/2.2.2.4# debug-control/2.2.2.2/2.2.2.4# debug-control/2.2.2.2/2.2.2.4# peer/2.2.2.2/2.2.2.4#

Using Site Manager

To configure BGP event message logging, complete the following steps:

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Click on

6. Set the following parameters:

• Peer Local/Remote

• Message Level

• Message Trace Switch

Click on descriptions beginning on page A-24

7. Click on OK. You return to the BGP DEBUG

8. Click on

debug-control

log-message-type {warning fault trace} log-packet-type {open keepalive} log-packet-type disabled back

Site Manager Procedure

The Protocols menu opens.

Protocols

. The IP menu opens.

BGP BGP DEBUG.

Add

Help

Apply

. The BGP menu opens.

The BGP DEBUG Parameters window opens.

. The NEW BGP DEBUG window opens.

or see the parameter

Parameters window.

, and then click on

. You return to the Configuration Manager

Done

window.

3-30

308628-15.0 Rev 00

Configuring Global BGP Parameters

Configuring EBGP Route Flap Damping

The frequent change of network reacha bili ty inf ormatio n th at can be caused b y an unstable route is commonly referred to as route flap. Route flap damping is a technique for suppressing information about unstable routes.

When configuring the damping of unstable EBGP routes, the route flap damping mechanism performs the following functions:

• Determines the stability of an EBGP rout e

• Suppresses the use and advertisement of unstable EBGP routes

• Unsuppresses a route that has regained stability

To determine the stability of a route in the IP routing table, BGP maintains a penalty value for the route based on its recent history. Each time the route flaps (that is, each time it is withdrawn from the routing tabl e), BGP increments this penalty value. During the period of time that the route does not flap, BGP decrements the penalty val u e. I n thi s way, as the instability of the route increases, the penalty value rises. As the route becomes more stable, the penalty value falls.

To monitor BGP for route flaps, you set the route damping parameter on a BGP accept policy th at matches the r oute. BGP cre ates a pen alty val ue for th e route an d sets the value to 0. BGP increments the penalty value each time that the route flaps.

BGP provides a route flap damping template that allows you to control the way BGP evaluates a route’s penalty and how, based on the penalty, it decides to suppress or unsuppress the route. The template consists of the following parameters:

• Cutoff threshol d. If the penalt y value rises above the valu e that you speci fy as

• Reuse threshold. If the pen al ty val ue fall s bel ow the value that you specify as

• Reachable decay. If the route is still rea chable af ter the pe riod of ti me that you

• Unreachable decay. If the route is still unreachable after the period of time

308628-15.0 Rev 00

the cutoff threshold, BGP suppresses the route.

the reuse threshold, BGP unsuppresses the route.

specify, BGP reduces the penalty value by half. (BGP uses a half-life decay algorithm to decrement the penalty value.)

that you specify, BGP reduces the penalty value by half. ( BGP uses a half -li fe decay algorithm to decrement the penalty value.)

3-31

Configuring IP Exterior Gateway Protocols (BGP and EGP)

• Maximum holddown time. If the rou te remains supp ressed after t he number of minutes that you specify, BGP unsuppresses it (even if the penalty value exceeds the threshold). This action does not change the penalty value.

• Memory limit. If the status of the route remains unchanged, either reachable or unreachable, after the number of minutes that you specify, BGP unsuppresses the route (if it is suppressed) and resets the penalty value and flap count to 0.

A route that has been suppressed may not be immediately available for

Note:

use as soon as the penalty value crosses the configured threshold.

You can also use the BCC or Site Manage r to crea te a route f lap damping t emplate and a BGP accept policy for route flap damping.

Using the BCC

T o create a route fl ap damping te mplate, go t o the BGP prompt (for e xample,

ip; bgp damping-template name <name>

name

) and enter:

is a unique name identifying the template (from 1 through 20 characters).

box;

A template-specific prompt appears. For example, the following command creates a route flap damping template

named steady:

bgp#

damping-template name steady

damping-template/steady#

To change the default values of the parameters for a template that you created, go to the template-specific prompt (for example,

damping-template/steady

Table 3-6 l

ists the parameters and values that you can enter

) and enter:

box; ip; bgp;

3-32

308628-15.0 Rev 00

Configuring Global BGP Parameters

Table 3-6. Route Flap Damping Template Parameters

Parameter

cutoff-threshold

reuse-threshold

reachable-decay

unreachable-decay

max-hold-down

memory-limit

Default Value

2000

750

Meaning

Specifies a maximum threshold (1 through 20000) for the route penalty value

Specifies a minimum threshold (1 through 20000) for the route penalty value

Specifies the number of minutes (1 through 45) that the route can remain reachable before BGP decrements the penalty value by half

Specifies the number of minutes (1 through 45) that the route c an re ma in unreachable befo re BGP decrements the penalty value by half

Specifies the maximum number of minutes (1 through 60) that a route can remain suppressed

Specifies the maximum number of minutes (1 through 60) tha t the s tatus of a ro ute c an rem ain unchanged—reachable or unreachable—before BGP resets the penalty value and flap count to 0 and unsuppresses the route

To apply the template to an inbound BGP routing update:

Go to the BGP prompt (for example,

box; ip; bgp)

308628-15.0 Rev 00

Create and name a BGP accept policy.

For instructions on creating a BGP accept policy, see “

Accept Policy” on page 5-2.

An accept policy-specific prompt appears.

Configuring a BGP

3-33

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Modify the accept policy by entering the following commands:

action accept route-damping enabled route-damping-template {default

is the name of the template that you want to apply to the unstable route.

name

Match the acc ept policy to the unstable route.

<name>

}

For instructions, see “

Specifying Match Criteri a for a BGP Acc ept Policy ” on

page 5-6.

For example, the following command sequence creates a BGP accept policy named fred that causes BGP to apply the route flap damping template steady to updates for network 3.4.0.0:

accept fred

bgp# accept/fred/bgp# accept/fred/bgp# accept/fred/bgp# accept/fred/bgp# match/bgp/accept/fred# network/3.4.0.0/255.255.0.0/exact/bgp/accept/fred#

action accept route-damping enabled route-damping-template steady match

network 3.4.0.0/16/exact

3-34

308628-15.0 Rev 00

Using Site Manag er

To create a route flap damping template, complete the following steps:

You do this System responds

Configuring Global BGP Parameters

Site Manager Procedure

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Choose

6. Click on

7. Click on

8. Set the following parameters:

• Name

• Cutoff Threshold

• Reuse Threshold

• Reachable Decay

• Unreachable Decay

• Max HoldDown

• Memory Limit

Click on descriptions beginning on page A-47.

9. Click on OK. You return to the Route Flap Dampening

10. Click on

Protocols

. The IP menu opens.

Policy Filters BGP-4 Accept Policies

Route Flap

Add

Help

Done

. The Policy Filters menu opens.

. The BGP-4 menu opens.

. The BGP4 Accept Policy Filters window

. The Route Flap Dampening List window

. The Route Flap Dampening

or see the parameter

. You return to the BGP4 Accept Policy

The Protocols menu opens.

opens.

Configuration window opens.

List window.

Filters window.

308628-15.0 Rev 00

3-35

Configuring IP Exterior Gateway Protocols (BGP and EGP)

To apply a route flap damping template to an existing BGP accept policy, complete the following steps:

Site Manager Procedure

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Choose

6. Click on the accept policy to which you want to apply route flap damping.

7. Set the following parameters:

• BGP Route Flap Dampening

Click on descriptions beginning on page A-45

Note: The BGP Route Flap Dampening Template parameter is grayed out unless the BGP Route Flap Dampening parameter is set to ENABLE.

8. Click on

Protocols

. The IP menu opens.

Policy Filters BGP-4 Accept Policies

Template

Help

Apply

. The Policy Filters menu opens.

. The BGP-4 menu opens.

. The BGP4 Accept Policy Filters window

or see the parameter

, and then click on

Done

The Protocols menu opens.

opens. The parameter values for that policy

appear in the window.

. You return to the Configuration Manager

window.

3-36

308628-15.0 Rev 00

Configuring Global BGP Parameters

Assigning Weight Classes and Values to an AS

You can assign a weight class to any AS number and a weight value to a weight class. Weights provide a way either to prefer or to avoid routes that pass through certain ASs. The weights of each AS in a path are added, and the path with the smallest total weight is the preferred path. An assigned weight can range from 1 through 15 plus an infi ni ty value. Any path containing an AS wei ght of i nfi nity is avoided.

AS weight classes allow you to assign multiple weight values to the same AS. This feature allows you to consider an AS path differently for different networks. For example, consider a situation in which two networks—192.32.1.0 and

192.32.2.0—are both reachable by two paths. The first path to each network shares a common AS—AS 5. The second path to each network also shares a common AS—AS 10. If you want to favor AS 5 in the path to 192.32.1.0 and AS 10 in the path to 192.32.2.0, you c an assign one wei ght class to AS 5 in the path to network 192.32.1.0 and another class to AS 5 in the path to 192.32.2.0.

When a BGP router receives a new route, it evaluates the route against any existing accept policies. If after this evaluation the path still is to be used, the router calculates the total weight of the path. Configure the same AS weights on all BGP routers in an AS.

You can use Site Manager to assign a weight and a weight class to an AS by completing the following st eps:

308628-15.0 Rev 00

Site Manager Procedure

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Click on

Protocols IP

. The IP menu opens.

BGP Weights

Add

. The BGP menu opens.

. The BGP AS Weight Parameters wind ow

. The BGP AS Weights window opens.

The Protocols menu opens.

opens.

(continued)

3-37

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Site Manager Procedure

You do this System responds

6. Set the following parameters:

•

Weight Value 1 through 8

Click on descriptions beginning on page A-20

7. Click on OK. You return to the BGP AS Weight

8. Click on

or see the parameter

Help

. You return to the Configuration Manager

Done

(continued)

Parameters window.

window.

3-38

308628-15.0 Rev 00

Chapter 4

Configuring BGP Peers

A BGP speaker forms a neighbor relat ionship by estab lishing a TCP connecti on to another BGP speaker or peer, based on local configuration information. You can configure a BGP peer-to-peer session by setting the BGP parame te rs de scr ibed in the following sections:

Topic Page

Defining a Peer-to-Peer Session 4-2 Initiating a Peer-to-Peer Session 4-4 Negotiating the BGP Version 4-6 Keeping the Connection Alive 4-8 Setting the External Advertisement Timer 4-9 Specifying a Holddown Time 4-11 Setting a Minimum AS Origination Interval 4-12 Overriding the Local AS Number 4-14 Specifying a Maximum Update Size 4-14 Specifying a Time-to-Live Value 4-16 Specifying the Next-Hop Router 4-17 Setting the Route Echo Switch 4-18 Disabling and Reenabling Loop Detection 4-20 Configuring Peers over an Unnumbered Point-to-Point Link 4-21 Configuring and Enabling MD5 Authentication 4-22

308628-15.0 Rev 00

4-1

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Defining a Peer-to-Pe er Session

To define a peer-to-peer session, specify the following:

• Local IP interface address

• Remote IP interface addres s

• AS number of the autonomous system in which the remote BGP peer is

located

In configurations where BGP speakers reside on routers that have multiple network connections over multiple IP interfaces (the typical case for IBGP speakers), consider using the address of the router’s circuitless (virtual) IP interface as the local peer address. In this way, you ensure that BGP is reachable as long as there is an active circuit on the router.

If the remote peer is located in a different AS from the local peer, the remote address must be on the same subnet as the local address. (To override this restriction, see “ and the remote peer are located in the same AS, BGP assumes that you are configuring an IBGP session and does not impose this restriction.

Enabling Multihop Connections” on page 3-13.)

If the local peer

Using the BCC

4-2

You can use the BCC or Site Manager to specify this information.

To define a peer-to-peer session, go to the BGP prompt (for example,

bgp

) and enter

peer local

local_address remote_address as_number

<local_address>

remote

<remote_address>

<as_number>

is the IP address of the local interface.

is the IP address of the remote interface.

is the number of the AS in which the remote peer is located.

box; ip;

For example, the following command defines a session with a remote peer in AS 5. The local IP interface is 2.2.2.2. The interface for the remote peer is 2.2.2.5.

bgp#

peer local 2.2.2.2 remote 2.2.2.5 as 5

peer/2.2.2.2/2.2.2.5#

308628-15.0 Rev 00

Using Site Manager

To define a peer-to-peer session, complete the following steps:

You do this System responds

Configuring BGP Peers

Site Manager Procedure

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Click on the IP interface for which you want to edit BGP peer parameters.

6. Click on

7. Click on

8. Set the following parameters:

• Peer Address

•

• Local Address

•

Click on descriptions beginning on page A-11

9. Click on OK. You return to the BGP Peer List window.

10. Click on

Protocols

. The IP menu opens.

BGP Peers

BGP Peers Add

Peer AS

Peer Mode

Help

Apply

. The BGP menu opens.

. The IP Interface List for BGP window

. The BGP Peer List window opens.

. The BGP Peer window opens.

or see the parameter

, and then click on

Done

The Protocols menu opens.

opens.

. Y ou retu rn to the IP Inte rface List for BGP

window.

308628-15.0 Rev 00

4-3

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Initiating a Peer-to-Peer Session

A BGP speaker ini tiates a pe er-t o-p eer connect ion b y period icall y is suing an o pen message. messages.

BGP speakers respond to connection requests by returning open

In Figure 4-1

, for example, BGP speaker A sends an open message to BGP speaker B to request a connection; BGP speaker B responds by sending an open message to BGP speaker A.

Open message A to B

BGP

speaker A

Speaker A

BGP

speaker A

Speaker A

Open message B to A

Keepalive message A to B

Keepalive message B to A

BGP

speaker B

Speaker B

BGP

speaker B

Speaker B

IP0022B

Figure 4-1. Establishing and Confirming a Connection Between BGP

Peers

All BGP speakers respond to connection requests from other speakers. By default, BGP attempts to initiate a connection on each interface configured for

peer-to-peer communicat ion. If the atte mpt is unsucces sful, BGP retrie s every 120 seconds (default interval).

4-4

You can use the BCC or Site Manager to change the default retry interval or disable the initiation function by setting the retry parameter to 0.

308628-15.0 Rev 00

Using the BCC

Configuring BGP Peers

To set the retry interval, go to a BGP peer prompt (for example,

peer/2.2.2.2/2.2.2.3

retry

interval

attempts to initi ate a pee r-to- peer se ssion. Settin g thi s value to 0 disab les at tempt s to initiate a peer-to-peer session.

For example, the following command causes BGP to retry every 60 seconds to establish a peer-to-peer session between IP interfaces 2.2.2.2 and 2.2.2.3:

peer/2.2.2.2/2.2.2.3# peer/2.2.2.2/2.2.2.3#

Using Site Manager

To set the retry in terval, complete the following steps:

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Click on the IP interface for which you want to edit BGP peer parameters.

6. Click on

7. Click on the peer for which you want to edit parameters.

8. Set the Click on description on page A-14

9. Click on

box; ip; bgp;

) and enter

interval>

is the number of seconds (from 1 through 2,147,483,647) between

retry 60

Site Manager Procedure

The Protocols menu opens.

Protocols

. The IP menu opens.

BGP Peers

Connect Retry Timer

Help

Apply

. The BGP menu opens.

. The IP Interface List for BGP window

opens.

. The BGP Peer List window opens.

The parameters for that peer appear in the window.

parameter.

or see the parameter

, and then click on

. Y ou retu rn to the IP Inte rface List for BGP

Done

window.

308628-15.0 Rev 00

4-5

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Negotiating the BGP Version

BGP peers negotiate the version of BGP that they will use to exchange routing information. If you enab le both BGP-3 and BGP-4, t he rou ter fi rst at tempts to use BGP-4. If the BGP peer is not a BGP-4 speaker, the router uses BGP-3.

By default, BGP considers BGP-4 as both the minimum and maximum acc eptable version for negotiation.

You can use the BCC or Site Manager to specify BGP-3 as the minimum or maximum acceptable version.

Using the BCC

To specify the minimum version, go to a BGP peer prompt

bgp; peer/2.2.2.2/2.2.2.3

min-version

version

is either

version>

bgp3

) and

bgp4

enter:

(default).

T o spe cify the maxi mum versi on, go to a BGP pe er prompt

bgp; peer/2.2.2.2/2.2.2.3

max-version

version

is either

version>

bgp3

) and

bgp4

enter:

(default).

(for example,

box; ip;

4-6

308628-15.0 Rev 00

Using Site Manager

To specify the minimum and maximum version of BGP, complete the following steps:

You do this System responds

Configuring BGP Peers

Site Manager Procedure

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Click on the IP interface for which you want to edit BGP peer parameters.

6. Click on

7. Click on the peer for which you want to edit parameters.

8. Set the following parameters:

•

Click on descriptions starting on page A-13

9. Click on

Protocols

. The IP menu opens.

BGP Peers

Min BGP Version Max BGP Version

Help

Apply

. The BGP menu opens.

. The IP Interface List for BGP window

. The BGP Peer List window opens.

or see the parameter

, and then click on

Done

The Protocols menu opens.

opens.

The parameters for that peer appear in the window.

. Y ou retu rn to the IP Inte rface List for BGP

window.

308628-15.0 Rev 00

4-7

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Keeping the Connection Alive

After a session is es tabli sh ed, BGP peer s peri odical ly is sue kee paliv e message s to maintain the connection. By default, BGP issues a keepalive message every 30 seconds.

You can use the BCC or Site Manager to specify how often BGP issues a keepalive message on this peer-to-p eer session or to disable t he keepalive f unction by setting the value to 0.

Using the BCC

T o set the k eepa li ve interv al, go to a BGP peer prompt (fo r exa mple,

peer/2.2.2.2/2.2.2.3

keepalive

seconds

2,147,483,647) on this peer ses sion. Set ting this value to 0 disables the sending of keepali ve messages.

For example, the following command causes BGP to send a keepalive message every 10 seconds on interface 2.2.2.2 to the peer at 2.2.2.5:

peer/2.2.2.2/2.2.2.5/5# peer/2.2.2.2/2.2.2.5/5#

Using Site Manager

To set the keepalive interval, complete the following steps:

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

box; ip; bgp;

) and enter:

seconds>

specifies how often BGP sends a keepalive message (from 1 through

keepalive 10

Site Manager Procedure

The Protocols menu opens.

Protocols IP

. The IP menu opens.

BGP Peers

. The BGP menu opens.

. The IP Interface List for BGP window

opens.

(continued)

4-8

308628-15.0 Rev 00

Configuring BGP Peers

Site Manager Procedure

You do this System responds

5. Click on the IP interface for which you want to edit BGP peer parameters.

6. Click on

7. Click on the peer for which you want to edit parameters.

8. Set the on on page A-15

9. Click on

BGP Peers

KeepAlive Timer

Help

or see the parameter description

Apply

. The BGP Peer List window opens.

parameter . Click

, and then click on

Done

. Y ou retu rn to the IP Inte rface List for BGP

(continued)

The parameters for that peer appear in the window.

window.

Setting the Extern al Ad ve rtisement Ti mer

After a connection is established, the BGP speaker uses one or more update messages to send the entire IP routing table (compliant with local BGP announce policies). BGP, however, does not require the entire routing table to be sent again. Therefore, the BGP speaker must keep a current version of the routing information rece ived fro m all i ts peer s for as long as t he connec tion to each pe er is valid. This information is updated by means of update messages whenever changes occur.

By default, BGP examines the routing table for changes every 30 seconds. If a change has occurred, BGP issues an update message on the connection.

You can use the BCC or Site Manager to specify a value for the external advertisement timer.

Using the BCC

To specify a value for the external advertis ement timer, go to a BGP peer prompt (for example ,

advertise-time

seconds

interval at which BGP examines th e routing table for changes before issuing an update message on this peer session.

308628-15.0 Rev 00

box; ip; bgp; peer/2.2.2.2/2.2.2.3

seconds>

) and enter:

is an integer (from 1 through 2,147,483,647) specifying the minimum

4-9

Configuring IP Exterior Gateway Protocols (BGP and EGP)

For example, the following command sets the external ad vertisement timer to 20 seconds for the peer session established between interfaces 2.2.2.2 and 2.2.2.5:

peer/2.2.2.2/2.2.2.5# peer/2.2.2.2/2.2.2.5#

Using Site Manager

To s pecify a value for the external advertisement timer, complete the following steps:

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Click on the IP interface for which you want to edit BGP peer parameters.

6. Click on

7. Click on the peer for which you want to edit parameters.

8. Set the parameter. Click on parameter description on page A-14.

9. Click on

advertise-time 20

Site Manager Procedure

The Protocols menu opens.

Protocols IP

. The IP menu opens.

BGP Peers

External Advertisement Tim er

Apply

. The BGP menu opens.

. The IP Interface List for BGP window

opens.

. The BGP Peer List window opens.

The parameters for that peer appear in the window.

Help

or see the

, and then click on

Done

. Y ou retu rn to the IP Inte rface List for BGP

window.

4-10

308628-15.0 Rev 00

Specifying a Hold down Time

The holddown time is the amount of time that either peer waits for a keepalive or update message before declaring the connection down.

A BGP speaker that initiates a connection inserts a holddown time value into the open message. The peer responds with an open message that also contains a holddown time value. If the BGP speakers establish a session, they use the lesser value (which must be greater than 2 seconds). There are two exceptions to this rule:

• If one peer sends a nonzero holddown time, the peers use the nonzero holddown time on the session.

• If both peers send zero holddown times, the peers observe no holddown time on the session.

By default, BGP inserts a value of 90 seconds into the open message. You can use the BCC or Site Manager to reset the holddown time value or to

disable the holddown function by setting this value to 0.

Configuring BGP Peers

Using the BCC

T o specify a hol ddo wn time, go to a BGP peer prompt (f or exam ple,

peer/2.2.2.2/2.2.2.3 holddown

seconds

for a keepalive or update message before declaring the connection down. For example, the following command sets the holddown time to 60 seconds for

the peer session established between interfaces 2.2.2.2 and 2.2.2.5:

peer/2.2.2.2/2.2.2.5# peer/2.2.2.2/2.2.2.5#

308628-15.0 Rev 00

box; ip; bgp;

) and enter:

is the number of s eco nds (f rom 3 t hrough 2,1 47,483,647 ) tha t BGP w ai ts

holddown 60

4-11

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Using Site Manager

To s pecify a holdd own ti me, complete the following steps:

Site Manager Procedure

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Click on the IP interface for which you want to edit BGP peer parameters.

6. Click on

7. Click on the peer for which you want to edit parameters.

8. Set the

Help

page A-15.

9. Click on

Protocols IP

. The IP menu opens.

BGP Peers

Holdtime

or see the parameter description on

Apply

. The BGP menu opens.

. The IP Interface List for BGP window

. The BGP Peer List window opens.

parameter. Click on

, and then click on

Done

The Protocols menu opens.

opens.

The parameters for that peer appear in the window.

. Y ou retu rn to the IP Inte rface List for BGP

window.

Setting a Minimum AS Origination Interval

By default, a BGP speaker that issues an update to adve rtise a change in the AS must wait at least 15 seconds before advertising a subsequent change.

Using the BCC

4-12

You can use the BCC or Site Manager to specify a different interval.

To specify a different interval, go to a BGP peer prompt (for example,

bgp; peer/2.2.2.2/2.2.2.3

min-originate-time

seconds

is an integer (from 1 through 2,147,483,647) indicating the minimum

) and enter:

seconds>

number of seconds that BGP waits between advertisements.

308628-15.0 Rev 00

box; ip;

For example, the following command causes BGP to wait at least 30 seconds between updates on the peer session established between interfaces 2.2.2.2 and

2.2.2.5:

peer/2.2.2.2/2.2.2.5# min-originate-time 30 peer/2.2.2.2/2.2.2.5#

Using Site Manager

To change the minimum AS origination interval, complete the following steps:

You do this System responds

Configuring BGP Peers

Site Manager Procedure

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Click on the IP interface for which you want to edit BGP peer parameters.

6. Click on

7. Click on the peer for which you want to edit parameters.

8. Set the parameter. Click on parameter description on page A-16

9. Click on

Protocols IP

. The IP menu opens.

BGP Peers

Min AS Origination Interval

Apply

. The BGP menu opens.

. The IP Interface List for BGP window

. The BGP Peer List window opens.

Help

or see the

, and then click on

Done

The Protocols menu opens.

opens.

The parameters for that peer appear in the window.

. Y ou retu rn to the IP Inte rface List for BGP

window.

308628-15.0 Rev 00

4-13

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Overriding the Local AS Number

By default, a BGP speaker that issues an open message to initiate a peer-to-peer session uses the AS number that you set with the Local AS parameter.

You can use Site Manager to specify a different AS number (overriding the default) by completing the following steps:

Site Manager Procedure

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Click on the IP interface for which you want to edit BGP peer parameters.

6. Click on

7. Click on the peer for which you want to edit parameters.

8. Set the parameter. Click on parameter description on page A-16

9. Click on

Protocols IP

. The IP menu opens.

BGP Peers

Local AS to Advertise to Peer

Apply

. The BGP menu opens.

. The IP Interface List for BGP window

. The BGP Peer List window opens.

Help

or see the

, and then click on

Done

Specifying a Maximum Update Size

By default, a BGP speaker sends update messages with a maximum size of 800

You can reset the maximum update message size.

bytes. advertises a single route is larger than the configured message size, the actual message size can override the configured value.

The Protocols menu opens.

opens.

The parameters for that peer appear in the window.

. Y ou retu rn to the IP Inte rface List for BGP

window.

If the update message th at

4-14

You can use the BCC or Site Manager to specify a maximum upd ate message s ize.

308628-15.0 Rev 00

Using the BCC

Configuring BGP Peers

To specify a maximum update size, go to a BGP peer prompt (for example,

ip; bgp; peer/2.2.2.2/2.2.2.3 max-update-size

bytes

this peer session. For example, the following command specifies a maximum size of 950 bytes for

updates sent on the peer session established between interfaces 2.2.2.2 and

2.2.2.5:

peer/2.2.2.2/2.2.2.5# peer/2.2.2.2/2.2.2.5#

Using Site Manager

To specify a maximum update size, complete the following steps:

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

5. Click on the IP interface for which you want to edit BGP peer parameters.

6. Click on

7. Click on the peer for which you want to edit parameters.

8. Set the Click on description on page A-16

9. Click on

box;

) and enter:

bytes>

is the maximum size (from 64 through 4096) of updates that BGP sends on

max-update-size 950

Site Manager Procedure

The Protocols menu opens.

Protocols

. The IP menu opens.

BGP Peers

Peer Max Update Size

Help

Apply

. The BGP menu opens.

. The IP Interface List for BGP window

opens.

. The BGP Peer List window opens.

The parameters for that peer appear in the window.

parameter .

or see the parameter

, and then click on

. Y ou retu rn to the IP Inte rface List for BGP

Done

window.

308628-15.0 Rev 00

4-15

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Specifying a Time-to-Live Value

Each IP data packet includes a time-to-live (TTL) value. The TTL value specifies the maximum number of hops that the pack et i s all o wed to tra verse in the network before an intermediate router discards the packet. The TTL counter prevents packets from looping endlessly through the network. For instructions on setting the global IP TTL value, see Configuring IP, ARP, RARP, RIP, and OSPF Services.

By default, BGP implements a TTL value as follows:

• IBGP peers use the TTL value set for global IP.

• BGP enforces the one-hop rule for EBGP peers, that is, the remote peer must

be located on a directly attached network.

If you enable multihop connections for EBGP peers, EBGP peers also use the TTL value set for global IP (see “Enabling Multihop Connections” on page 3-13).

You can specify a TTL value for a BGP session that overrides the TTL value set for IP. To specify a TTL value for a BGP peer connection, go to a BGP peer prompt (for example,

ttl

<hops>

box; ip; bgp; peer/2.2.2.2/2.2.2.3

) and enter:

4-16

hops

is the time-to-live value (expressed as the number of hops from 1 through

255) that BGP inserts in outbound updates sent on this peer session. The default value is 0 (to use the same TTL value set for global IP).

For example, the following command sequence enables multihop connections for all EBGP peers, configures a session with a remote peer i n AS 5, and causes BGP to insert a TTL value of 4 in each outbound update sent over the peer connection:

multi-hop enabled

bgp#

peer local 2.2.2.2 remote 3.3.3.3 as 5

bgp# peer/2.2.2.2/3.3.3.3# peer/2.2.2.2/3.3.3.3#

ttl 4

308628-15.0 Rev 00

Specifying the Next-Hop Router

A BGP update message has a path attribute called NEXT_HOP that specifies the IP address of the router to use as the next hop for the advertised destinations. By default, the NEXT_HOP at tribute can specify the IP address of a route r other than this BGP router.

Configuring the peer to advertise a next hop other than itself can eliminate extra forwarding hops. However, you can configure the router to always send the local BGP peer’s IP address as the next hop in all advertisements.

Using the BCC

To specify whether the BGP peer router sends its own address as the next hop in advertisements, go to a BGP peer prompt (for example,

peer/2.2.2.2/2.2.2.3

) and enter:

Configuring BGP Peers

box; ip; bgp;

next-hop-self

state

For example, the following command causes BGP to send the local peer’s I P address as the next hop in all advertisements:

peer/2.2.2.2/2.2.2.5# peer/2.2.2.2/2.2.2.5#

Using Site Manager

To specify whether the BGP peer router sends its own address as the next hop in advertisements, complete the following steps:

You do this System responds

1. In the Configuration Manager window, choose

2. Choose

3. Choose

4. Choose

is either

Protocols IP

BGP Peers

<state>

enabled

. The IP menu opens.

. The BGP menu opens.

. The IP Interface List for BGP window

disabled

next-hop-self enabled

(default).

Site Manager Procedure

The Protocols menu opens.

opens.

(continued)

308628-15.0 Rev 00

4-17

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Site Manager Procedure

You do this System responds

5. Click on the IP interface for which you want to edit BGP peer parameters.

6. Click on

7. Click on the peer for which you want to edit parameters.

8. Set the

Help

page A-18

9. Click on

BGP Peers

Next Hop Self

or see the parameter description on

Apply

. The BGP Peer List window opens.

parameter . Clic k on

, and then click on

Setting the Route Echo Switch

The peer route echo switch controls the way that the router echoes a BGP route that is chosen for forwarding. Echoing in this case means advertising the route back to the peer from which it was received.

By default, the router advertises the route back as unreachable. However, you can configure the router to advertise the route back as reachable and to include the local AS.

(continued)

The parameters for that peer appear in the window.

Done

. Y ou retu rn to the IP Inte rface List for BGP

window.

4-18

You can use the BCC or Site Manager to set the peer route echo switch.

308628-15.0 Rev 00

Avaya IP Exterior Gateway Protocols User Manual

Specifications and Main Features

Frequently Asked Questions

User Manual

Configuring IP Exterior Gateway Protocols (BGP and EGP)

Contents

Figures

Tables

Preface

Before You Be gin

Text Conventions

Acronyms

Related Publications

How to Get Help

Autonomous Systems a nd Gateway Protocols

Border Gateway Protocol (BGP)

Exterior Gateway Protocol (EGP)

Classless Interdomain Routing

BGP Concepts and Terminology

Peer-to-Peer Sessions

Stub and Multihomed Autonomous Systems

Interior BGP Routing

IBGP Route Reflector

Equal-Cost Multipath

BGP Updates

Path Attributes

BGP/OSPF Interaction

BGP-4 Confederations

BGP-4 TCP MD5 Message Authentication

BGP Implementation Notes

Starting BGP with the BCC

Step 1: Configuring Global BGP

Step 2: Defining a Peer-to-Peer Connection

Starting IP and BGP with Site Manager

Deleting BGP with Site Manager

Deleting BGP-3 and BGP-4 with Site Manager

Starting IP an d EGP with Site Manager

Deleting EGP from the Router

Disabling and Reenabling BGP

Supplying a BGP Router ID

Identifying the Local AS

Disabling and Reenabling IBGP Support

Specifying Route Types for IBGP Advertise m ents

Enabling BGP Interaction with OSPF and RIP

Setting the Update Interval Timer

Allowing Redundant Connections

Enabling Multihop Connections

Disabling and Reenabling Dynamic Policy Configuration

Configuring the BGP Soloist Slot Mask

Disabling and Reenabling Rou te Aggregation

Enabling and Disabling Black Hole Punching

Disabling and Reenabling th e BGP-4 MED Attribute

Configuring BGP-4 Confede rations

Disabling BayRS Local Preference Calculation and Route Selection

Calculating BGP-4 Local Preference Values

Best-Route Selection

Configuring BGP Message Logging

Configuring EBGP Route Flap Damping

Assigning Weight Classes and Values to an AS

Defining a Peer-to-Pe er Session

Initiating a Peer-to-Peer Session

Negotiating the BGP Version

Keeping the Connection Alive

Setting the Extern al Ad ve rtisement Ti mer

Specifying a Hold down Time

Setting a Minimum AS Origination Interval

Overriding the Local AS Number

Specifying a Maximum Update Size

Specifying a Time-to-Live Value

Specifying the Next-Hop Router

Setting the Route Echo Switch