StudentGuide Implementing Broadband Aggregation On Cisco10k Vol2

BBAGG
Volume 2
Implementing
Broadband Aggregation
on Cisco 10000 Series
Version 1.0
Student Guide
The products and specifications, configurations, and other technical information regarding the products in this manual
are subject to change without notice. All statements, technical information, and recommendations in this manual are
believed to be accurate but are presented without warranty of any kind, express or implied. You must take full
responsibility for their application of any products specified in this manual.
LICENSE
PLEASE READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THE MANUAL,
DOCUMENTATION, AND/OR SOFTWARE (MATERIALS). BY USING THE MATERIALS YOU AGREE TO
BE BOUND BY THE TERMS AND CONDITIONS OF THIS LICENSE. IF YOU DO NOT AGREE WITH THE
TERMS OF THIS LICENSE, PROMPTLY RETURN THE UNUSED MATERIALS (WITH PROOF OF PAYMENT)
TO THE PLACE OF PURCHASE FOR A FULL REFUND.
Cisco Systems, Inc. (Cisco) and its suppliers grant to you (You) a nonexclusive and nontransferable license to use
the Cisco Materials solely for Your own personal use. If the Materials include Cisco software (Software), Cisco
grants to You a nonexclusive and nontransferable license to use the Software in object code form solely on a single
central processing unit owned or leased by You or otherwise embedded in equipment provided by Cisco. You may make
one (1) archival copy of the Software provided You affix to such copy all copyright, confidentiality, and proprietary
notices that appear on the original. EXCEPT AS EXPRE SSLY AUTHORIZED ABOVE, YOU SHALL NOT: COPY,
IN WHOLE OR IN PART, MATERIALS; MODIFY THE SOFTWARE; REVERSE COMPILE OR REVERSE
ASSEMBLE ALL OR ANY PORTION OF THE SOFTWARE; OR RENT, LEASE, DISTRIBUTE, SELL, OR
CREATE DERIVATIVE WORKS OF THE MATERIALS.
You agree that aspects of the licensed Materials, including the specific design and structure of individual programs,
constitute trade secrets and/or copyrighted material of Cisco. You agree not to disclose, provide, or otherwise make
available such trade secrets or copyrighted material in any form to any third party without the prior written consent
of Cisco. You agree to implement reasonable security measures to protect such trade secrets and copyrighted Material.
Title to the Materials shall remain solely with Cisco.
This License is effective until terminated. You may terminate this License at any time by destroying all copies of the
Materials. This License will terminate immediately without notice from Cisco if You fail to comply with any
provision of this License. Upon termination, You must destroy all copies of the Materials.
Software, including technical data, is subject to U.S. export control laws, including the U.S. Export Administration Act
and its associated regulations, and may be subject to export or import regulations in other countries. You agree to
comply strictly with all such regulations and acknowledge that it has the responsibility to obtain licenses to export, reexport, or import Software.
This License shall be governed by and construed in accordance with the laws of the State of California, United States
of America, as if performed wholly within the state and without giving effect to the principles of conflict of law. If
any portion hereof is found to be void or unenforceable, the remaining provisions of this License shall remain in full
force and effect. This License constitutes the entire License between the parties with respect to the use of the
Materials
Restricted Rights - Ciscos software is provided to non-DOD agencies with RESTRICTED RIGHTS and its supporting
documentation is provided with LIMITED RIGHTS. Use, duplication, or disclosure by the U.S. Government is subject
to the restrictions as set forth in subparagraph C of the Commercial Computer Software - Restricted Rights clause at
FAR 52.227-19. In the event the sale is to a DOD agency, the U.S. Governments rights in software, supporting
documentation, and technical data are governed by the restrictions in the Technical Data Commercial Items clause at
DFARS 252.227-7015 and DFARS 227.7202.
DISCLAIMER OF WARRANTY. ALL MATERIALS ARE PROVIDED AS IS WITH ALL FAULTS. CISCO AND
ITS SUPPLIERS DISCLAIM ALL WARRANTIES, EXPRESSED OR IMPLIED, INCLUDING, WITHOUT
LIMITATION, THOSE OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT OR ARISING FROM A COURSE OF DEALING, USAGE, OR TRADE PRACTICE.
IN NO EVENT SHALL CISCO OR ITS SUPPLIERS BE LIABLE FOR ANY INDIRECT, SPECIAL,
CONSEQUENTIAL, OR INCIDENTAL DAMAGES, INCLUDING, WITHOUT LIMITATION, LOST PROFITS OR
LOSS OR DAMAGE TO DATA ARISING OUT OF THE USE OR INABILITY TO USE THIS MANUAL, EVEN IF
CISCO OR ITS SUPPLIERS HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. In no event
shall Ciscos or its suppliers liability to You, whether in contract, tort (including negligence), or otherwise, exceed the
price paid by You. The foregoing limitations shall apply even if the above-stated warranty fails of its essential
purpose.
The following information is for FCC compliance of Class A devices: This equipment has been tested and found to
comply with the limits for a Class A digital device, pursuant to part 15 of the FCC rules. These limits are designed to
provide reasonable protection against harmful interference when the equipment is operated in a commercial
environment. This equipment generates, uses, and can radiate radio -frequency energy and, if not installed and used in
accordance with the instruction manual, may cause harmful interference to radio communications. Operation of this
equipment in a residential area is likely to cause harmful interference, in which case users will be required to correct the
interference at their own expense.
The following information is for FCC compliance of Class B devices: The equipment described in this manual
generates and may radiate radio -frequency energy. If it is not installed in accordance with Ciscos installation
instructions, it may cause interference with radio and television reception. This equipment has been tested and found
to comply with the limits for a Class B digital device in accordance with the specifications in part 15 of the FCC rules.
These specifications are designed to provide reasonable protection against such interference in a residential
installation. However, there is no guarantee that interference will not occur in a particular installation.
You can determine whether your equipment is causing interference by turning it off. If the interference stops, it was
probably caused by the Cisco equipment or one of its peripheral devices. If the equipment causes interference to radio
or television reception, try to correct the interference by using one or more of the following measures:
Turn the television or radio antenna until the interference stops.
Move the equipment to one side or the other of the television or radio.
Move the equipment farther away from the television or radio.
Plug the equipment into an outlet that is on a different circuit from the television or radio. (That is, make certain
the equipment and the television or radio are on circuits controlled by different circuit breakers or fuses.)
Modifications to this product not authorized by Cisco Systems, Inc. could void the FCC approval and negate your
authority to operate the product.
The following third-party software may be included with your product and will be subject to the software license
agreement:
CiscoWorks software and documentation are based in part on HP OpenView under license from the Hewlett-Packard
Company. HP OpenView is a trademark of the Hewlett-Packard Company. Copyright 1992, 1993 Hewlett-Packard
Company.
The Cisco implementation of TCP header compression is an adaptation of a program developed by the University of
California, Berkeley (UCB) as part of UCBs public domain version of the UNIX operating system. All rights reserved.
Copyright 1981, Regents of the University of California.
Network Time Protocol (NTP). Copyright 1992, David L. Mills. The University of Delaware makes no
representations about the suitability of this software for any purpose.
Point-to-Point Protocol. Copyright 1989, Carnegie-Mellon University. All rights reserved. The name of the
University may not be used to endorse or promote products derived from this software without specific prior written
permission.
The Cisco implementation of TN3270 is an adaptation of the TN3270, curses, and termcap programs developed by
the University of California, Berkeley (UCB) as part of UCBs public domain version of the UNIX operating system.
All rights reserved. Copyright 1981-1988, Regents of the University of California.
Cisco incorporates Fastmac and TrueView software and the RingRunner chip in some Token Ring products. Fastmac
software is licensed to Cisco by Madge Networks Limited, and the RingRunner chip is licensed to Cisco by Madge NV.
Fastmac, RingRunner, and TrueView are trademarks and in some jurisdictions registered trademarks of Madge
Networks Limited. Copyright 1995, Madge Networks Limited. All rights reserved.
XRemote is a trademark of Network Computing Devices, Inc. Copyright 1989, Network Computing Devices, Inc.,
Mountain View, California. NCD makes no representations about the suitability of this software for any purpose.
The X Window System is a trademark of the X Consortium, Cambridge, Massachusetts. All rights reserved.
Cisco Systems has more than 200 offices in the following countries and regions. Addresses, phone numbers, and fax
numbers are listed on the Cisco Web site at www.cisco.com/go/offices.
Argentina Australia Austria Belgium Brazil Bulgaria Canada Chile China PRC Colombia Costa Rica Croatia Czech
Republic Denmark Dubai, UAE Finland France Germany Greece Hong Kong SAR Hungary India Indonesia Ireland
Israel Italy Japan Korea Luxembourg Malaysia Mexico The Netherlands New Zealand Norway Peru Philippines Poland
Portugal Puerto Rico Romania Russia Saudi Arabia Scotland Singapore Slovakia Slovenia South Africa Spain Sweden
Switzerland Taiwan Thailand Turkey Ukraine United Kingdom United States Venezuela Vietnam Zimbabwe
Copyright
2003, Cisco Systems, Inc. All rights reserved. AccessPath, AtmDirector, Browse with Me,
CCDA,
CCDE, CCDP, CCIE, CCNA, CCNP, CCSI, CD-PAC, CiscoLink, the Cisco Net Works logo, the
Cisco
Powered Network logo, Cisco Systems Networking Academy, Fast Step, Follow Me Browsing,
FormShare, FrameShare, GigaStack, IGX, Internet Quotient, IP/VC, iQ Breakthrough, iQ Expertise, iQ FastTrack, the
iQ logo, iQ Net Readiness Scorecard, MGX, the Networkers logo, Packet, RateMUX, ScriptBuilder, ScriptShare,
SlideCast, SMARTnet, TransPath, Unit y, Voice LAN, Wavelength Router, and WebViewer are trademarks of Cisco
Systems, Inc.; Changing the Way We Work, Live, Play, and Learn, Discover All Thats Possible, and Empowering the
Internet Generation, are service marks of Cisco Systems, Inc.; and Aironet, ASIST, BPX, Catalyst, Cisco, the Cisco
Certified Internetwork Expert Logo, Cisco IOS, the Cisco IOS logo, Cisco Systems, Cisco Systems Capital, the Cisco
Systems logo, Enterprise/Solver, EtherChannel, EtherSwitch, FastHub, FastSwitch, IOS, IP/TV, LightStream, MICA,
Network Registrar, PIX, Post -Routing, Pre-Routing, Registrar, StrataView Plus, Stratm, SwitchProbe, TeleRouter, and
VCO are registered trademarks of Cisco Systems, Inc. and/or its affiliates in the U.S. and certain other countries.
All other brands, names, or trademarks mentioned in this document or Web site are the property of their respective
owners. The use of the word partner does not imply a partnership relationship between Cisco and any other company.
(0104R)
Book Title, Revision xx.x: Student Guide

Copyright 2003, Cisco Systems, Inc.
All rights reserved. Printed in USA.
Course Overview
Intended Audience
This course is for technical professionals who need to know how to
implement broadband aggregation on the Cisco 10000 Series router.
The following are considered the primary audience for this course:
Customer technicians
Cisco System Engineers (SEs)
System Integrators (SIs)
Course Level
This course is basic and intermediate training for the topics that it
covers.
Prerequisites
Students attending this course should have successfully completed the
following training:
Interconnecting Cisco Network Devices (ICND) or equivalent

experience
Campus ATM (CATM) or equivalent experience
Basic DSL End To End Architecture either video on demand or

leader-led or equivalent experience
2003 Cisco Systems, Inc.
Version 1.0
Additional Information
Cisco Systems Technical Publications
You can print technical manuals and release notes directly from the
Internet. Go to http://www.cisco.com/univercd/home/home.htm.
Find the Cisco Systems product for which you need documentation.
Then locate the specific category and model or version for your
hardware or software product. Using Adobe Acrobat Reader, you can
open the manuals and release notes, search for the sections you need,
and print them on most standard printers. You can download Acrobat
Reader free from the Adobe Systems website, www.adobe.com.
Documentation sets and CDs are available through your local Cisco
Systems sales office or account representative.
Cisco Systems Service
Comprehensive network support is available from Cisco Systems

Service & Support solutions. Go to
http://www.cisco.com/public/support_solutions.shtml for a listing of
services.
vi
Version 1.0
Implementing Broadband Aggregation
Course Agenda
Day 1
Broadband Aggregation Architectures
RBE and RFC 1483 Routing
PPPoA
Day 2
PPPoE
Cisco Aggregation Optimization Features
AAA Service
Day 3
L2TP
Cisco 10000 Series Router Hardware Overview
Cisco 10000 Series Router Software Overview
Version 1.0
vii
viii
Version 1.0
Course Introduction and Objectives
Overview
Description
This course is intended for customer technicians and system
integrators who need to implement various broadband aggregation
technologies on Cisco routers. This course also enables Cisco System
Engineers (SEs) to present and demonstrate various broadband
aggregation technologies on Cisco routers for customers. Students
learn about RBE, PPPoA, PPPoE, and L2TP, and learn how to
configure and verify operation of these technologies on Cisco routers.
This course also explains the Cisco 10000 Series router hardware
architecture and software features.
The course is instructor-led and includes hands-on lab exercises.
Lecture topics are reinforced with supporting student exercises.
This course focuses on implementing broadband aggregation
technologies on the Cisco 10000 Series router, however, most learning
experiences from this course may be applied to o ther Cisco routers
that support these technologies.
Objectives
After completing this course, you will be able to do the following:
Compare and contrast the various broadband aggregation

architectures available with Cisco routers
Explain how RBE and RFC 1483 routing work, describe their
typical architectures and benefits, and configure them on Cisco
routers
Explain how PPPoA and PPPoE work, along with descriptions of

their typical architecture and benefits, and configure them on
Cisco routers
Version 1.0
ix
Explain and configure various methods for optimizing subscriber

connections including PVC range, auto detect PPPoX
encapsulation, VC class, ATM PVC autoprovisioning, and BBA
groups
Explain AAA services available on Cisco routers and RADIUS

servers and configure AAA services on Cisco routers
Explain how L2TP works, describe its typical architecture and

benefits, and configure it on Cisco routers
Describe the Cisco 10000 Series router and explain the features
and functions of system-wide hardware and software components
Identify and describe system modules and services on the Cisco

10000 Series router that are utilized in broadband aggregation
deployment scenarios
Version 1.0
Contents
Course Overview ........................................................................................................... v
Course Agenda ............................................................................................................ vii
Course Introduction and Objectives........................................................................ ix

Overview...................................................................................................................... ix
Module 1 Broadband Aggregation Architectures ..........................................11

Overview................................................................................................................... 11
Broadband Aggregation Introduction ......................................................................... 12
Retail and Wholesale Services ................................................................................. 112
VC Service............................................................................................................... 116
ATM Bridging and Routing Methods ....................................................................... 118
PPP Review ............................................................................................................. 120
PPP Broadband Access Methods .............................................................................. 124
PTA......................................................................................................................... 126
L2TP ....................................................................................................................... 128
AAA ........................................................................................................................ 130
Managed LNS ......................................................................................................... 132
Remote Access into MPLS ....................................................................................... 134
SSG and SESM ....................................................................................................... 136
Summary ................................................................................................................ 140
Review Questions .................................................................................................... 141
Module 2 RBE and RFC 1483 Routing...............................................................21

Overview................................................................................................................... 21
Typical RBE Architecture.......................................................................................... 22
RFC 1483 Bridging Protocol Stack............................................................................. 24
How Does RBE Work? ............................................................................................... 28
RBE Configuration .................................................................................................. 212
RBE Advantages and Disadvantages ....................................................................... 218
Typical RFC 1483 Routing Architecture .................................................................. 222
RFC 1483 Routing Protocol Stack ............................................................................ 224
Version 1.0
xi
How Does RFC 1483 Routing Work? ........................................................................ 226

RFC 1483 Routing Configuration ............................................................................. 228
RFC 1483 Routing Advantages and Disadvantages ................................................. 232
Summary ................................................................................................................ 234
Module 3 PPPoA .....................................................................................................31

Overview................................................................................................................... 31
Typical PPPoA Architecture ...................................................................................... 32
PPPoA with PTA Protocol Stack ................................................................................ 36
PPPoA with Tunneling Protocol Stack ..................................................................... 310
How Does PPPoA Work with PTA? .......................................................................... 312
How Does PPPoA Work with Tunneling? ................................................................. 314
PPPoA IP Address Management.............................................................................. 316
PPPoA Configuration .............................................................................................. 318
PPPoA Advantages and Disadvantages ................................................................... 328
Summary ................................................................................................................ 332
Module 4 PPPoE......................................................................................................41
Overview................................................................................................................... 41
Typical PPPoE Architecture ...................................................................................... 42
PPPoE Protocol Stack................................................................................................ 46
How Does PPPoE Discovery Work?............................................................................ 48
PPPoEoA with PTA Protocol Stack .......................................................................... 410
PPPoEoA with Tunneling Protocol Stack ................................................................. 414
How Does PPPoE Work with PTA? .......................................................................... 416
How Does PPPoE Work with Tunneling? ................................................................. 418
PPPoE IP Address Management.............................................................................. 420
PPPoEoA Configuration .......................................................................................... 422
PPPoE Advantages and Disadvantages ................................................................... 434
PPPoEoE and PPPoEo892.1q................................................................................... 438
PPPoEoE and PPPoEo892.1q Configuration ............................................................ 440
Summary ................................................................................................................ 442
Module 5 Cisco Aggregation Optimization Features ....................................51

Overview................................................................................................................... 51
xii
Version 1.0
Optimization Features Introduction .......................................................................... 52

Minimizing ATM PVC Provisioning ........................................................................... 54
PVC Range................................................................................................................ 56
VC Class ................................................................................................................. 514
ATM PVC Autoprovisioning .................................................................................... 518
Autosense PPPoX Encapsulation ............................................................................. 522
PPPoE Profiles ........................................................................................................ 528
Summary ................................................................................................................ 532
Module 6 AAA Services.........................................................................................61

Overview................................................................................................................... 61
Introduction to AAA .................................................................................................. 62
Authentication .......................................................................................................... 68
Authorization .......................................................................................................... 610
Accounting .............................................................................................................. 612
AAA-Supported Protocols ........................................................................................ 614
RADIUS Attributes ................................................................................................. 616
Radius Files ............................................................................................................ 620
AAA Implementations ............................................................................................. 628
RADIUS Protocol..................................................................................................... 632
Cisco Implementation of AAA.................................................................................. 644
Troubleshooting Aids............................................................................................... 656
Cisco IOS Commands .............................................................................................. 658
UNIX Commands .................................................................................................... 670
Module 7 L2TP .........................................................................................................71

Overview ................................................................................................................... 71
L2TP Overview.......................................................................................................... 72
L2TP Components..................................................................................................... 74
L2TP Tunnel and Session Identifiers......................................................................... 76
Encapsulations Supported......................................................................................... 78
L2TP Message Format............................................................................................. 710
Incoming Call Sequence........................................................................................... 712
Forwarding PPP Frames ......................................................................................... 716
Call Disconnect Sequence ........................................................................................ 718
Typical L2TP Scenarios........................................................................................... 720
Version 1.0
xiii
L2TP Configuration Overview ................................................................................. 724

L2TP Tunnel Attributes .......................................................................................... 726
L2TP Configuration Without RADIUS..................................................................... 728
L2TP Configuration with RADIUS .......................................................................... 736
Tunnel Verification ................................................................................................. 750
Summary ................................................................................................................ 758
Module 8 Cisco 10000 Series Router Hardware Overview ..........................81

Overview ................................................................................................................... 81
Cisco 10000 Series Router Introduction ..................................................................... 82
Broadband Aggregation Deployment Scenarios ......................................................... 84
Cisco 10000 Series Router Components Overview...................................................... 86
Chassis Description ................................................................................................... 88
Modules Used with Broadband Aggregation ............................................................ 814
Cisco 10000 Series Router Architecture Overview ................................................... 818
Functional Block Diagram ....................................................................................... 820
Router Buffer Management ..................................................................................... 824
Router Backplane.................................................................................................... 826
Performance Routing Engine-2 ................................................................................ 830
PRE-2 Front Panel .................................................................................................. 832
PRE-2 Architecture ................................................................................................. 834
PRE-2 Packet Flow.................................................................................................. 842
PXF Technology and Operation ............................................................................... 850
PRE Comparison ..................................................................................................... 860
High Availability ..................................................................................................... 862
PRE Redundancy..................................................................................................... 864
Cisco 10000 Series Router Broadband Aggregation Line Cards................................ 874
ATM Line Cards ...................................................................................................... 876
ATM Line Card Common Features .......................................................................... 882
Assigning VPI/VCIs for ATM VC Scaling ................................................................. 888
LAN Line Cards ...................................................................................................... 892
Packet over SONET Line Cards..............................................................................8106
Common POS/SDH Line Card Features .................................................................8112
Summary ...............................................................................................................8114
Review Questions ...................................................................................................8115
xiv
Version 1.0
Module 9 Cisco 10000 Series Router Software Overview............................91

Overview................................................................................................................... 91
Software Architecture................................................................................................ 92
Software components................................................................................................. 94
Cisco 10000 Router Software ..................................................................................... 96
Supported Encapsulations ....................................................................................... 914
Frame Relay Support .............................................................................................. 918
Broadband Features and Scaling ............................................................................. 920
Leased-Line Features and Scaling ........................................................................... 928
High Availability and Management Functionality ................................................... 934
QoS Features and Functions.................................................................................... 936
Class-Map Match Options ....................................................................................... 938
Policy-Map Keywords .............................................................................................. 940
Policy-Map Actions .................................................................................................. 942
QoS Facts ................................................................................................................ 946
Policing Considerations ........................................................................................... 952
VC Scaling with QoS ............................................................................................... 954
System Status and Alarms ...................................................................................... 958
Checking the Data Path .......................................................................................... 966
System-Wide Statistics and Performance................................................................. 980
Summary ................................................................................................................ 996
Glossary .......................................................................................................................... 1
Technology Acronyms .................................................................................................... 2
Cisco 10000 Series Router Acronyms ............................................................................. 5
Appendix A Review Question Answers........................................................... A1

Appendix Contents .................................................................................................... A1
Module 1 Broadband Aggregation Architectures ..................................................... A2
Module 2 RBE and RFC 1483 ................................................................................. A4
Module 3 PPPoA ..................................................................................................... A7
Module 4 PPPoE ................................................................................................... A10
Module 5 Cisco Aggregation Optimization Features.............................................. A13
Module 7 AAA Services......................................................................................... A14
Module 7 L2TP ..................................................................................................... A16
Module 8 Cisco 10000 Series Router Hardware Overview ..................................... A18
Version 1.0
xv
Appendix B Router Starting Configurations ..................................................B1

Appendix Contents .................................................................................................... B1
P1R1 Configurations ................................................................................................. B2
P1R2 Configurations ............................................................................................... B16
P1R3 Configuration ................................................................................................. B30
Core Routers Configurations ................................................................................... B32
PC CPE Configurations ........................................................................................... B36
xvi
Version 1.0
Module 8
Overview
Description
In this module you learn about use of the Cisco 10000 Series Router
hardware in broadband aggregation implementations. This module
includes descriptions and capabilities of the chassis, PRE-2, and line cards
used with broadband aggregation, as well as functional block diagrams of
the hardware and packet processing.
Objectives
After completing this module, you will be able to do the following:
Describe how the Cisco 10000 router is used in typical broadband

deployments
Describe the Cisco 10000 router chassis components
Identify Cisco 10000 router functional components, interconnections,

and operation
Describe PRE-2 architecture and operation, including the route

processor, forwarding processor, and PXF
Trace the flow of a packet through the PRE-2
Describe the Cisco 10000 router high-availability hardware and

functions
Describe the features and functions of Cisco 10000 router line cards
used with broadband aggregation deployments
Version 1.0
81
Module 8
Cisco 10000 Series Router Introduction

The Cisco 10000 Series Router is an industry-leading service provider edge
aggregation router that provides high-performance IP services, maximum
platform scalability, and high availability.
Deployment Scenarios
The Cisco 10000 router uses the following networking technologies to
aggregate subscribers at the edge of the service provider network:
Leased-line edge aggregation
ATM
Frame Relay
Broadband aggregation
This training module focuses on the broadband aggregation utilization of

the Cisco 10000 router.
82
Version 1.0
Module 8
LEASED LINE
F
R
A
M
E
A
T
M
BROADBAND
Edge service aggregation support
Leased Line
ATM
Frame Relay
Broadband
Version 1.0
83
Module 8
Broadband Aggregation Deployment Scenarios

The Cisco 10000 router is ideally suited for aggregating broadband
subscriber connections including those subscribers that connect to the
router in a DSL environment.
Subscriber Connection Termination

The Cisco 10000 router may be used to terminate various types of
subscriber connections, including the following:
Route Bridge Encapsulation (RBE)
PPP over ATM (PPPoA)
PPP over Ethernet (PPPoE), including PPPoEoA, PPPoEoE, and

PPPoEo802.1q
PTA and Tunneling Support

Subscriber connections may be terminated on an aggregation router and
routed to the final service destination or tunneled to a service destination
using the Cisco 10000 router. When used with tunneling, the following
methods may be used:
Layer 2 Tunneling Protocol (L2TP)
Remote Access to Multiprotocol Label Switching (RA-MPLS)
Any Transport over MPLS (AToM) (future support)
When using L2TP, the Cisco 10000 router may be deployed as both the
L2TP access concentrator (LAC) and L2TP network server (LNS).
84
Version 1.0
Module 8
PPP sessions
PE
Internet
PTA
Termination
VPN
LS
P
M
ISP One.com
L2TP
L2TP LNS
Internet
oM
AT
PPP sessions
Cisco 10000:
A universal
broadband
access server
PE
ISP Two.com
ATM VCs
ISP Three.com
Version 1.0
85
Module 8
Cisco 10000 Series Router Components Overview

This section provides an overview of the Cisco 10000 router chassis and its
major components, as well as an introduction to the processing engine and
line cards used to support broadband aggregation.
86
Version 1.0
Module 8
Chassis Description
Chassis Components
Modules for Broadband Aggregation
Line Cards
Processor
Version 1.0
87
Module 8
Chassis Description
The Cisco 10000 router is a 10-slot router optimized to meet the broadband
and leased-line aggregation requirements of ISPs. The Cisco 10000 router
contains two dedicated slots: one for an active Performance Routing Engine
(PRE), and one for a redundant (standby) PRE.
The remaining eight slots are for interface modules. The interface modules
can be placed in any slot. The router supports both full- and half-height
hot-swappable interface modules.
Chassis
The chassis is NEBS Level 3 compliant and meets Telcordia Technologies
specifications for temperature extremes, vibration, earthquake,
electromagnetic interference, fire safety, electrostatic discharge , and so on.
It also provides redundant external alarms for service provider
deployments.
_____________________________ Note __________________________
The chassis supports the external timing connector; however this
functionality is not a part of the line cards or processing engine and
therefore not used.
____________________________________________________________
Power
The Cisco 10000 router supports redundant power entry modules (PEMs).
The modules may be either AC or DC but not a mixture of both.
Cooling
The Cisco 10000 router cooling system uses redundant fans with a load sharing design and supports front-to-back airflow.
88
Version 1.0
Module 8
Chassis Description
Chassis Description
Blower
Module
Connector
External
Timing
Connector
(not used)
Blower
Module
External
Alarm
Connector
T3/E3
PEMs
DC Power
Connector
Air Filter
PREs
Front View
Rear View
Version 1.0
89
Module 8
Chassis Description (continued)

Power Entry Modules
The Cisco 10000 router has the option to use redundant PEMs. This option
provides for high availability because only one module is required.
When two PEMs are installed, the DC units operate in a diode-shared
mode while the AC modules are load-shared. The failure of a single hotswappable PEM does not cause disruption of system operation because the
other PEM takes over. These units distribute 48V within the chassis; while
each circuit card contains its own onboard regulation.
The illustration depicts both PEMs and provides tables that describe the
functions of their LEDs.
810
Version 1.0
Module 8
Chassis Description
Power Entry Modules
Status LEDs
Status LEDs
DC PEM
Status LED
Status
Description
Power
Green
Power is on
Off
No power
Yellow
Power is off
or
Wiring not
properly
connected
Fault
Miswired
On
-48V and
return wires
are reversed
Version 1.0
AC PEM
Status LED
Status
Description
Power
Green
Power is on
Off
No power
Yellow
Power is off
Fault
or
Replace PEM
811
Module 8
Chassis Description (continued)

Blower Module
Cooling
The cooling system uses redundant fans with a load -sharing design housed
in a blower module. A fan failure generates an interrupt to the PRE for the
syslog. In addition, there are thermal sensors on the PRE to back up a fan
failure signal. The unit can operate for an indefinite period with a single
fan failure; although this results in the loss of cooling redundancy.
Blower Module Replacement
The blower module is easily replaced. If a blower module requires servicing

from a fan failure, the system can operate without fans and without
overheating for 2 minutes, which is ample time to hot-swap the blower
assembly.
Critical Temperature Alarm
The default action for a critical temperature alarm is to continue operation.

If you want the system to shutdown on this alarm condition, you must
configure the router to do so.
To control the action of the Cisco 10000 router when the air intake or core
temperature reaches a critical temperature condition, use the following
commands:
(config)#facility-alarm intake-temperature critical exceedaction shutdown
(config)#facility-alarm core-temperature critical exceed-action
shutdown
812
Version 1.0
Module 8
Chassis Description
Blower Module
Status LEDs
Fan Status LEDs
Status
Fan OK
Green
Fan Failure
Yellow
Multi-Fan Failure
Yellow
Version 1.0
813
Module 8
Modules Used with Broadband Aggregation

The Cisco 10000 router has a wide selection of line card and processor
modules that may be used for service provider edge router aggregation
with leased-line, ATM, Frame Relay, and broadband aggregation
installations. These modules enable the Cisco 10000 to provide highperformance IP services, maximum platform scalability, and high
availability.
The following types of modules are used with broadband aggregation
deployments.
ATM Modules
The Cisco 10000 ATM line cards are especially suited for aggregating ATM
subscriber connections from DSLAMs. The line cards provide both the
performance and the port density to scale networks efficiently and reliably.
They feature a high-performance segmentation and reassembly (SAR)
adapted for various applications, including advanced traffic management,
cell scheduling, and integrated buffer management.
8xE3/DS3 ATM card 8-port E3/DS3 ATM line card
4xOC-3 ATM card 4-port OC3c/STM-1 ATM line card operating in

SONET or SDH mode
1xOC-12 ATM card 1-port OC12 ATM line card operating in SONET
or SDH mode
POS Modules
The Cisco 10000 packet over SONET (POS) line cards are high-capacity,
high-performance line cards that enable service providers to offer
dedicated Internet access services and peering to other service providers.
814
6xOC-3c/STM-1 POS/SDH card 6-port OC-3c/STM-1 line card

operating in POS or SDH mode
1xOC-12/STM-4 POS/SDH card 1-port OC-12/STM-4 line card

operating in POS or SDH mode
1xOC-48c/STM-16 POS/SDH card 1-port OC-48c/STM-16 line card

operat ing in POS or SDH mode
Version 1.0
Module 8
ATM Cards
4xOC-3 ATM, 1xOC-12 ATM,

8xE3/DS3 ATM
POS Cards
6xOC-3 POS, 1xOC-12 POS, 1xOC-48 POS
LAN Cards
GigE, GigE HH, 8xFE HH
Processor
PRE-2
Version 1.0
815
Module 8
Modules Used with Broadband Aggregation (continued)

LAN Modules
The LAN modules enable service providers to aggregate subscriber
connections and connect to ISPs using Ethernet technology. DSL access
multiplexers (DSLAMs) can now aggregate subscriber connections using
Ethernet to the aggregation router. These LAN modules may also be used
to connect to backbone or intra-POP routers.
Gigabit Ethernet card 1-port, full-height Gigabit Ethernet line card
Gigabit Ethernet half-height card 1-port, half-height Gigabit

Ethernet line card
8-port Fast Ethernet half-height card
PRE-2
The Cisco 10000 router Performance Routing Engine 2 (PRE-2) is the nextgeneration route processor for the Cisco 10000 router. Using the PRE-2,
the Cisco 10000 delivers line -rate performance for more than 60,000
simultaneous sessions with critical per-subscriber-service features enabled,
such as security and traffic policing.
The PRE-2 enables the Cisco 10000 router to provide full PPP termination
and aggregation (PTA), LAC and LNS functionality.
_____________________________ Note __________________________
Prior to the PRE-2, the Cisco 10000 using the PRE-1 was ideally suited
for leased-line services or as an LNS in broadband aggregation
deployments.
____________________________________________________________
816
Version 1.0
Module 8
Modules Used with Broadband Aggregation (continued)
ATM Cards

8xE3/DS3 ATM
POS Cards
LAN Cards
Processor
PRE-2
Version 1.0
817
Module 8
Cisco 10000 Series Router Architecture Overview

The pages that follow provide an overview of the Cisco 10000 router
architecture. The following topics are presented:
818
Functional Block Diagram
Buffer Management
Router Backplane
Version 1.0
Module 8

Buffer Management
Router Backplane
Version 1.0
819
Module 8

The major functional components that make up the Cisco 10000 router are
the PRE, Iron Bus, and line c ards.
PRE-2
The PRE is designed for reliability and high availability. It uses an
advanced route processor redundancy (RPR) feature for automatic failover.
The PRE is composed of two main sections, a route processor and a
forwarding processor.
Route Processor
The route processor p rovides standard Cisco IOS functionality for:
Chassis management
System initialization
Routing protocol updates
Route processor r edundancy (RPR)
CLI and SNMP functionality
Forwarding Processor
The forwarding processor provides the following IP functions:
IP forwarding
Packet buffering
Layer 3 features
QoS features
Three main components make up the forwarding processor:
820
Parallel eXpress Forwarding (PXF) engine a 2-dimensional array of

64 CPUs that forward IP packets. Processor-intensive tasks such as
policy routing, quality of service (QoS), and statistics collection are
segmented and distributed to columns of multiple processors.
Packet Buffers buffer packets processed by the PXF engine .
Cobalt ASIC provides buffer management and Iron Bus data flow
control. It controls the flow of packets from the line c ards to the
forwarding processor, ensuring that the PXF does not become
overloaded. In addition, it manages the queuing and dequeuing of
packets to the 256MB packet buffers under the direction of the PXF.
Version 1.0
Module 8
Routing protocols
CLI
SNMP
Chassis management
Initialization
Route
processor
IP forwarding
L3 features
Packet buffering
QoS features
PRE-2
PXF
Parallel eXpress Forwarding engine
Cobalt buffer
and I/O control
Iron Bus
Media interface
Framers and multiplexers
Facility data links
Link-level clocking
SAR (for ATM only)
Line card
Packet
buffers
Version 1.0
Line card
Line card
821
Module 8
Functional Block Diagram (continued)

Line Cards
The system supports up to 8 full-height or 16 half-height line cards.
Because all packet forwarding is done on the PRE, the line cards are
required only to perform basic packet manipulation and packet buffering,
and to send the packet across the high-speed backplane to the PRE. These
functions include:
Media interface
Framing
Multiplexing
Link-level clocking
Facility data links
Segmentation and reassembly (SAR) for ATM line cards
Iron Bus
The Iron Bus is the systems primary data path between line cards and the
PRE-2. It is composed of point-to-point links from each PRE to each line
card half-slot.
822
Version 1.0
Module 8
Functional Block Diagram (continued)
Routing protocols
CLI
SNMP
Chassis management
Initialization
Route
processor
IP forwarding
L3 features
Packet buffering
QoS features
PRE-2
PXF
Parallel eXpress Forwarding engine
Cobalt buffer
and I/O control
Iron Bus
Media interface
Framers and multiplexers
Facility data links
Link-level clocking
SAR (for ATM only)
Line card
Packet
buffers
Version 1.0
Line card
Line card
823
Module 8
Router Buffer Management

Packet buffering and queuing are the mechanisms by which a router stores
packets when transient overloads occur; these mechanisms play a key role
in implementing QoS features.
Input Buffering
Line cards contain large input buffers that absorb transient overloads,
reducing the possibility of spurious packet loss. Under times of heavy load ,
the PRE-2 applies backpressure to the line card, and its buffers absorb the
overhead.
Note that large input buffers add latency, while very small buffers can
result in increased packet loss. The input buffers operate in FIFO mode
only.
Output Buffering and Queuing

The Cisco 10000 router primary output buffer pool is located on the PRE-2.
QoS features are provided through PXF management of this buffer.
The PXF marks outbound packets, based on their destination interface and
QoS priority, and then forwards them to the output buffer pool. The output
scheduler selects these packets for transmission, based on QoS parameters
and their destination interface.
To ensure a predictable output, line card queues are kept short, and flow
control is used on the data path between the output interface and the PXF
output scheduler.
824
Version 1.0
Module 8
Input buffers hold

packets if L3
forwarding
is congested
Input pipelining buffers apply

backpressure to line cards
if L3 is overloaded
TX
Buffers
QoS
Scheduler
TX
Buffers
Output Packet
Buffer
Line Card
PRE
Packet flow
Backpressure
Route Lookup
& L3 features
Input Buffering
RX
Buffers
Round Robin
Scheduler
RX
Buffers
Line Card
256-MB packet
buffers on PRE
Small output buffers

on line cards for high
speed pipelining
Drains input buffers

faster than trunk speed
Version 1.0
825
Module 8
Router Backplane
The Cisco 10000 backplane is composed of multiple backplane connections:
Backplane Ethernet for control path and Iron Bus links for data paths.
Control Paths
The backplane Ethernet provides the control or maintenance path for PREto-line card communication. This backplane provides functions such as the
following:
Line card discovery
Interface configuration
Card reset
Network management
Debug
Inter-PRE communication is handled by a RPR bus that performs the

following functions:
Synchronization of running and startup configuration files
Time-of-day synchronization
Keepalive messages between redundant PREs
Iron Bus
The Iron Bus is the primary data path between the PREs and line cards for
the Cisco 10000 system. It is star-wired from each PRE (Cobalt 2 ASIC) to
each line card (Barium ASIC). Independent Iron Bus paths from each line
card to each PRE provide enhanced high system availability (EHSA)
functionality. As a result, PRE cut-over does not require bus or line card
reset.
Using the PRE-2, each Iron Bus is capable of operating at 3.2 Gbps per
full-height slots and 1.6 Gbps per half-height slot in each direction; that is,
6.4 and 3.2 Gbps full-duplex, respectively. This architecture avoids the
problem of backplane oversubscription.
The Iron Bus uses a packet-oriented serial protocol with link-level and
channel-level flow control. As a result, the actual bandwidth available to a
line card is slightly less than the full maximum bandwidth indicated here.
826
Version 1.0
Module 8
Router Backplane
Router Backplane
Point-to-Point
IronBus links
between line
cards and PRE
PRE-A
RPR bus PRE-B
Cobalt ASIC
Cobalt ASIC
T
Each line card
connects to
both PREs for
redundancy
Barium ASIC
Barium ASIC
Line Card 1/0
Line Card 1/1
Backplane
Ethernet
provides
control plane
communication
Version 1.0
...
Barium ASIC
Line Card 8/1
827
Module 8
Router Backplane (continued)

Iron Bus Modes
Each of the eight line card slots in the Cisco 10000 chassis has eight pairs
of communication lines to each PRE slot. With the Cobalt 2 ASIC, each
pair of lines is clocked at 400 Mbps, in each direction, for a total of
3.2 Gbps bidirectional.
Bound versus Unbound Mode
The Cobalt 2 ASIC can adjust for half-height or full-height line cards on a
per-slot basis.
Bound mode uses all eight pairs of connections in a full slot as a single
group. This means that full-height line cards can receive a maximum of
3.2 Gbps in each direction.
Unbound mode is typically used for half-height line cards. In this mode,
the eight pairs of line are split up into two groups of four pairs e ach.
Each group services a half-height line card to provide up to a maximum
of 1.6 Gbps of data transfer in each direction.
Wide versu s Narrow Mode
The Cobalt 2 ASIC operates in either wide mode or narrow mode. This is a
function of the bandwidth requirements of the line card inserted into the
half-slot.
828
Wide mode uses all four pairs of communication lines per half slot, for a
total of 1.6 Gbps in each direction per half slot.
Narrow mode is supported for lower speed interfaces and uses only two
pairs of communication lines. This results in 800 Mbps in each
direction per half slot.
Version 1.0
Module 8
Router Backplane
Iron Bus Modes
Full Height (Bound Mode)
Line
Card
Line
Card
PRE
Half Height (Unbound Mode)
PRE
Line
Card
3.2 Gbps
each way per full slot
Narrow
or Wide
Mode
1.6 Gbps
each way per
half slot
Total Backplane Bandwidth
Total Backplane Bandwidth
3.2 Gbps * 8 slots

* 2 directions
1.6 Gbps * 16 half slots

* 2 directions
= 51.2 Gbps
= 51.2 Gbps
Version 1.0
829
Module 8
Performance Routing Engine-2

The pages that follow explain various features and functions of the PRE-2:
830
PRE-2 Front Panel
PRE-2 Architecture
PRE-2 Packet Flow
PXF Technology and Operation
PRE Comparison
Version 1.0
Module 8
PRE-2 Front Panel

PRE-2 Architecture
PRE-2 Packet Flow
PRE Comparison
Version 1.0
831
Module 8
PRE-2 Front Panel

The PRE-2 front panel has multiple components and LEDs for managing
the Cisco 10000 router. The LEDs provide quick indications of the stat us of
the PRE .The LEDs are identified and described in the graphic that
follows.
The following components are on the PRE-2 front panel:
832
Console/Aux ports provide management access using serial interfaces.
Ethernet port provides out-of-band management access using a Fast

Ethernet interface.
PCMCIA two PCMCIA disks may be inserted for storing Cisco IOS
images, configuration files, and log files.
Status Windows provide status of the PRE-2, such as IOS RUN on

the active PRE and IOS STBY on the standby PRE. Other indications
are given as the PRE is booting.
ACO Alarm Cut-Off switch is used to shut off an external alarm.
Status and Fail LEDs indicates boot and failure status.
Version 1.0
Module 8
PRE-2 Front Panel
PRE-2 Front Panel
Aux
Console
Ethernet
LED
Status
Description
Ethernet port Activity LEDs
Green
Packets are being transmitted and received
Off
No activity
Green
Carrier detected, the ports able to pass

traffic
Off
No carrier detected, ports are not able to

pass traffic
PCMICA slot 0
Green
Slot 0 active
PCMICA slot 1
Green
Slot 1 active
Critical, Major and Minor LEDs
Off
No alarm
Yellow
Indicates an alarm condition
Ethernet port Link LEDs
Alarm LEDs
ACO
Status
Fail
Alarm Cutoff (ACO) switch
PCMICA
Status
Status
Windows
Fail
Pressing the switch disables the audible

alarm
Flashing yellow
System is booting
Green
PRE is active (primary)
Flashing green
PRE is standby (secondary)
Off
No power to PRE
Yellow
A major failure has disabled the PRE
Off
PRE is operating properly
Version 1.0
833
Module 8
PRE-2 Architecture
The PRE-2 hardware is divided into two logical and physically separate
components:
Route processor
Forwarding processor.
Route Processor
The primary function of the route processor is to manage the system and
build the tables necessary for the forwarding processor to make forwarding
decisions.
The primary functions include:
Receiving all routing updates and building the Forwarding Information

Base (FIB) tables for the system
Managing the PXF and line card microcode that is bundled with the
main Cisco IOS software image
Monitoring system components to ensure proper functioning
The route processor contains the following components:
Main CPU
2 MB of NVRAM
System controller
Time-of-day clock
64 MB of bootflash
Support for 48-MB or 128-MB flash disk
100-Mbps Fast Ethernet for net management
Fast Ethernet connection for the RPR bus through the backplane
Fixed 1 GB of ECC SDRAM
Ethernet management backplane
Main CPU
Control plane processing on the Cisco 10000 router PRE-2 is handled by a

generic RM7000B CPU. In the PRE-2 implementation, this CPU operates
at 500 MHz with internal Layer 1 and Layer 2 caches (32 KB and 256 KB,
respectively) as well as a 4-MB external Layer 3 cache.
834
Version 1.0
Module 8
PRE-2 Architecture
PRE-2 Architecture
MIPS
1-GB
ECC Column
Memory
RM7000B
500 MHz
T3
T3
T3
System
Controller
IOS
DRAM
64 154-MHz
PXF Processor
Elements
Flash,
NVRAM,
Ether, etc.
Cobalt 2 ASIC
Route
Forwarding
Processor Processor
1024 MB of
ECC DRAM
T3
Line
Cards
1.6/3.2 Gbps
line card
interconnect
Control
Control
SDRAM
Buffer
Buffer
SDRAM
256-MB ECC
Packet Buffer
Version 1.0
835
Module 8
PRE-2 Architecture (continued)

Route Processor (continued)
System Controller
The system controller acts as the traffic officer for the route processor. The
system controller includes many of the functional components necessary to
communicate with the devices on the route processor and within the
system.
The system controller provides the following components and functions:
Integrated memory controller that provides a direct memory access

(DMA) channel for accessing memory
Dual serial interfaces that connect to the console and auxiliary

connections
Two built-in Ethernet MAC addresses that connect to the PRE-2 and to
the Ethernet port on the front panel
Two PCI buses that provide configuration and control of the PXF
complex and some other onboard components
PCI Bus 0
The following functions occur over PCI Bus 0:
836
Downloading of PXF microcode to the PXF CPUs on system startup or

microcode reloads.
The RM7000B CPU keeps the PXF CPUs forwarding information

continuously updated.
Provides the connection to an onboard Ethernet controller for

connection to the backplane Ethernet (BPE). The BPE is used to
configure and monitor all line cards in the system.
Version 1.0
Module 8
PRE-2 Architecture
MIPS
1-GB
ECC Column
Memory
RM7000B
500 MHz
T3
T3
T3
System
Controller
IOS
DRAM
64 154-MHz
PXF Processor
Elements
Flash,
NVRAM,
Ether, etc.
Cobalt 2 ASIC
Route
Forwarding
Processor Processor
1024 MB of
ECC DRAM
T3
Line
Cards
1.6/3.2 Gbps
line card
interconnect
Control
Control
SDRAM
Buffer
Buffer
SDRAM
256-MB ECC
Packet Buffer
Version 1.0
837
Module 8

Route Processor (continued)
PCI Bus 1
PCI Bus 1 operates at twice the speed of PCI Bus 0. This bus connects
between the system controller and the Cobalt 2 ASIC on the forwarding
processor.
PCI Bus 1 is the punt path between the route processor and forwarding
processor. Punting a packet occurs when the forwarding processor must
send a packet to the route processor for handling. This could be as simple
as a packet destined for the router itself, such as a routing update, or
something that the PXF does not know how to forward. The punt path is
not used for most IP forwarding because it is less efficient than using the
forwarding processor.
Route Processor Local Resources
Several components on the route processor perform specific functions

without any connection to other components outside of the route processor.
838
I/O field-programmable gate array (FPGA) acts as a gateway with a

low-speed connection to several devices, including NVRAM, bootflash,
ID EEPROM, and LEDs. Another function of the I/O FPGA is
management of all of the environmental monitoring, watchdog timers,
PRE-2 interrupt handling, and redundancy signaling to the secondary
PRE-2.
Bootflash 64-MB of bootflash is large enough to handle multiple

copies of the Cisco IOS software images and still have space for log files
and stored configurations. This component is fixed and is not
upgradeable or field replaceable.
NVRAM 2 MB of NVRAM for storing system configuration

information.
ID EEPROM provides 4 KB of storage for board identification, voltage

monitoring, and reset functions.
Version 1.0
Module 8
PRE-2 Architecture
MIPS
1-GB
ECC Column
Memory
RM7000B
500 MHz
T3
T3
T3
System
Controller
IOS
DRAM
64 154-MHz
PXF Processor
Elements
Flash,
NVRAM,
Ether, etc.
Cobalt 2 ASIC
Route
Forwarding
Processor Processor
1024 MB of
ECC DRAM
T3
Line
Cards
1.6/3.2 Gbps
line card
interconnect
Control
Control
SDRAM
Buffer
Buffer
SDRAM
256-MB ECC
Packet Buffer
Version 1.0
839
Module 8

The forwarding processor handles all packet forwarding for the system and
consists primarily of the following:
Field-Programmable Gate Array (FPGA).
Four Toaster 3 PXF application-specific integrated circuits (ASICs)
Cobalt 2 ASIC
Packet forwarding is handled by the Cobalt 2 ASIC and the PXF ASICs.
Fast Packet FPGA
The FPGA (not shown in the graphic) provides the following functions:
Allows the route processor to program the PXF CPUs packet handling
function using PCI Bus 0
Aggregates and controls Cobalt 2, PXF, and line card interrupts and
resets
Monitors line card ready signals to aid in detecting online insertion and
remo val (OIR) events in the chassis
Toaster 3 ASIC (PXF)
Collectively, the four Toaster 3 ASICs are the PXF engine. The PXF is
responsible for all packet processing and forwarding in the Cisco 10000
router.
Internal to the Toaster 3 ASICs are individual CPUs operating a 154 MHz.
The CPUs are arranged into eight columns and eight rows, for a total of 64
CPUs.
Each column of CPUs has its own dedicated 128 MB column memory, in
which are stored the data structures and lookup tables needed for packet
processing and forwarding. A total of 1 GB of column memory resides on
the PRE-2.
Cobalt 2 ASIC
The Cobalt 2 ASIC is a Cisco custom-designed ASIC with two major

functions.
840
Provides the serial connections to all 16 of the line card half-slots
Manages packet flow to and from the PXF, as well as managing packet
buffer memory and any punting to the route processor.
Version 1.0
Module 8
PRE-2 Architecture
MIPS
1-GB
ECC Column
Memory
RM7000B
500 MHz
T3
T3
T3
System
Controller
IOS
DRAM
64 154-MHz
PXF Processor
Elements
Flash,
NVRAM,
Ether, etc.
Cobalt 2 ASIC
Route
Forwarding
Processor Processor
1024 MB of
ECC DRAM
T3
Line
Cards
1.6/3.2 Gbps
line card
interconnect
Control
Control
SDRAM
Buffer
Buffer
SDRAM
256-MB ECC
Packet Buffer
Version 1.0
841
Module 8
PRE-2 Packet Flow

Incoming Packets
Inbound p ackets from a line card enter the PRE-2 forwarding processor
through the Iron Bus to the Cobalt 2 ASIC. The Cobalt 2 stores the packet
in high-speed, internal packet memory and removes the 64-byte packet
header. The packet header is then mated with a PXF control field, called a
context, and sent to the first device in the PXF.
Packet Processing
When the packet header and context enter the first Toaster 3 ASIC, they
are assigned to one of eight rows. Row assignment is done in a strict roundrobin fashion. The choice of row makes no difference to the packet header
context combination because all CPUs in a column provide exactly the
same function. All packet headers must go through all eight columns
within a row.
Resultant Operations
After the packet header makes its way through all eight columns of a row,
it will be marked for one of four operations. These operations are:
Forward
Feedback
Punt
Drop
The modified packet header is then sent back to the Cobalt 2 ASIC to be
buffered for future action, that is forward, feedback, punt or drop.
842
Version 1.0
Module 8
PRE-2 Packet Flow
PRE-2 Packet Flow
2) Headers pass
FCRAM
FCRAM
FCRAM
FCRAM
through
Toaster 3 ASIC
3) Modified packet
OUT
IN
Toaster 3
Toaster 3
Toaster 3
Cobalt 2
ASIC
To Toaster
Complex
Input Packet
Memory
Toaster 3
From Toaster
Complex
Output Queue
Controller
Iron Bus Interface

1) Packet
enters from
line card
headers and
packet bodies are
moved into packet
buffer memory
Packet Buffer
SDRAM
4) Complete packets
are moved from
SDRAM to output
line cards
From Line Cards
To Line Cards
Version 1.0
843
Module 8
PRE-2 Packet Flow (continued)

Forward
The Cobalt 2 ASIC reads queue and dequeue information from the packet
context and then queues the packet header with the packet body from the
input packet memory to the external packet buffer. The packet is dequeued
and sent to the appropriate line card by the Cobalt 2 ASIC based on
scheduler instructions.
Feedback
When a packet requires more processing than can be accomplished in a

single pass through all eight columns, it is set up for feedback. During this
operation, the Cobalt 2 ASIC queues the packet header with the packet
body from the input packet memory to the external packet buffer. The
packet header with context is returned to the PXF for a second pass
through all eight columns. When processing is completed, the header is
joined with the remainder of the packet that is queued in the external
packet buffer.
Punt
Certain traffic types such as routing updates, Local Management

Interface (LMI), PPP control, and Simple Network Management Protocol
(SNMP) are handled by the route processor. These packets are first
processed by PXF where its determined that the traffic is destined for the
route processor. The packets context is then marked for punting. The
Cobalt 2 ASIC treats the route processor as any other interface and queues
the packet header and body in the external packet buffer for transfer
across PCI Bus 1 to the route processors system controller.
_____________________________ Note __________________________
Punting is sometime referred to as diversions.
____________________________________________________________
Drop
Some packets may be unroutable or may have a IP CRC error or some

other sort of error in which case, the header context for these packets is
flagged for drop. When these packets are advanced to the Cobalt 2 ASIC,
the packet headers and pointers to the corresponding locations in input
packet memory are cleared.
844
Version 1.0
Module 8
PRE-2 Packet Flow
2) Headers pass
FCRAM
FCRAM
FCRAM
FCRAM
through
Toaster 3 ASIC
3) Modified packet
OUT
IN
Toaster 3
Toaster 3
Toaster 3
Cobalt 2
ASIC
To Toaster
Complex
Input Packet
Memory
Toaster 3
From Toaster
Complex
Output Queue
Controller
Iron Bus Interface

1) Packet
enters from
line card
headers and
packet bodies are
moved into packet
buffer memory
Packet Buffer
SDRAM
4) Complete packets
are moved from
SDRAM to output
line cards
From Line Cards
To Line Cards
Version 1.0
845
Module 8

_____________________________ Note __________________________
Not all features and functions may be available in all releases.
____________________________________________________________
PRE-2 Packet Flow Single-Pass Operations

The following operations occur with a single pass of the packet header
through the PXF.
846
Layer 2 classification This includes PPPoA, PPPoE, RBE, RFC 1483,

Frame Relay, POS, Gigabit Ethernet, serial and channelized
connections
CEF switching/ Reverse Path Forwarding (RPF) Strict RPF requires

only a single pass
Queue/dequeue PXF queue/dequeue commands require only one pass
I/O statistics collection This includes Layer 2, QoS, access control

list (ACL), and routing statistics
Input ACLs Input ACLs require a single pass unless they are too
large to fit in a single turbo ACL table
Load balancing Per-packet load balancing is a single-pass feature

after Cisco IOS release 12.0(22)S
Input or o utput QoS normally a single-pass operation; exceptions are

given under the Multipass heading
MPLS traffic engineering (TE), Virtual Private Network (VPN) and tag
switching These functions require only a single pass as long the tags
do not represent an aggregate route
QoS policy propagation through the Border Gateway Protocol (QPPB)
L2TP Traffic from an LNS requires a single pass on the LAC
Multilink PPP (MLP) Packets that are received in sequence do not

require feedback
Policy based r outing
Version 1.0
Module 8
PRE-2 Packet Flow
PRE-2 Packet Flow Single-Pass Operations
Single Pass Operations
L2 classification (PPPoA,
PPPoE, RBE, RFC 1483,
Frame Relay, POS, GigE,
serial, channelized)
RPF strict only

Queue/dequeue
I/O statistics collection
Input ACL security
Single Pass Operations

Input or output QoS
MPLS TE, VPN (no
aggregate route), tag
switching
QPPB
L2TP (on LAC)
MLP (received packets in
order)
Policy-based routing
Load balancing
Version 1.0
847
Module 8

PRE-2 Packet Flow Multipass Operations
The following operations require multiple passes through the PXF:
Multicast traffic One feedback for setup and one for teardown, plus
one per packet replicated
IP fragmentation A feedback is required for each fragment
MLP received out of sequence packets and outgoing packets A

feedback is required for each out-of-order received packet and each
packet transmitted in an MLP bundle
Output or split ACL ACLs applied to an output interface, or ACLs

that are too big to fit in a single turbo ACL table
Input and o utput QoS/split If a packet will be subject to QoS on both

the input and output interfaces, it will require a feedback. If the QoS
criteria is too large to fit in a single table, a feedback will be required
MPLS aggregate de-encapsulation In this environment both a tag

information base (TIB) and Forwarding information Base (FIB) lookup
is required, so a feedback is required
NetFlow accounting Accumulating statistics requires a feedback, as

does sending NetFlow export packets. Four feedbacks are required to
age an entry for a new entry
Generic route encapsulation (GRE) tunneling
L2TP LNS functionality and tunneling switching each require two

passes
Layer 2 header > 48 bytes If the Layer 2 header including tag stack
exceeds 48 bytes a feedback is required
ICMP responses ICMP response generated by the PXF require two

passes: one pass to determine that the PXF is unable to perform a
function, such as cant fragment, and a second pass to generate the
ICMP response
PRE-2 Packet Flow Punting Operations

Locally destined traffic to the route processor can be anything from routing
updates and keepalives to management traffic, such as Integrated Link
Management Interface (ILMI), LMI, and Telnet. Echo replies require route
processor functionality as does punting for an IP address during adjacency
establishment.
848
Version 1.0
Module 8
PRE-2 Packet Flow
PRE-2 Packet Flow Multipass and Punting Operations
Multipass Operations
Multipass Operations
GRE tunneling (on de-encap)
LT2P: tunnel switch, LNS
Layer 2 header > 48 bytes
Multicast traffic
IP Fragmentation
Strict RPF after 12.0(22)S
(includes tag stack)
MLP output sequence

numbers
ICMP responses: cant

fragment, echo request, TTL
MLP Input on out of sequence

only
Output or a split ACL

Input and output QoS or a
split QoS table
MPLS Aggregate de-encap

(TIB then FIB)
Netflow accounting
Punting Operations
Locally directed traffic
Glean adjacency (punt for IP
address during adjacency
establish)
Echo reply Response to
ping
Version 1.0
849
Module 8

PXF Technology Overview
Parallel eXpress Forwarding (PXF) is a powerful adaptive networkprocessing technology that balances maximum forwarding performance
with a flexible feature set. PXF on the Cisco 10000 router enables multiple
million packet-per-second forwarding.
PXF makes use of the expedited IP lookup and forwarding algorithms
introduced with Cisco Express Forwarding, while offering expanded
functionality and accelerated performance through the implementation of a
parallel architecture. Using an array of CPUs, the PXF processor applies
the combination of parallel processing and pipelining techniques to the
Cisco Express Forwarding algorithms to efficiently handle a variety of
complex services and operations.
Benefits of PXF Technology

Implementation of PXF technology in the Cisco 10000 offers many benefits,
including the following:
850
Forwarding processor focuses on providing extremely fast, high-touch

packet processing in hardware.
Forwarding processor makes use of reprogrammable PXF technology in

combination with custom ASIC forwarding.
PXF provides many benefits of and advantages o ver ASIC-based

forwarding, such as
High throughput
Low latency with high-touch packet services
Shorter time to market for new features
Easier fixing of problems
PXF CPUs run custom microcode that is downloaded at boot time and
can be rewritten with every Cisco IOS software image change .
New features are easy to code into hardware, and any problems that
are found are easy to remedy with a new Cisco IOS software image.
Version 1.0
Module 8
PXF a powerful adaptive network-processing

technology
Benefit of PXF technology
Extremely fast packet processing in hardware

Reprogrammable PXF technology
High throughput
Low latency
Shorter time to market for new features
Easier fixing of problems
PXF CPUs run custom microcode
New features easy to code into hardware
Version 1.0
851
Module 8
PXF Technology and Operation (continued)

PXF Components
The primary components of the PXF are the Toaster 3 ASIC and column
memory.
Toaster 3 ASIC
Internal to each Toaster 3 ASIC is an array of 16 individual CPUs

arranged into two columns of eight CPUs each. Together, the four Toaster
3 ASICs provide eight columns of CPUs, for a total of 64 CPUs.
Each column performs specific functions as it processes packets.
Sometimes a packet cannot be fully processed in one pass. In this instance,
the packet is fed back through all eight columns in the PXF array for a
second pass.
Column Memory Usage
Each column of CPUs has its own dedicated 128 MB of ECC-protected Fast
Cycle RAM (FCRAM) column memory, where the data structures and
lookup tables needed for packet processing and forwarding are stored.
When the route processor sends a forwarding table update to the fast
packet FPGA, the column memory will store the forwarding information
for use by that specific PXF column.
Because each column of CPUs in the PXF can be programmed to provide a
specific series of functions, the column memory for each column will be
unique. However, because all eight CPUs in any column will be providing
the same packet-processing features, they can share a single column
memory so that only one copy of the forwarding information for any feature
needs to be stored.
852
Version 1.0
Module 8
PXF Components
Toaster 3 PXF
ASIC
Toaster
Toaster 33 PXF
PXF
ASIC
Toaster 3 PXF
ASIC
ASIC
Toaster 3 PXF
ASIC
Current Context
Column Column
Memory Memory
Next Context
Column Column
Memory Memory
Column Column
Memory Memory
IM
IM
IM
IM
IM
IM
IM
IM
IM
IM
IM
IM
IM
IM
IM
IM
99
88
IM
IM
IM
IM
IM
IM
IM
2
IM
IM
IM
IM
Output Mux
13
12
33
22
Input Demux
IM
IM
IM
IM
Output Mux
13
12
IM
IM
Input Demux
IM
IM
Output Mux
13
12
IM
IM
Input Demux
13
12
Output Mux
IM
IM
IM
Input Demux
Processor
Core
Instruction
Memory
IM
Column Column
Memory Memory
IM
IM
IM
IM
IM
IM
IM
IM
11
10
11
10
11
IM
IM
IM
IM
IM
IM
IM
IM
10
11
10
15
14
15
15
14
14
15
15
IM
IM
IM
IM
IM
IM
IM
IM
14
14
15
14
Feedback
Version 1.0
853
Module 8

PXF Packet Flow
_____________________________ Note __________________________
The drawing that follows is simplified to show only four rows and
columns of the PXF to better demonstrate packet flow.
____________________________________________________________
Packet Assignment to Rows
When a packet header and context enter the PXF, they are assigned to one
of the eight rows. The packet headers are assigned to rows in a strict
round -robin fashion. Every packet header with context will pass through
all columns within a row. The choice of a row has no effect on the packet
header processing because functionality is identical for all CPUs in a
particular column irrespective of the row that the packet traverses.
Packet Processing by CPUs
Each packet header spends 192 clock cycles (154-MHz clock cycle) or
approximately 1.24 microseconds processing time at a CPU in a row. The
packet header plus context is then forwarded to the next CPU (column) in
the row. Each CPU in the row is configured to provide a different series of
functions for the packet header.
Once every 24 clock cycles (192 cycles divided by 8 rows), a new context
will enter the next PXF row. Therefore, each packet within a column is
offset from its adjacent row by 24 cycles. The result is that the PXF can
simultaneously process up to 64 packet headers with contexts.
Completing the Process
After the packet header plus context makes its way through all eight
columns of CPUs, one of two things can happen.
854
Single pass Typically, all of the necessary processing of the packet

header will be completed and the header will be sent back to the
Cobalt 2 ASIC to be buffered for future transmission.
Multiple passes Occasionally, more processing will be required on a

packet header than can be handled in the fixed amount of time it
spends in a single CPU, or a feature in one PXF column may change
the behavior of an earlier column. In this case, the packet is sent back
to the Cobalt 2 ASIC to be returned to the first PXF column for a
second pass through all of the columns. This reprocessing is known as a
feedback.
Version 1.0
Module 8
P2
P3
P4
Data
Input
Shared
Column
Memory
Shared
Column
Memory
Shared
Column
Memory
Shared
Column
Memory
CPU0
Complex
P1
CPU1
Complex
P1
CPU2
Complex
P1
CPU3
Complex
CPU4
Complex
P2
CPU5
Complex
CPU6
Complex
P2
CPU7
Complex
CPU8
Complex
P3
CPU9
Complex
P3
CPU10
Complex
P3
CPU11
Complex
CPU12
Complex
CPU13
Complex
CPU14
Complex
CPU15
Complex
P4
P2
P4
P4
P1
P2
P3
P4
Output header buffer
P1
Input header buffer
PXF Packet Flow
P1
Data
Output
P2
P3
P4
Feedback Path
Multiple on-chip processors using pipelining and parallelism to

maximize use of external data memories
Version 1.0
855
Module 8

Column Functionality
The following is an example of the packet header processing that would
take place as each column in a PXF row is traversed. In reality there are
numerous functional paths and functions within the PXF.
_____________________________ Note __________________________
The PXF is controlled by a microcode image that is part of the Cisco
IOS software, therefore the functionality of the microcode and each
column may vary with each release.
____________________________________________________________
Column 0
The packet header with context is received from the Cobalt 2 ASIC. The
arriving context contains a pointer to the packet body being held in the
input packet memory and information on the input interface. In this
column an inbound virtual channel common index (VCCI) for use by the
PXF is assigned in the packets context. In addition, Layer 2 MTU and IP
CRC checks are performed, and a pointer to the IP header is established.
Finally, inbound interface statistics are accumulated in this column.
Column 1
Column 1s column memory contains the Mtree table. An Mtree fixed-time

lookup takes place to obtain the destination route pointer, outbound
encapsulation and VCCI. The packets context is updated with the
appropriate information. To help put this in perspective, the inbound and
outbound VCCIs that are now known will be used in the appropriate
column to help determine features to be applied to a packet, such as ACLs
and QoS. In addition an RPF check is performed to determine whether the
packets source address and VCCI agree.
Column 2
In this column the need to process inbound and or outbound ACLs for the
packet is determined based on VCCI information. If an outbound ACL
needs to be processed, the feedback bit is set in the header context. In
addition, multilink tracking is conducted in the form of checking and
writing sequence numbers. Up to 2000 packets can be tracked per bundle.
856
Version 1.0
Module 8
Column Functionality
inbnd VCCI, L2 mtu chk, pointer to IP header,

IPCRC, inbnd int stats
Mtree constant time lookup, dest route ptr, outbnd
encap & VCCI for features such as QoS & ACls,
RPF check does SRC address agree with VCCI
Que/Deque pkt, VTMS, Traffic Shaping,

PQ, WRED based on column #4 info
Input VCCI + pkt hdr = Input ACL? Out ACL set =

yes - set feedback, MLP tracking, check & write seq #,
2K packets per MLP bundle
VCCI determines policy maps & classes ie. packet
classification, Input or Output policy only 1pass,
Input & output policy set feedback
VTMS, Dequeue wheel,

reschedule, Activate, deactivate
Queue
Implement policies from #3, rate limit,mark,

Que selection, process inbound or outbound ACL,
control and track IP frag
MAC rewrite based on header
encap, L2 out stats, frame header,
DLCI, ATM header
0
IM
1
IM
2
IM
3
IM
4
IM
Version 1.0
5
IM
6
IM
7
IM
857
Module 8

Column Functionality (continued)
Column 3
The VCCI with packet header is used to determine policy maps and c lasses
for a packet. If a packet will be subject to both input and output QoS, then
a feedback bit will be set in the header context.
Column 4
In column 4 the policies determined within column 3 are implemented,

such as rate limiting, marking, and queue selection. In addition, inbound
or outbound ACLs are processed (outbound ACLs are processed on the
second pass) and IP fragmentation is controlled and tracked.
Column 5
MAC rewrite takes place here, based on the header encapsulation frame
header, data-link connection identifier (DLCI) or ATM header. In addition,
Layer 2 outbound statistics are accumulated.
Column 6
Operation of the versatile time management system (VTMS) takes place

here, in addition to traffic shaping, queuing, and drop precedence
calculations. Packet queue and dequeue commands are added to the packet
context.
Column 7
Operation of the VTMS also takes place in this column. Instructions for the
dequeue wheel that is, rescheduling, activating and deactivating a queue
also take place in this column. Columns 6 and 7 use shared memory to
share VTMS information.
858
Version 1.0
Module 8
Column Functionality (continued)
inbnd VCCI, L2 mtu chk, pointer to IP header,

IPCRC, inbnd int stats
Mtree constant time lookup, dest route ptr, outbnd
encap & VCCI for features such as QoS & ACls,
RPF check does SRC address agree with VCCI
Que/Deque pkt, VTMS, Traffic Shaping,

PQ, WRED based on column #4 info
Input VCCI + pkt hdr = Input ACL? Out ACL set =

yes - set feedback, MLP tracking, check & write seq #,
2K packets per MLP bundle
VCCI determines policy maps & classes ie. packet
classification, Input or Output policy only 1pass,
Input & output policy set feedback
VTMS, Dequeue wheel,

reschedule, Activate, deactivate
Queue
Implement policies from #3, rate limit,mark,

Que selection, process inbound or outbound ACL,
control and track IP frag
MAC rewrite based on header
encap, L2 out stats, frame header,
DLCI, ATM header
0
IM
1
IM
2
IM
3
IM
4
IM
Version 1.0
5
IM
6
IM
7
IM
859
Module 8
PRE Comparison
The graphic that follows provides a relative comparison of the PRE-1 and
the PRE-2 performance routing engines used in the Cisco 10000 router.
The key points include:
860
Increased packets per second throughput from 2.8 to 6.2 Mpps.
Increased route p rocessor memory size from 512 to 1024 MB.
Increased packet buffer size from 128 to 256 MB.
The ability to simultaneously process 64 packet headers versus 32

packet headers.
Increased route processor clock speed from 267 to 500 MHz and
increased Toaster clock speed from 100 to 154 MHz.
Line card interconnect has increased from 1.6 to 3.2 Gbps allowing for
use of the OC-48 POS line card and half-height Gigabit Ethernet line
card.
Version 1.0
Module 8
PRE Comparison
PRE Comparison
PRE-1
PRE-2
2.8 Mpps
6.2 Mpps
512 MB
1024 MB
1024 MB
1024 MB
128 MB
256 MB
32
64
Route Processor Clock
267 MHz
500 MHz
Toaster Clock
100 MHz
154 MHz
Line Card Interconnect
1.6 Gbps
3.2 Gbps
Benchmark PPS
Route Processor
Memory Size
Memory Size
Packet Buffer
Toaster Processors
PRE2
Version 1.0
861
Module 8
High Availability
The following features on the Cisco 10000 router provide high-availability
functionality:
862
Point-to-point wiring with the Iron Bus between line cards and the
PREs, preventing failures on one line card from interrupting traffic on
other line cards
Redundant PREs (know as RPR+) with automatic failover
Separate control and data planes that limit faults in one plane from
affecting the other
Redundant AC or DC power supplies
Redundant cooling
Automatic protection switching (APS) between redundant Packet-overSONET (POS) line cards
Front access to serviceable cards and components
Version 1.0
Module 8
High Availability
High Availability
RPR+ - Redundant processing engines with

automatic failover
Iron Bus - Dedicated point-to-point wiring
between line cards and each of the PREs
Separate control and data planes
Redundant AC or DC power supplies
Redundant cooling
APS for redundant POS cards
Version 1.0
863
Module 8
PRE Redundancy
The pages that follow describe how PRE redundancy operates in the
Cisco 10000 router.
Two PRE Slots

Using two PRE modules in the Cisco 10000 router provides a redundant
PRE environment with one PRE functioning as the active PRE and the
other as the standby PRE. During bootup, The PRE in slot 0A will become
the active PRE and the PRE in slot 0B the standby unit.
_____________________________ Note __________________________
There is no load sharing between the PREs.
____________________________________________________________
Status LEDs
The status LED on the standby PRE blinks green; the LED on the active
PRE is on continuously. In the broadband environment, RPR+ high
availability is supported.
864
Version 1.0
Module 8
PRE Redundancy
PRE Redundancy
Slot
0A
0B
Two PRE slots

One PRE active, one
PRE in standby
No load sharing
Same Cisco IOS
image on both PREs
Version 1.0
Status LED
Active PRE = solid green
Standby PRE = blinking green
865
Module 8
PRE Redundancy (continued)

Active PRE
The active PRE accepts all traffic from the line cards via the Iron Bus. The
PXF is responsible for executing the forwarding plane while the route
processor continues to execute the control plane. In addition the active
PRE monitors the health of the standby unit via the Route Processor
Redundancy (RPR) and Inter-Process Ethernet (IPE, Fast Ethernet bus)
buses. The Backplane Ethernet (BPE) permits the PRE to communicate
control information with the line cards.
Standby PRE
The standby PRE is in monitor mode monitoring the health of the active
PRE. The forwarding processor is initialized; that is, microcode is loaded
and the configuration file is loaded and synchronized. PPP and route states
are held in an initialization state and are not synchronized.
866
Version 1.0
Module 8
PRE Redundancy
Active PRE
IronBus connected to network interfaces
PXF forwarding processor forwards all
traffic
Route processor executes control plane
Monitors health of standby PRE
Standby PRE
Forwarding processor initialized -
IPE
PRE A
(Active)
RPR
PRE B
(Standby)
BPE
LC
microcode loaded
Cisco IOS startup and running configuration
synchronized
Monitors health of active PRE
Version 1.0
867
Module 8

Active PRE Operation Steady State
The active PREs route processor runs Cisco IOS software, controls the
routing protocols, provides the management interface, and monitors the
health of the standby PRE. The forwarding processor handles all inbound
and outbound traffic; that is, it is managing and controlling the queuing
process plus the active Iron Bus links to the line cards.
When operating in RPR+ mode, the active PRE automatically synchronizes
the startup and running configuration files in addition to the configregister and bootvar with the standby PRE. This is the default mode of
operation, known as standard.
_____________________________ Note __________________________
You can alter the Cisco 10000 router from operating in standard
synchronization mode using the redundancy configuration mode
commands.
____________________________________________________________
868
Version 1.0
Module 8
PRE Redundancy
Active PRE Operation Steady State
Active Route Processor runs IOS
Manages system resources, runs routing protocols

Provides network management interface
Monitors health of standby PRE
Active PRE forwards all traffic
Iron Bus connected to network interfaces
With RPR+, by default the active PRE synchronizes the

following files to the standby PRE:
startup-config
running-config
config-register
bootvar
Version 1.0
869
Module 8

Standby PRE Operation Steady State
Integrity
On the standby PRE with RPR+, the Cisco IOS image is loaded and the
configuration file has been processed. However, the interfaces Layer 2 and
Layer 3 protocols are held in an initialization state.
The standby PRE reports its state to the active PRE in addition to
monitoring the state of the active PRE, using RPR bus status signals and a
keepalive signal over the IPE.
Synchronization
The standby PRE synchronizes the following with the active PRE, using
the IPE bus:
Startup configuration
Running configuration
Bootvar
Config-register
Time of day
Wait for Switchover Events
Any of the following events will cause a switchover to and activation of the
standby PRE:
870
Failure of the active RP can result in a crash signal, watchdog timeout

or keep-alive failure.
Removal of the active PRE will result in the change in the redundancy
signal, causing a failover to take place.
Maintenance or upgrades resulting in an operator-initiated failover will

result in the standby PREs becoming active .
Version 1.0
Module 8
PRE Redundancy
Standby PRE Operation Steady State
Maintain integrity
Report health to active PRE via keepalives

Monitor local failure indication signals
Maintain synchronization from active PRE
startup-config
running-config
config-register
bootvar
time of day
Wait for switchover events
Failure of active PRE crash signal, watchdog timeout, keepalive

failure
Removal of active PRE redundancy signal change

Operator-initiated switchover maintenance or upgrade
Version 1.0
871
Module 8

Action on PRE Switchover
The following events occur during a switchover from one PRE to the other
PRE:
872
If a failure is detected, cutover is initiated by the standby PRE.
Line cards shift their backplane interface connection to the new active
PRE.
Line cards are not reset or reloaded.
Line cards reconnect within 5 seconds.
Interfaces change state to up .
Layer 2 and Layer 3 protocols initialize.
Version 1.0
Module 8
PRE Redundancy
Action on PRE Switchover
Cutover initiated by the standby

Line cards shift their backplane interfaces to the
new primary
Line cards are not reset or reloaded

Line cards reconnect within 5 seconds
Interfaces change state to up
Layer 2 and Layer 3 protocols initialize
Version 1.0
873
Module 8
Cisco 10000 Series Router Broadband Aggregation Line Cards

The pages that follow provide information about Cisco 10000 Series router
line cards that are suitable for broadband aggregation deployments.
874
ATM cards Overview each of the available ATM line cards for the
Cisco 10000 router followed by an overview of optimizing VC scaling for
these cards
4-port OC-3c/STM-1 ATM
1-port OC-12 ATM
8-port E3/DS3 ATM
LAN cards Overview of the available LAN line cards for the
Cisco 10000 router
1-port Gigabit Ethernet full-height card
1-port Gigabit Ethernet half-height card
8-port Fast Ethernet half-height c ard
POS cards Overview of the available POS cards for the Cisco
10000 router
1-port OC-12 POS/SDH card
6-port OC-3c/STM-1 POS/SDH card
1-port OC-48c/STM-16 POS/SDH card
Version 1.0
Module 8
ATM Cards

8xE3/DS3 ATM
ATM VCs
Optimizing VC Scaling on ATM Line Cards
LAN Cards
POS Cards
Version 1.0
875
Module 8
ATM Line Cards

4-Port OC-3c/STM-1 ATM Line Card
The 4-port OC-3c/STM-1 ATM line card is a standards-based ATM solution
supporting line rate (155 Mbps) performance at 64-byte packets. It allows
Internet service providers to offer line -rate Internet access via ATM virtual
circuits. When used in broadband aggregation deployments, it is an
effective solution for aggregating subscriber connections from DSLAMs.
Hardware Features
Full-height, single-slot, four-port
155-Mbps SONET/SDH OC-3 STS-3c/STM-1c framing format
Single-mode, interme diate-reach optics with an LC connector
Full-duplex OC-3 fiber-speed performance
64 MB of transmit and receive buffers
Conforms to ATM Forum 155-Mbps physical layer specification
LEDs
Fail Yellow indicates the line card power-on self test (POST) fails or a
failure during operation. Off indicates the line card is working properly.
Loopback Yellow indicates the port data path is in loopback and not
available for normal operation. Off when not in loopback.
Alarm Yellow indicates an alarm condition exists at the

corresponding port.
Carrier Detect Green indicates a carrier is detected at the

corresponding port. Off indicates a loss of signal (LoS).
High-Availability Features Supported
876
1+1 redundancy per port line card using automatic protection switching
(APS)
1+1 line-card redundancy for card failover
OIR
RPR+
SONET-based alarms
Version 1.0
Module 8
ATM Line Cards
4-Port OC-3c/STM-1 ATM Line Card
Hardware Features
Four port
LC duplex connector
Single mode
Intermediate reach
SONET/SDH OC-3 STS-3c/STM1c framing format
Conforms to ATM Forum
physical layer specifications
64 MB Tx and Rx Buffers
LEDs
High-Availability Features
1+1 redundancy per port line

card using APS
OIR
RPR+
SONET based alarms
Version 1.0
877
Module 8
ATM Line Cards (continued)

1-Port OC-12 ATM Line Card
The 1-port OC-12 ATM line card is a standards-based ATM solution
supporting line rate (622 Mbps) performance at 64-byte packets. It
provides an excellent uplink between leased-line customers and ISP
backbone ATM devices.
Hardware Features
Full-height, single-slot, single-port
622-Mbps SONET/SDH STS-12c/STM-4c framing format
Single-mode, intermediate -reach optics with SC connector
Full-duplex OC-12 fiber-speed performance
Conforms to ATM Forum 622-Mbps physical layer specification
LEDs
Fail Off (Card OK)/Yellow (Card failure)
Enable Green (Port traffic enabled)/Off (Port traffic disabled)
Alarm Off (No alarms at oc-12 level)/Yellow (OC-12 level alarm)
Loop Off (Loopback disabled)/Yellow (Port in loopback, no data traffic)
Receive Green (Port receiving traffic)/Off (No traffic)
Transmit Green (Transmitting traffic)/Off (No traffic)
Carrier Green (Carrier detected)/Off (No carrier)
878
1+1 redundancy per port line card using automatic protection switching
(APS)
1+1 line-card redundancy for card failover
OIR
RPR+
SONET-based alarms
Version 1.0
Module 8
ATM Line Cards
1-Port OC-12 ATM Line Card
Hardware Features
Single port
SC duplex connector
Single mode
Intermediate reach
SONET/SDH OC-12 STS-
12c/STM-4c framing format
Conforms to ATM Forum
physical layer specifications

LEDs
1+1 redundancy per port line

card using APS
OIR
RPR+
SONET based alarms
Version 1.0
879
Module 8
ATM Line Cards (continued)

8-Port E3/DS3 ATM Line Card
The Cisco 10000 Series 8-Port E3/DS3 ATM line card provides high-density
connectivity and deployment flexibility for the Cisco 10000 Series router.
When used in broadband aggregation deployments, it is an effective
solution for aggregating subscriber connections from DSLAMs.
Hardware Features
Full height, single slot, eight ports
75-ohm coaxial cable to a length of 450 feet, with the capability for both
remote and local side loopback.
E3 framer providing 34.368Mbps, G.751 or G.832 E3 application, G.751

E3 physical layer convergence procedure (PLCP)
DS3 framer provides: 44.736Mbps, C-bit parity and M23 based ATM
Direct Mapping (ADM), C-bit parity and M23 based PLCP
LEDs
Loopback Yellow indicates the port data path is in loopback and not
available for normal operation. Off when not in loopback.
Alarm Yellow indicates an alarm condition exists at the

corresponding port.

880
OIR
RPR+
Version 1.0
Module 8
ATM Line Cards
8-Port E3/DS3 ATM Line Card
Hardware Features
8-port DS3 or E3
Line build out: 450 ft of 75-ohm
coax
E3 framer provides 34.368 Mbps
DS3 framer provides 44.736 Mbps
Conforms to ATM Forum physical
layer specifications
LEDs
OIR
RPR+
Version 1.0
881
Module 8
ATM Line Card Common Features

The following describes features that are common to the Cisco 10000 router
line cards.
Card Functions
The Cisco 10000 router ATM line cards functionality is focused on Layer 2
(ATM) services, and they rely on the PRE to provide Layer 3 services. The
line cards receive and transmit ATM cells on the physical interfaces while
transmitting and receiving packets from the backplane.
ATM Features
The Cisco 10000 router ATM line cards support the following ATM
features:
8 VPI and 16 VCI bits
User-Network Interface (UNI) Versions 3.x and 4.0
Integrated Local Management Interface (ILMI) Version 4.0 including

permanent virtual connection (PVC) auto discovery feature
Standard ATM traffic management categories (VBR-nrt and UBR)
RFC 2684 AAL5 logical link control
Standard F4/F5 OAM
Supports AAL5 data transports: SNAP and MUX
VCs per Line Card

The following are the number of VCs supported by each line card type:
882
4-port OC-3 16,000 VCs per port ; 32,000 VCs per card
Single-port OC-12 16,000 VCs per card
8-port E3/DS3 4,000 VCs per port ; 32,000 VCs per card
Version 1.0
Module 8
ATM Features
VCs per Line Card
Supports VBR-nrt and UBR
4-port OC-3 16,000 VCs per port,
traffic
32,000 VCs per card
Supports 8 VPI and 16 VCI bits

PVCs and permanent virtual
1-port OC-12 16,000 VCs per

card
8-port E3/DS3 4000 VCs per port,
path (PVPs)
32,000 VCs per card
Supports AAL5 data transport

(SNAP and MUX)
Supports F4 and F5 OAM

UNI 3.x/4
ILMI management and
SAR and QoS Features

Per-VC queuing/shaping for VBRnrt, including CBWFQ and PQ
Group queuing and shaping for

UBR VCs
autodiscovery
CAC support for VBR-nrt, needs

available bandwidth
Map IP CoS to ATM QoS
Version 1.0
883
Module 8
ATM Line Card Common Features (continued)

SAR
The line cards feature a high-performance programmable segmentation

and reassembly (SAR) that is adapted for various applications including
advanced traffic management, cell scheduling, and integrated buffer
management.
The programmable feature of the SAR allows flexibility for software
upgrades and the ability to support new standards.
QoS and Shaping Features
Mapping of IP class of service (CoS) to ATM quality of service (QoS)

enables the effective management of traffic flows across heterogeneous
IP and ATM networks.
Per-VC and per-VP traffic shaping are supported in the highperformance PXF network processor. Providing traffic shaping on a
per-VC and per-VP basis allows flexibility and control over every VC
and VP configured.
884
For UBR VCs, queuing and shaping are done on a group basis.
For VBR-nrt PVCs, queuing and shaping, including CBWFQ and

PQ, are provided on a per-VC basis.
Supports Call Admission Control (CAC) by not permitting the

configuration of VBR-nrt PVCs if the requested bandwidth is not
available on the port.
Version 1.0
Module 8
ATM Features
VCs per Line Card
Supports VBR-nrt and UBR
4-port OC-3 16,000 VCs per port,
traffic
32,000 VCs per card
Supports 8 VPI and 16 VCI bits

PVCs and permanent virtual
1-port OC-12 16,000 VCs per

card
8-port E3/DS3 4000 VCs per port,
path (PVPs)
32,000 VCs per card
Supports AAL5 data transport

(SNAP and MUX)
Supports F4 and F5 OAM

UNI 3.x/4
ILMI management and

Per-VC queuing/shaping for VBRnrt, including CBWFQ and PQ
Group queuing and shaping for

UBR VCs
autodiscovery
CAC support for VBR-nrt, needs

available bandwidth
Map IP CoS to ATM QoS
Version 1.0
885
Module 8

Packet Layer Features
The following packet layer features are supported:
Application of the following QoS to each VBR-nrt PVC and to all UBR
PVCs as a group using Modular QoS CLI (MQC)
Priority queuing (PQ)
Class-based weighted fair queuing (CBWFQ)
Weighted random early d etection (WRED]
Committed access rate (CAR)
Setting the ATM cell loss priority (CLP) bit.
Multicast
Access control lists (ACLs)
Multiprotocol Label Switching (MPLS)
Frame mode MPLS over ATM
MPLS cell mode with label controlled ATM (LC-ATM) interface

support.
886
This feature allows service providers with existing ATM backbone

to upgrade their switches with label switch controller to support the
MPLS control plane.
Version 1.0
Module 8

Application of QoS to each VBR-nrt PVC and to UBR PVCs as a
group via MQC
ATM CLP
Multicast
ACLs
MPLS
Frame mode MPLS over ATM
MPLS cell mode with LC-ATM
Version 1.0
887
Module 8
Assigning VPI/VCIs for ATM VC Scaling

Overview
The Cisco 10000 router ATM line cards support the full range of VPI/VCI
values: 8 VPI bits and 16 VCI bits. The SAR on the ATM line cards use the
VPI/VCI values assigned to VCs along, with a physical port number as a
unique identifier for the VCs configured on the line card. There are some
restrictions on how VCs are assigned that , if not selected properly, can
result in reduced VC counts from the theoretical maximum.
The following table gives the maximum number of VCs supported for each
line card type.
8-Port DS3/E3
4-Port OC3
1-Port OC12
Maximum VCs/port
4000
8000
16000
Maximum VCs/card
32,000
32,000
16000
_____________________________ Note __________________________

To attain maximum VC density no pxf queuing should be configured
on the ports.
____________________________________________________________
Internal Logical Identifier

To enable the SAR to support the same VPI/VCI values per physical
interface, the external VPI/VCI values assigned to VCs are translated into
an internal 32-bit logical header. The logical header is divided into a 25-bit
tag and 7-bit offset, with both parts pointing to unique channel descriptors.
Each VC configured on the line card uses one channel descriptor.
The bit designations are listed below for the illustration, from left to right.
888
Tag/
Offset
Bits
Field
Description
Tag
3129
N/A
Not used
Tag
2824
PHY
Physical port identifier
Tag
2316
VPI
8 bits of VCI
Tag
158
VCI
Upper 9 bits of VCI (BCD values 32384 to 128)
Offset
70
VCI
Lower 7 bits of VCI (BCD values 64 to 1)
Version 1.0
Module 8
ATM VC scaling
Consider VPI/VCI selections
when creating PVCs
31
SAR translates physical port
28
N/A PHY#
23
15
VPI
76
VCI
ID + VPI/VCI into 32-bit logical

header
Logical header divided into
Tag
Offset
tag and offset to determine

channel descriptor
Tag is based on
Physical field
Entire VPI field
Upper 9 bits of the VCI
Offset = bits 0 7 of VCI
Version 1.0
889
Module 8
Assigning VPI/VCIs for ATM VC Scaling (continued)

Channel Descriptor Usage
Channel descriptors are grouped into 512 pages with each page containing
128 channel descriptors. The tag portion of the internal logical header
points to one of 512 pages and the offset points to one of 128 channel
descriptors within a page.
Therefore, when a VC is configured on a port of the ATM line card, it is the
combination of the port identifier, VPI and upper 9 bits of VCI that results
in a unique page number. The Cisco 10000 router ATM line cards support
a maximum of 512 channel descriptor pages
Achieving Maximum Scalability

To achieve maximum VC scalability, you should assign VCIs for a given
VPI utilizing a contiguous numbering sche me in groups of 128 VCs
(bits 0 7 of the VCI) that utilize all 128 channel descriptors in a page.
The following two examples illustrate the number of channel descriptor
pages that are used to configure 200 VCs on an ATM port.
Example 1
Configured VPI/VCI combinations on an ATM port are as follows:
1/100, 1/200, 1/300, etc. through 1/4000
2/100, 2/200, etc. through 2/4000
Etc. through 5/4000
This example requires 155 channel descriptor pages

Example 2
Configured VPI/VCIs on an ATM port are as follows:
1/32, 1/33, 1/34, etc., through 1/231
This example requires 2 channel descriptor pages.

With example 1, if this VC numbering scheme were used on all four ports
of the 4-port OC3 ATM line cards, then all 512 channel descriptor pages
would be used before all VCs were configured on the line card. In practice,
depleting the maximum number of channel descriptor pages should not
occur as VCs are usually defined by incrementing the VCI sequentially on
a given VPI
890
Version 1.0
Module 8
Assigning VPI/VCIs for ATM VC Scaling (continued)
Channel Descriptors
One per VC
Grouped into 512 pages
Each page contains 128
descriptors
Achieving maximum
scaling
Assign VCs in groups of
128 that utilize all channel

descriptors in a page
Example of 200 VCs
PVCs 1/32 through 1/231

Uses 2 pages
Version 1.0
TAG
OFFSET
Search
Channel Descriptor 0
Page 0
TAG 0
TAG 1
Channel Descriptor 127
TAG 511
Page 511
Channel Descriptor
Page #
CD #
891
Module 8
LAN Line Cards

Gigabit Ethernet Line Card
The Cisco 10000 router Gigabit Ethernet interface line card is an effective
solution for Intra-POP interconnections among Internet service providers.
The Gigabit Ethernet line card in the Cisco 10000 router provides a costeffective high-performance uplink to backbone routers such as the
Cisco 12000 series Gigabit Switch Router (GSR).
Hardware Features
Full-height, single slot
Single 1-Gbps port
Complies with 802.3z standards
1-Gbps full duplex
Choice of SC Gigabit Interface Converter (GBIC) transceivers: SX,

LX/LH, and ZX
Receive buffering 16MB
LEDs
Fail Solid yellow indicates the line card power-on self test (POST)
fails or a failure during operation. Off indicates the line card is
working properly. A blinking Fail LED is an indication of a defective or
incompatible GBIC.
Link Green indicates carrier detected and the port is able to pass
traffic. If negotiation is enabled at both end, it indicates successful
completion and the port can pass traffic. Off indicates that no carrier
signal is detected, negotiation failed, or the port is administratively
down.
Rx Green indicates that packets are being received.
Tx Green indicates that packets are being transmitted.
High Availability Features Supported
892
OIR
RPR+
Version 1.0
Module 8
LAN Line Cards
Gigabit Ethernet Line Card
Hardware Features
Full-height 1-port
Gigabit Ethernet
1 Gbps full duplex
GBIC transceiver with SC

connector
1000BASE-SX (802.3z specs)

1000BASE-LX/LH (802.3z
specs)
1000BASE-ZX
Receive buffering: 16 MB
LEDs
OIR
RPR+
Version 1.0
893
Module 8
LAN Line Cards (continued)

Gigabit Ethernet Line Card (continued)
GBIC Specifications
The table lists the specification about the GBIC transceivers that may be
used on the Gigabit Ethernet line card.
The 1000BASE-LX/LH GBIC exceeds IEEE 802.3 5-km reach requirement
for 1000BASE-LX over single-mode fiber (SMF). If multimode fiber (MMF)
is used, then a mode-conditioning patch cord is required on both ends for a
link distance of tens of meters.
The 1000BASE-ZX GBIC has a reach of 70 km over SMF. The reach can be
extended to 100 km using premium or dispersion-shifted fiber.
894
Version 1.0
Module 8
LAN Line Cards
GBIC Specifications
GBIC
1000BASE-SX
1000BASE-LX/LH
1000BASE-ZX
Wavelength
(nm)
Fiber
850
MMF
1300
SMF
Core Size
(microns)
62.5
62.5
50.0
50.0
62.5
50.0
50.0
9/10
SMF
9/10
MMF
1550
Version 1.0
Reach
(meters)
220
275
500
550
550
550
550
10 km
70 km to
100 km
895
Module 8

Gigabit Ethernet Half-Height Line Card
The Cisco 10000 router Gigabit Ethernet Half-Height line card is an
effective solution for Intra-POP interconnections among Internet service
providers and provides a cost-effective high performance uplink to
backbone routers such as the Cisco 12000 series Gigabit Switch Router
(GSR). This line card doubles the Gigabit Ethernet capabilities of the
Cisco 10000 Series router.
Hardware Features
Half-height slot using the Cisco 10000 router carrier card
Single 1-Gbps port
Complies with 802.3z standards
1-Gbps full duplex
Choice of small form factor pluggable (SFP) transceivers: SX and

LX/LH
LEDs
Fail Solid yellow indicates the line card power-on self test (POST)
fails or a failure during operation. Off indicates the line card is
working properly. A blinking Fail LED is an indication of a defective or
incompatible GBIC.
Link Green indicates carrier detected and the port is able to pass
traffic. If negotiation is enabled at both end, it indicates successful
completion and the port can pass traffic. Off indicates that no carrier
signal is detected, negotiation failed, or the port is administratively
down.
896
OIR
RPR+
Version 1.0
Module 8
LAN Line Cards
Gigabit Ethernet Half-Height Line Card
Hardware Features
Half-height single-port
Gigabit Ethernet
1 Gbps full duplex
SFP transceiver with LC

connector
1000BASE-SX (802.3z specs)

1000BASE-LX/LH (802.3z
specs)
Line rate with 64 byte packets

LEDs
OIR
RPR+
Version 1.0
897
Module 8

Gigabit Ethernet Half-Height Line Card (continued)
SFP Specifications
The table lists the specification about the SFP transceivers that may be
used on the Gigabit Ethernet half-height line card.
_____________________________ Note __________________________
A mode-conditioning patch cord is required. If you use an ordinary
patch cord with multimode fiber (MMF), 1000BASE -LX/LH SFP
transceivers, for a short link distance (tens of meters), this can cause
transceiver saturation, resulting in an elevated bit error rate (BER). In
addition, when you use the LX/LH SFP transceivers with 62.5-micron
diameter MMF, you must install a mode-conditioning patch cord
between the transceiver and the MMF cable on both the transmit and
receive ends of the link. The mode-conditioning patch cord is required
for link distances of less than 300 meters.
____________________________________________________________
898
Version 1.0
Module 8
LAN Line Cards
SFP Specifications
SFP
Wavelength
(nm)
Fiber
850
MMF
1000BASE-SX
GLC-SX-MM
MMF
1000BASE-LX/LH
GLC-LH-SM
1300
SMF
Core Size
(microns)
62.5
62.5
50.0
50.0
Reach
(meters)
220
275
500
550
62.5
50.0
50.0
8/10
550
550
550
10 km
See note on text page
Version 1.0
899
Module 8

Gigabit Ethernet Common Features
The following are features common to both the full-height and half-height
Gigabit Ethernet line cards.
Media Access Control (MAC) with full-duplex operation and flow

control
Hardware address filtering on received frames of up to 4096 address

entries
Ethernet encapsulation formats:
Ethernet V2
802.2 SAP
802.2 SNAP
802.1Q support for 4096 VLANs
Line rate at 64-byte packets
MTU 9180 bytes
802.3x flow control (Tx and Rx negotiated separately)
Flow control enables connected Ethernet ports to control traffic

rates during congestion by allowing congested nodes to pause link
operation at the other end. If one port experiences congestion and
cannot receive any more traffic, it notifies the other port to stop
transmitting until the condition clears. When the local device
detects any congestion at its end, it can notify the link partner or
the remote device of the congestion by transmitting a pause frame.
On receiving a pause frame, the remote device stops transmitting
any data packets. This prevents any loss of data packets during the
congestion period.
Accepts Ethernet frames for its unicast address, and it participates in

other multicast protocols such as CDP.
_____________________________ Note __________________________
The Gigabit Ethernet card on the Cisco 10000 defaults to autonegotiation

On, that is negotiation of send and receive capabilities.
____________________________________________________________
8100
Version 1.0
Module 8
LAN Line Cards
Gigabit Ethernet Common Feature
Gigabit Ethernet Features
Full-duplex operation and flow control
MTU of 9180 bytes

Autonegotiation
Hardware address filtering up to 4096 address entries

802.3x flow control

Ethernet V2
802.2 SAP
802.2 SNAP
64-bit counters
Hot Standby Router Protocol (HSRP)
Version 1.0
8101
Module 8

8-Port Fast Ethernet Half-Height Line Card
With the 8-Port Fast Ethernet half-height card, the Cisco 10000 router
serves the Ethernet aggregation market with maximum flexibility for a
variety of densities and aggregation topologies.
Hardware Features
Half-height slot using the Cisco 10000 router carrier card
Eight ports o f Fast Ethernet
Each port configurable for either full- or half-duplex operation
512 content-addressable memory (CAM) addresses per port
16-MB receive packet memory
Error-Correction Code (ECC) protection for the processor local memory and
packet memory
LEDs
Link Green indicates a valid Ethernet link.
Activity Green indicates presence of transmitted and received

Ethernet packets.
8102
OIR
RPR+
Version 1.0
Module 8
LAN Line Cards
8-Port Fast Ethernet Half-Height Line Card
Hardware Features
Half-height slot
Eight ports of Fast Ethernet
Configurable per port half/full
duplex
ECC protection for the
processor and packet memory
512 CAM addresses per port

LEDs
OIR
RPR+
Version 1.0
8103
Module 8

8-Port Fast Ethernet Half-Height Line Card (continued)
Fast Ethernet Features
8104
Media Access Control (MAC) with full-duplex operation and flow

control
Hardware address filtering on received frames of up to 4096 address

entries
802.3x flow control
Ethernet encapsulation formats
Ethernet V2
802.2 SAP
802.2 SNAP
MTU of 9180 bytes
Autonegotiation
64-bit counters
Hot Standby Router Protocol (HSRP)
Version 1.0
Module 8
LAN Line Cards
8-Port Fast Ethernet Half-Height Line Card (continued)
Fast Ethernet Features
Full-duplex operation and flow control
Hardware address filtering up to 4096 address entries

802.3x flow control
Ethernet V2
802.2 SAP
802.2 SNAP
MTU of 9180 bytes

Autonegotiation
64-bit counters
HSRP
Version 1.0
8105
Module 8
Packet over SONET Line Cards

1-Port OC-12 POS/SDH Line Card
The Cisco 10000 router OC-12/STM-4 POS/SDH line card is a highcapacity, high-performance interface that provides connectivity from the
aggregation router to the core. This line card provides Internet service
providers with significant performance improvements in their existing
fiber networks.
Hardware Features
1-port OC-12/STM-4 SONET interface
Operates at 622 Mbps full duplex
SC duplex connector supports single-mode, intermediate-reach fiber

(maximum distance 15 km)
Unit power budget: 28W
LEDs
Carrier Green indicates the carrier signal is detected. Off indicates a

loss of signal (LoS).
8106
Per-port SONET Automatic Switching Protection (ASP)
Per-port SDH Multiplex Section Protection (MSP)
OIR
RPR+
Version 1.0
Module 8
1-Port OC-12 POS/SDH Line Card
Hardware Features
Single port OC-12/STM-4
SONET/SDH
Operates at 622 Mbps, fullduplex
SC duplex connector
Single mode
Intermediate reach (15 km)

LEDs
SONET ASP, SDH MSP

OIR
RPR+
Version 1.0
8107
Module 8
Packet over SONET Line Cards (continued)

6-Port OC-3c/STM-1 POS/SDH Line Card
The Cisco 10000 router six-port OC-3c/STM-1 POS line card is a highcapacity, high-performance interface that enables service providers to offer
dedicated Internet access services and peering to other service providers at
OC-3c/STM-1 rates. This line card provides a powerful combination of
scalability and flexibility, enabling service providers to substantially
increase the data load over their optical transport infrastructure.
Hardware Features
6-port OC-3c/STM-1 SONET interface
Operates at 155 Mbps full-duplex
LC duplex connector supports single-mode, intermediate-reach fiber

(maximum distance 15 km)
Receive buffering 16 MB
LEDs
Loopback Yellow indicates the data path is in loopback, one per port
Alarm Yellow indicates the presence of a port alarm, one per port

8108
Per-port SDH Multiplex Section Protection (MSP)
OIR
RPR+
Version 1.0
Module 8
Hardware Features
6-port OC-3c/STM-1
SONET/SDH
Operates at 155 Mbps, fullduplex
LC duplex connector
Single mode
Intermediate reach (15 km)

LEDs
SONET ASP, SDH MSP

OIR
RPR+
Version 1.0
8109
Module 8
Packet over SONET Line Cards (continued)

The Cisco 10000 router 1-port OC-48c/STM-16 POS/SDH line card provides
a high-performance interface from the Cisco 10000 router to the core
network. This line card enables Internet service providers to substantially
increase the data load over their optical transport infrastructure.
Hardware Features
One port OC-48c/STM-16 SONET interface
Operates at 2.488 Gbps full duplex
SC connectors support
single-mode, short-reach fiber (maximum distance 2 km)
single-mode, long-reach fiber (maximum distance 80 km)
LEDs
POS Green indicates the card is operating in POS mode
Enable Green indicates the port administratively enabled
Carrier Green indicates the carrier signal is detected. Off indicates a

loss of signal (LoS).
SRP, Sync, Wrap, Pass Thru currently not supported when installed
in the Cisco 10000 chassis.
High Availability Features
8110
Per port SDH Multiplex Section Protection (MSP)
OIR
RPR+
Version 1.0
Module 8
Hardware Features
1-port OC-48c/STM-16
SONET/SDH
Operates at 2.488 Gbps. fullduplex
SC connector
Single-mode short-reach (2 km)

Single-mode long-reach (80 km)

LEDs
High Availability Features
SONET ASP, SDH MSP

OIR
RPR+
Version 1.0
8111
Module 8
Common POS/SDH Line Card Features

The following features are common to the three Cisco 10000 router
POS/SDH line cards.
Standards-compliant SONET interface: GR253, ITU G.707 and G.957

compliant
Alarm processing
Performance monitoring
Error counts for B1, B2, B3
Threshold crossing alerts (TCA) for B1, B2, B3 with settable

threshold
Synchronization
Local (internal) or loop timed (recovered from network)
Clock accuracy OC-3 and OC-48 POS cards: 4.6 ppm
Clock accuracy OC-12 POS card: 20 ppm
Local (diagnostic) and line (network) loopback
Encapsulation
8112
Loss of signal (LOS), loss of frame (LOF), line alarm indicator signal
(LAIS), path alarm indicator signal (PAIS), loss of pointer (LOP),
line remote defect indicator (LRDI), p ath remote defect indicator
(PRDI), signal failure (SF), signal degrade (SD), line remote error
indicator (line FEBE), path remote error indicator (path FEBE)
IETF RFC 1661, PPP
IETF RFC 1662, PPP in HDLC-like framing
IETF RFC 1490, Frame Relay encapsulation
NEBS Level 3 compliant
Version 1.0
Module 8
SONET/SDH Features
SONET/SDH standards compliant

Alarm processing
Error Counts and threshold cross alerts
Performance monitoring
Synchronization
Local (internal) or loop timed (recovered from network)

Clock accuracy OC-3 and OC-48 POS cards: 4.6 ppm
Clock accuracy OC-12 POS card: 20 ppm
Local (diagnostic) and line (network) loopback

Encapsulation
IETF RFC 1661, PPP

IETF RFC 1662, PPP in HDLC-like Framing
IETF RFC 1490, Frame Relay encapsulation
NEBS Level 3 compliant
Version 1.0
8113
Module 8
Summary
In this module, you learned the following:
8114
The broadband deployments of the Cisco 10000 router
The major chassis components of the Cisco 10000 router
The functional components of the Cisco 10000 router and their

interconnections and operation
The PRE-2 architecture and operation, including the forwarding

processor, router processor, and PXF
The flow of a packet through the Cisco 10000 router
The Cisco 10000 routers high availability hardware and functions
The features and functions of Cisco 10000 router line cards used in
broadband aggregation deployments
Version 1.0
Module 8
Review Questions
Review Questions
1. Which of the following statements is not true about the Cisco 10000
chassis?
a. The chassis has eight slots for line cards.
b. The chassis supports two PRE modules.
c. ATM line cards should be inserted in slots 1 4.
d. Half-height line cards may be used in any slot.
2. What are the two main sections of the PRE?
a. _____________________________________
b. _____________________________________
3. Which functions are performed by the route processor? Choose three.
a. Chassis management
b. System initialization
c. Route processor redundancy
d. Packet buffering
4. Which functions are performed by the forwarding processor? Choose
three.
a. Routing protocol updates
b. IP forwarding
c. Layer 3 features
d. QoS features
5. The _________________________ is the primary data path between the
PREs and line cards.
6. The backplane bandwidth between the PRE-2 and a line card slot is
a. 3.2 Gbps
b. 1.6 Gbps
c. 800 Mbps
d. 51.2 Gbps
Version 1.0
8115
Module 8
7. Which of the following statements is not true about the PXF?

a. The PXF is made up of 64 CPUs.
b. The CPUs are arranged into eight rows and eight columns.
c. Each column of CPUs has its own dedicated 128 MB column
memory.
d. Each column of CPUs is internal to Toaster ASICs.
8. What are the four resultant operations of the PXF?
a. _______________________________
b. _______________________________
c. _______________________________
d. _______________________________
9. Which of the following statements are true about PRE redundancy?
Choose three.
a. The active PRE is in slot 0A,
b. The standby PRE monitors the state of the active PRE
c. The standby PRE maintains of files with the active PRE
d. The standby PRE initiates cutover upon failure of the active PRE
10. Which of the following functions is common to all line cards? Choose
three.
a. OIR
b. APS
c. RPR+
d. Fail LED
8116
Version 1.0
Module 9
Overview
Description
In this module you learn about the Cisco 10000 Series Router software.
Included in this module are software elements, supported system features,
and supported Layer 2 connectivity options. This module also presents
Cisco IOS commands that display PXF statistical information. You will
perform hands-on exercises to observe the system hardware and software
components and will learn to use commands that display PXF statistics.
Objectives
After completing this module, you will be able to do the following:
List and describe the purpose of the major elements within the Cisco
10000 router operating system
List the major software services available with use of the PRE-2 in the
Cisco 10000 router
Describe the features, functions and interaction of QoS mechanisms

supported on the Cisco 10000 Series router
Use Cisco IOS commands to determine system-level status and alarms
Use the appropriate commands to display PXF statistics
Version 1.0
91
Module 9
Software Architecture
Overview
In the Cisco 10000 Series Router, the route processor, forwarding
processor, and line cards form a distributed system similar to other Cisco
products, including the Cisco 7500 and Cisco 12000 Series Routers. Each
of these components uses software that is shipped as part of the Cisco IOS
release.
Functional Components
The Cisco IOS software provides the following functions that are used by
the system components:
System control Provided by Cisco IOS 12.2(16)BX or later running on

the route processor
Packet forwarding Provided by Parallel eXpress Forwarding (PXF)

pipeline by means of custom microcode running on the PXF processors.
Line card and interface control Line card resources are controlled by
a Cisco-developed lightweight kernel called LC-DOS. LC-DOS provides
a lightweight execution environment easily tailored to the highly
embedded line card application.
The microcode for packet forwarding and for line card and interface control
is bundled with the Cisco IOS software.
92
Version 1.0
Module 9
Route processor, forwarding processor, and

line cards form a distributed system
System software functional components:
System control
Route processor
Packet forwarding
PXF microcode
Line card and interface control
LC-DOS lightweight kernel
Version 1.0
93
Module 9
Software components
Route ProcessorCisco IOS
The route processor is running Cisco IOS Release 12.2(16)BX or later that
enables the router to support broadband and leased-line aggregation
applications. It provides system management, routing protocols, and PXF
control.
Forwarding ProcessorTMC Microcode

The PXF processors run custom microcode developed for the Cisco 10000
router. This microcode, which is called the Toaster microcode or TMC
microcode, is optimized for IP forwarding and is bundled with the Cisco
IOS image.
Line CardsLC-DOS
The LC-DOS kernel provides a basic execution environment upon which
more specific applications can be built. It includes:
94
A scheduler
An exception handler
Memory management facilities
Task management facilities
A Cisco IOS interprocess communication (IPC) client for

communication with Cisco IOS software on the route processor
Support for card hot-swap and download
Version 1.0
Module 9
Software components
Software Components
Route Processor
System management, routing protocols, drivers, PXF control

Forwarding Processor: Toaster TMC microcode
PXF microcode implements the forwarding path

Forwarding features not implemented in microcode are not
supported
Bundled with Cisco IOS image

Line Card Software: LC-DOS
Line card and interface management

Not involved in packet forwarding
Lightweight kernel, unrelated to Cisco IOS
Currently bundled with Cisco IOS software
Version 1.0
95
Module 9
Cisco 10000 Router Software

Overview of Image Names
The following identify the Cisco IOS software releases that were released
with each type of Performance Routing Engine (PRE) module:
96
The original PRE shipped with the leased-line images c10k-p6-mz

12.0(9)SL 12.0(21)ST.
The PRE-1 shipped with the C10k-p10 image starting with code release
12.0(21)ST and continuing on with the c10k-p10-mz 12.0(21)ST
12.0(28)S images. These images are all strictly for leased-line
functionality. Currently the last planned release of new features for the
PRE-1 will be in release 12.0(28)S. This will be followed up with
maintenance releases only.
Broadband and leased-line functionality has been released with the

PRE-2 and image c10k2-p11-mz 12.2(16)BX. Listed below are near
term releases for PRE-2 with broadband and leased line aggregation
functionality:
12.2(16)BX
12.3TX
12.2S (will contain all of the 12.3 functionality)
Version 1.0
Module 9
Image names
PRE c10k-p6-mz
12.0(9)SL 12.0(21)ST leased-line images
PRE-1 c10k-p10-mz
12.0(21)ST 12.0(28)S, leased-line images, followed by

maintenance releases
PRE-2 c10k2-p11-mz (broadband aggregation plus

leased-line functionality)
12.2(16)BX
12.3TX
12.2S (will contain all of the 12.3 functionality)
Version 1.0
97
Module 9
Cisco 10000 Router Software (continued)

Software Bundling
The line card and PXF microcode images are bundled with the Cisco IOS
software.
During system initialization, the images are extracted, decompressed, and
downloaded as part of the initialization process:
The line card image is downloaded by the backplane Ethernet.
The PXF microcode is downloaded o ver a data path that exists within
the PRE.
To upgrade either the PXF microcode or the line card control processor
code, you must upgrade the Cisco IOS software. The PXF microcode may
be reloaded dynamically by the route processor after certain PXF faults,
obviating the need to restart Cisco IOS after certain PXF pipeline faults.
98
Version 1.0
Module 9
Software Bundling
LC and PXF microcode images bundled with

Cisco IOS software
Images are extracted, decompressed, and
downloaded during system initialization
Line card and PXF microcode is upgraded with

new Cisco IOS release
Dynamic reload of PXF microcode occurs after

certain PXF faults
Version 1.0
99
Module 9

PXF Microcode
The Cisco 10000 router currently supports IP Version 4 packets only.
The PXF engine does not support the following; instead, they are processed
by the route processor:
IP Version 6
AppleTalk
Internetwork Packet Exchange (IPX)
DECnet
IS-IS Protocol
Bridging
In addition, the PXF pipeline diverts to the route processor packets

addressed to any of the routers interfaces. Packets of this nature are
Routing updates
Telnet
Cisco Discovery Protocol (CDP)
Frame Relay Local Management Interface (LMI) packets
PPP control packets
Tag forwarding information base (TFIB) updates
Some Internet Control Message Protocol (ICMP) packets are handled by

the PXF engine (for example, ICMP unreachable, echo requests, and so on).
910
Version 1.0
Module 9
PXF Microcode
All transit packets are forwarded by the PXF

IPv4 is the only supported L3 protocol
Packets addressed to routers interfaces are
diverted (punted) to Cisco IOS software by PXF
Routing updates, Telnet packets, CDP, FR LMI,

PPP control packets
Some ICMP packets are handled by fast
forwarder microcode
Echo request, network unreachable, redirect
Version 1.0
911
Module 9

Line Card Software
The LC-DOS kernel is responsible for line card and interface management;
it is no t involved in packet forwarding. The LC-DOS software is currently
bundled with the Cisco IOS image shipped for the platform. Line card
software functionality is described below.
LC-DOS Functionality
The LC-DOS kernel provides a basic execution environment upon which
more specific applications can be built. It includes:
912
A scheduler.
An exception handler.
Memory management facilities.
Task management facilities.
A Cisco IOS IPC client for communication with Cisco IOS software on
the route processor.
Support for card hot-swap and download.
The base LC-DOS kernel is enhanced to provide functionality common

to all Cisco 10000 router line cards.
Support for line card control processor bootstrap.
A driver for the backplane Ethernet chip.
Management and control of the line card ID EEPROM.
A driver for the Barium Iron Bus termination application-specific

integrated circuit (ASIC).
Each instance of LC-DOS is enhanced to include functions specific to

the particular line card on which it is targeted to run.
Version 1.0
Module 9
Line Card Software
LC-DOS core
Scheduler, exception handler, memory management
Cisco IOS IPC client
OIR and download support
Common line card functionality
Line card control processor bootstrap
Backplane Ethernet driver
Line card IDPROM interface
Barium/Vanadium ASIC (Iron Bus termination chip) driver
Line card and interface-specific control functions
Version 1.0
913
Module 9
Supported Encapsulations
Serial interfaces
Serial interfaces (channelized, POS, E3/DS3, E1/T1)
Cisco High-Level Data Link Control (HDLC).
PPP (RFC 1570).
914
Unsupported protocol field compression, address and control field

compression
Multilink PPP (MLP) 1250 bundles, up to 10 links per bundle. The

maximum number links cannot exceed 2500.
Frame Relay.
Generic routing e ncapsulation (GRE).
Multiprotocol Label Switching (MPLS) Virtual Private Network (VPN)

and traffic engineering (TE) functionality.
VPN, provider (P) and provider edge (PE) functionality (RFC

2547bis)
TE, headend and tailend functionality
Version 1.0
Module 9
Serial interfaces channelized, POS, E3/DS3, E1/T1

Cisco HDLC
PPP (RFC 1570)
Unsupported: address and protocol compression
MLP
1250 bundles, up to 10 links per bundle

Maximum number of links not to exceed 2500
Frame Relay
GRE
MPLS VPN and TE functionality
VPN, P and PE functionality (RFC 2547bis)

TE, headend and tailend functionality
Version 1.0
915
Module 9
Supported Encapsulations (continued)

Gigabit Ethernet and Fast Ethernet Interfaces
Advanced Research Projects Agency (ARPA), service access point

(SAP)/IP, SAP/ Subnetwork Access Protocol (SNAP), MPLS
802.1q VLAN Support
ARPA is directly supported in the PXF path
4096 VLANs per interface
PPP over Ethernet over Ethernet (PPPoEoE and PPPoEo802.1q)
ATM interfaces
916
RFC 1483 and RFC 2684 routed
PPP over ATM (PPPoA)
PPP over Ethernet over ATM (PPPoEoA)
Route Bridge Encapsulation (RBE)
Version 1.0
Module 9
Gigabit Ethernet and Fast Ethernet interfaces

ARPA, SAP/IP, SAP/SNAP, MPLS
802.1q VLAN support
PPPoEoE, PPPoEo802.1q
ATM interfaces (OC-3, OC-12, DS3/E3)

RFC 1483 and RFC 2684 routed
PPPoA
PPPoEoA
RBE
Version 1.0
917
Module 9
Frame Relay Support

Supported Features
4200 Frame-Relay sessions.
2-byte header format.
DTE, DCE modes.
All three LMI variants.
Frame Relay traffic shaping.
Multilink Frame Relay (MLFR) (FRF.16) 1250 bundles, up to 10 links

per bundle. The maximum number of links is not to exceed 2500.
Unsupported features
918
Frame Relay switching
FRF.11, FR/ATM Interworking (FRF.5 or FRF.8)
Version 1.0
Module 9
Frame Relay Support
Frame Relay Support
Supported features:
4200 Frame Relay sessions

2-byte header format
DTE, DCE modes
All three LMI variants
Frame Relay traffic shaping
MLFR (FRF.16)
1250 bundles, up to 10 links per bundle

Maximum number of links not to exceed 2500
Unsupported features:
Frame Relay switching

FRF.11 and FR/ATM Interworking (FRF.5 or FRF.8)
Version 1.0
919
Module 9
Broadband Features and Scaling

Subscriber Sessions
The Cisco 10000 router supports autodetection of subscriber connections to
simplify provider configurations.
PPPoA
PPPoEoA
RBE
The following numbers of subscriber sessions are supported:
61,500 sessions with PPP termination and aggregation (PTA), Layer 2

Tunneling Protocol (L2TP) network server (LNS), L2TP access
concentrator (LAC) termination (simple configurations)
32,000 sessions RA-MPLS, managed LNS, VLANs, Service Selection

Gateway (SSG)
Tunneling
The following tunnel capabilities are supported:
920
L2TP 10,000 tunnels, maximum, for any combination inbound or

outbound
L2TP 61,500 PPP sessions per tunnel, maximum
L2TP tunnel switch functionality
GRE maximum of 384 tunnels
Version 1.0
Module 9
Subscriber Sessions
Autodetect
PPPoA, PPPoEoA, RBE
Sessions
61,500 sessions with PTA, LNS, LAC (simple configurations)

32,000 sessions with RA-MPLS, Managed LNS, VLANs, SSG
Tunneling
L2TP 10,000 tunnels for any combination inbound or

outbound
L2TP 61,500 PPP sessions per tunnel

L2TP Tunnel switch functionality
GRE maximum of 384 tunnels
Version 1.0
921
Module 9
Broadband Features and Scaling (continued)

ATM
PVC Autoprovisioning
The Cisco 10000 router supports ATM permanent virtual connection (PVC)
autoprovisioning. With this feature, digital subscriber line (DSL) wholesale
service providers can use a local configuration to dynamically provision
ATM virtual connections (VCs) for subscribers. Incoming traffic on the
virtual path identifier/virtual connection identifier (VPI/VCI) pair triggers
VC creation.
PVC Range
The PVC range command allows the definition of a range of PVCs to be

made available for automatic provisioning.
Oversubscription
The router allows the creation of more autoprovisioned PVCs than the
router permits to be active simultaneously, that is, 61,500 PVCs. Provided
that a sufficient number of ATM line cards are installed in the chassis, you
can configure up to 128,000 autoprovisioned PVCs instead of the 61,500
PVC system limit.
When the Cisco 10000 router is oversubscribed, use the idle-timeout
interface command to dynamically bring down inactive PVCs and allow
other subscribers to connect to the router.
922
Version 1.0
Module 9
ATM
PVC autoprovisioning
PVC range
Oversubscription
Version 1.0
923
Module 9

QoS
Per session Service Policy (Quick Connect)
This feature enables a subscriber management server (SMS), typically a

RADIUS server, to dynamically change the traffic policing parameters for
a user session. The RADIUS server maintains user profiles to define
subscriber parameters. The per-session rate limiting parameter is defined
in the RADIUS.
Rate limiting
Per-session rate limiting is a traffic regulation mechanism that allows you

to control the maximum rate of traffic sent or received on an interface for a
session. The rate limiting feature uses the modular quality of service (QoS)
CLI to provide input and output policing rates for each session. The router
uses policing to manage the access bandwidth policy for PPPoA, PPPoE,
PPP in L2TP (LNS only), and RBE subscriber-based sessions. For these
subscriber-based sessions, the service policy is applied to a predefined
configuration template known as the virtual template interface.
Per-user multiservice rate limiting allows for the control of the maximum
rate of traffic for each user behind a multiservice subscriber connection.
This rate limiting feature also uses the modular QoS CLI to provide input
and output policing rates for each user. Users are distinguished from each
other by the use of unique ACLs for each user behind the subscriber.
802.1p Classification and Marking
The class of service (CoS)-based packet matching and marking feature

enables the router to interoperate with switches to deliver end-to-end QoS.
The IEEE 802.1p standard allows QoS to classify inbound Ethernet
packets, based on the value in the CoS field, and to explicitly set the value
in the CoS field of outbound packets.
924
Version 1.0
Module 9
QoS
Per-session service policy for PTA and PPP in L2TP (LNS side)
Rate limiting
Per-user multiservice
Per-session
Quick connect
Per-session service policy using RADIUS VSAs
802.1p classification and marking

Per-session CBWFQ
65,536 queues per system
32 Queues per interface (includes class default and pk_priority)
Dynamic bandwidth selection
Version 1.0
925
Module 9

QoS (continued)
Per-Session CBWFQ
The Cisco 10000 router supports class-based weighted fair queuing

(CBWFQ) for virtual access interfaces. A virtual access interface can
inherit the service policy of the VC that the virtual access interface uses.
Any virtual access interface that uses that VC is subject to the queuing,
policing, and marking actions defined in the VC service policy. You apply a
service policy with queuing-related actions to a VC, not to a virtual
template. The Cisco 10000 router supports queuing only when you apply
the service policy to a VC.
______________________________ Note __________________________
You can apply a service policy without queuing-related actions to either
a VC or a virtual template, but not to both at the same time.
_____________________________________________________________
Queues
The system supports 65,536 queues with a maximum of 32 queues per

interface. Two of these queues are reserved, one for class default and one
for pk_priority traffic. The OC-48 line card requires the use of an
additional reserved control queue .
Dynamic Bandwidth Selection (DBS)
The Cisco 10000 router supports the dynamic bandwidth selection (DBS)
feature for ATM VCs. DBS dynamically changes ATM traffic-shaping
parameters based on a subscribers RADIUS profile. This profile contains
QoS traffic-shaping parameters such as peak cell rate (PCR), sustained cell
rate (SCR), and VC traffic management class (VBR or UBR).
926
Unspecified bit rate (UBR) service class The router applies the PCR
parameter to a UBR configured VC.
Variable bit rate nonreal time (VBR-nrt) service class The VBR-nrt
service class provides QoS to the VC in no atm pxf queuing mode. The
router applies the PCR and SCR parameters to a VBR-nrt configured
VC.
Version 1.0
Module 9
QoS
Per-session service policy for PTA and PPP in L2TP (LNS side)
Rate limiting
Per-user multiservice
Per-session
Quick connect
Per-session service policy using RADIUS VSAs
802.1p classification and marking

Per-session CBWFQ
65,536 queues per system
32 Queues per interface (includes class default and pk_priority)
Dynamic bandwidth selection
Version 1.0
927
Module 9
Leased-Line Features and Scaling

The following leased-line features and scaling numbers are for leased-line
functionality with the PRE-2 image.
Scaling
Up to 4,000 dedicated sessions PPP and HDLC
Up to 8,000 ATM VCs when using atm pxf queuing
800 BGP peers
Total routes
950,000 global routes
MPLS VPNs support 999 VPN routing and forwarding instances

(VRFs) containing 350,000 total routes
MLP 1,250 bundles, up to 10 interfaces per bundle, up to 2500 links
The 2,500 links include both MLP and multilink Frame Relay
(MLFR)
QoS
928
Priority queuing (PQ) and CBWFQ up to 65,536 queues
Police committed access rate (CAR) and rate limiting 2-color policer
with 2-color functionality
Access control lists (ACLs) All ACLs with more than 8 lines are
converted to turbo ACls.
Nested policy maps
Quality of service policy propagation through Border Gateway Protocol

(QPPB)
Class-based traffic shaping using modular QoS CLI (MQC)
Version 1.0
Module 9
Scaling
Up to 4,000 dedicated sessions PPP and HDLC

Up to 8,000 ATM VCs with ATM PXF queuing
800 BGP peers
950,000 global routes
MPLS VPN 999 VRFs with 350,000 routes
MLP 1,250 MLP bundles, up to 10 interfaces per bundle, up to 2,500 links
QoS
PQ/CBWFQ
Police (CAR, rate limiting)
ACLs
Nested policy maps
QPPB
Class-based traffic shaping
Version 1.0
929
Module 9
Leased-Line Features and Scaling (continued)

ATM QoS
VBR-nrt per VC queuing
VBR-nrt per VC weighted random early detection (WRED)
ATM class of service shaped UBR, UBR+, VBR-nrt, and committed bit
rate (CBR)
1000 groups, 8000 subinterfaces
6400 multicast routes
PE and P Functionality
Label Distribution Protocol (LDP)
Tag Distribution Protocol (TDP)
Supports up to 6 labels
Multicast
MPLS VPN
930
Version 1.0
Module 9
Leased-Line Features and Scaling (continued)
ATM QoS
VBR-nrt per VC queuing
VBR-nrt per VC WRED
ATM CoS shaped UBR, UBR, VBR-nrt & CBR
Multicast
1000 groups, 8000 subinterfaces
6400 multicast routes
MPLS VPN
PE and P functionality
LDP
TDP
Supports up to 6 labels
Version 1.0
931
Module 9
Lease-Line Features and Scaling (continued)

MPLS TE
932
Headend, tailend, midpoint functionality
Routes static, IS-IS and Open Shortest Path First (OSPF) single level
Autoroute calculation
AutoBandwidth
InterArea TE (OSPF)
Label VC (LVC) and TE over ATM interface (Frame Mode)
384 Tunnels (headend and midpoint)
Version 1.0
Module 9
Lease-Line Features and Scaling (continued)
MPLS TE
Headend, tailend, midpoint functionality
Routes static, IS-IS & OSPF single level
Autoroute calculation
AutoBandwidth
InterArea TE (OSPF)
LVC and TE over ATM interface (Frame mode)
384 Tunnels (headend & midpoint)
Version 1.0
933
Module 9
High Availability and Management Functionality

High Availability
The Cisco 10000 router currently supports a wide range of high availability
features. These features are briefly described below:
RPR+ Standby PRE booted with configuration loaded and interface

connections ready to switch over. Routing tables need to be built and
connections reestablished in RPR+.
Fast Software Upgrade (FSU) Reduced time to update software.

Processor switchover time will be based on RPR+.
Online insertion and removal (OIR) all modules
SONET automatic protection switching (APS), Synchronous Digital

Hierarchy (SDH) SDH multiservice switching path (MSP) 50 msec
switchover times for APS and MSP.
Hot Standby Routing Protocol (HSRP) with 802.1Q Supports 802.1q

VLANs in addition to interfaces.
Management
System management includes Netflow accounting, ongoing MIB
enhancements, and reverse path forwarding functionality.
934
Netflow accounting (v1, 5, 8)
MIB enhancements
RPF strict
Version 1.0
Module 9
High Availability
RPR+
Fast software upgrade
OIR (all modules)
SONET APS, SDH MSP
HSRP (with 802.1Q)
Management
Netflow accounting (v1, 5, 8)
MIB enhancements
RPF strict
Version 1.0
935
Module 9
QoS Features and Functions

The pages that follow list supported system-level QoS features and
functions.
936
Class-map match options
Policy-map keywords and actions
QoS facts
Version 1.0
Module 9
Class-map match options

Policy-map keywords and actions
QoS Facts
Version 1.0
937
Module 9
Class-Map Match Options

Listed below ar e the class-map match options supported on the Cisco
10000 router with PRE-2. The match option defines the matching criteria
for a class map.
938
IP Precedence Supporting precedence values 07
IP differentiated services code point (DSCP) Full DSCP support for

values 063
QoS group Up to 99 QoS groups
Input interface matches
Numbered and named ACLs Access group support for numbered and
named ACLs
IP Real-Time Transport Protocol (RTP)
CoS (802.1Q class of service/user priority values) Permits matching

on inbound CoS value (07).
Version 1.0
Module 9
Match criteria for a class map

IP Precedence
IP DSCP
QoS Group
Input interface matches
Numbered and named ACLs
IP Real-Time Transport Protocol (RTP)
COS (802.1Q class of service and user priority
values)
Version 1.0
939
Module 9
Policy-Map Keywords
The following are policy-map keywords for setting QoS parameters,
policing traffic, determining the method of congestion control, controlling
queue depth, and allocating bandwidth.
940
set atm-clp
set ip dscp
set ip precedence
set mpls experimental
set qos-group
police bps
random detect precedence
queue-limit
bandwidth
priority
shape
Version 1.0
Module 9
Policy-Map Keywords
Policy Map Keywords
Keywords for policy maps
Set
ATM CLP
IP DSCP
IP precedence
MPLS experimental
QoS group
Police bps
Random detect precedence
Queue-limit
Bandwidth
Priority
Shape
Version 1.0
941
Module 9
Policy-Map Actions
The Cisco 10000 router does not impose any restrictions on the
classification definitions you include in the class map. However, it does
limit the input and output policy actions that you can define in a policy
map. These limitations are based on the type of interface on which you
apply the service policy as indicated in the tables that follow.
The tables indicate the following types of interfaces:
Normal interface, including VBR VCs on ports configured in atm pxf

queuing mode
Tag/MPLS interface (MPLS VPN)
Virtual-access interface
ATM UBR VCs and VCs configured on ports in no atm pxf queuing
mode
The tables indicate one of the following possibilities for applying a service
policy to an interface:
Valid the action is valid for this interface.
Not Applicable (N/A) indicates that you cannot perform the action or
that it has no meaning in the context.
Not Available the action is not supported.
Input Policy Actions

The table that follows shows the possible policy actions that you may apply
to input interfaces.
942
Version 1.0
Module 9
Policy-Map Actions
Input Policy Map Actions
Action
Normal Interface/
MPLS Int.
VBR VC
UBR VC/
Virtual
Access Int. No PXF Queue
Bandwidth
N/A
N/A
N/A
N/A
Queue-limit
N/A
N/A
N/A
N/A
Priority
N/A
N/A
N/A
N/A
Shape
Not Available Not Available Not Available Not Available
Random-detect
N/A
Set Prec/DSCP
Valid
Set QoS-group
N/A
N/A
N/A
N/A
Valid
Valid
Valid
Valid
Valid
Valid
Set ATM-CLP
N/A
N/A
N/A
N/A
Set COS
N/A
N/A
N/A
N/A
Police
Valid
Valid
Valid
Valid
Set MPLS-exp Not Available Not Available Not Available Not Available
Version 1.0
943
Module 9
Policy Map Actions (continued)

Output Policy Actions
The table that follows shows the possible policy actions that you may apply
to output interfaces.
944
Version 1.0
Module 9
Policy-Map Actions
Output Policy Map Actions
Action
Normal Interface/
MPLS Int.
VBR VC
Virtual
UBR VC/
Access Int. No PXF Queue
Bandwidth
Valid
Valid
On VC not VAI
Queue-limit
Valid
Valid
Not Available Not Available
Priority
Valid
Valid
On VC not VAI
N/A
Shape
Valid
Valid
On VC not VAI
N/A
Random-detect
Valid
Valid
Not Available Not Available
Set Prec/DSCP
Valid
N/A
Valid
Valid
Set QoS-group
N/A
N/A
N/A
N/A
Set ATM-CLP
Valid
Not Available Not Available Not Available
Set COS
Valid
Not Available
Police
Valid
Set MPLS-exp
N/A
N/A
Valid
N/A
Valid
Valid
Valid
Not Available
N/A
N/A
Version 1.0
945
Module 9
QoS Facts
The following are supported QoS capabilities and scaling numbers to help
you plan and use the Cisco 10000 router in your network.
946
Up to 16,384 access lists per system with 8 access control entries

(ACEs) or less.
Up to 16 match statements per class-map.
Class maps per policy 256 (future 64). With the reduction to 64 class
maps per policy, the number of policy maps per system will change
from 256 to 4096.
Class maps per system: 1000 (includes default).
Policy-Maps per system 256 (future 4,000). This increase in policy maps
per system will occur in conjunction with the decrease in class maps
per policy from 256 to 64.
A packet subject to only one policy (source or d estination interface)

requires one pass through the PXF.
A packet whose source and destination interfaces each have a policy

requires two passes through the PXF.
Max QoS sessions : 8,000.
Version 1.0
Module 9
QoS Facts
QoS Facts
Supported QoS capabilities

Up to 16,384 access lists per system
Up to 16 match statements per class map
Class maps per policy 256 (future 64)
Class maps per system 1,000 (includes default)
Policy maps per system 256 (future 4,000)
A packet subject to only one policy (source or destination
interface) requires 1 pass
A packet whose source and destination interfaces each

have a policy requires 2 passes
Max QoS sessions: 8000
Version 1.0
947
Module 9
QoS Facts (continued)
948
Max sessions police in or police out: 32000
Max sessions 802.1Q match or mark: 4096
Max number of mini ACLs (8 entries or less) per system: 16000
Max number of turbo ACLs per system: 570
Max number of ACEs per ACL: 8000
Maximum committed bandwidth for an interface is 99%
Version 1.0
Module 9
QoS Facts
Max sessions police in or police out: 32000

Max sessions 802.1Q match or mark: 4096
Max number of mini ACLs (8 entries or less) per
system: 16000
Max number of turbo ACLs per system: 570

Max number of ACEs per ACL: 8000
Maximum committed bandwidth for an interface
is 99%
Version 1.0
949
Module 9

Traffic subject to QoS
All in-transit IP packets
Locally destined or originated IP packets that are not control packets
Queues
65,536 total queues
32 queues per interface
Queue numbers: 0 default, 31 pk_priority (OC-48 only, 30 control)
Queue sizes are a power of 2 from 32 to 16,384 packets (rounded up to next

legal value). That is, the legal value is allocated, and the specified value is
actually used.
950
Version 1.0
Module 9
QoS Facts
Traffic subject to QoS
All in-transit IP packets

Locally destined or originated IP packets that are
not control packets
64K Queues - 32 per interface
0 default, 31 pk_priority (OC-48 only, 30
control)
sizes are a power of 2 from 32 to 16384 packets
(rounded up to next legal value)
Version 1.0
951
Module 9
Policing Considerations
Consider the following when specifying the data rate for policing. The
committed rate includes framing overhead for non-ATM interfaces, and
with ATM interfaces it includes cell overhead.
For example, to specify a committed rate of 1 kbps supporting 64-byte
packets, use the following formulas:
952
Non-ATM interfaces 1 kbps * (64 + 4) * 8 = 544kbps (with 4 bytes of

framing overhead)
ATM interfaces 1 kbps * 2 * 53 * 8 = 848kbps (each 64-byte packet

uses 2 cells)
Version 1.0
Module 9
Committed rate includes framing overhead and

cell overhead
For example 1 kbps committed rate for 64 byte

packets
1 kbps * (64 + 4) * 8 = 544 kbps (with 4 bytes of
framing overhead)
1 kbps * 2 * 53 * 8 = 848 kbps (with cell overhead
each 64 byte packet uses 2 cells)
Version 1.0
953
Module 9
VC Scaling with QoS

Overview
High VC count mode (no atm pxf queuing) enables the Cisco 10000 router
to support 61,500 VCs with PPPoE, PPPoA, or RBE protocols. The High VC
count mode is set on a per-port basis, and it imposes certain limitations,
regardless of how the VCs are defined.
With Low VC count mode (atm pxf queuing), the supported VC count is
limited to 8,000 VCs; however, this mode permits extensive QoS
functionality.
______________________________ Note __________________________
If you intend to disable ATM PXF queuing, to ensure reliable operation,
you must enter the no atm pxf queuing command before you configure
any VCs on an interface. If you have already configured VCs on an
interface, and you need to disable ATM PXF queuing, remove the VCs
from the configuration and then change the ATM PXF queuing mode.
_____________________________________________________________
QoS Support with no atm pxf queuing

The Cisco 10000 router supports the following QoS features when no atm
pxf queuing is enabled on an interface or a UBR VC:
Rate limiting on each session in the input, output, or both input and
output directions
The set qos-group (input only), set ip precedence, and set ip dscp policy
map actions
The following QoS features are not supported when no atm pxf queuing
is enabled on an interface or a UBR VC:
954
Weighted fair queuing (WFQ)
Weighted random early detection (WRED)
Traffic shaping for IP and PPP
Version 1.0
Module 9
VC Scaling with QoS
VC Scaling with QoS
High versus low VC count

61,500 VCs vs 8,000VCs
Set on a per-port basis using the no atm pxf queuing
command
For PPPoA, PPPoE, PPP in L2TP, and RBE sessions
Supports
Rate limiting on each session in the input, output, or both
input and output directions

The set qos-group (input only), set ip precedence, and set ip
dscp policy map actions
Does Not Support

Version 1.0
955
Module 9
VC Scaling with QoS (continued)

QoS Support with atm pxf queuing
With atm pxf queuing, RBE sessions and virtual -access interfaces inhe rit
the service policy of the VC. The router supports the following QoS features
with atm pxf queuing and when the VC is a VBR VC:
956
Rate limiting on each session in the input, output, or in both the input
and output directions
The set qos-group (input only), set ip precedence, and set ip dscp
policy map actions
Traffic shaping
Version 1.0
Module 9
VC Scaling with QoS
VC Scaling with QoS (continued)
High versus low VC count

61,500 VCs vs 8,000VCs
Set on a per-port basis using the no atm pxf queuing
command
For PPPoA, PPPoE, PPP in L2TP, and RBE sessions
Supports
Rate limiting on each session in the input, output, or both
input and output directions

The set qos-group (input only), set ip precedence, and set ip
dscp policy map actions
Does Not Support

Version 1.0
957
Module 9
System Status and Alarms

Use the following commands to display system status and alarms:
958
show version displays boot image, code revision, system operational
show facility-alarm status displays temperature alarm
show environment displays system temperature, fan, and PEM
time, available interfaces, etc.
thresholds and system wide alarms

status
Version 1.0
Module 9
Commands to display system status and

alarms
show version
Code revision, system operational time, available

interfaces
show facility-alarm status
Temperature alarm thresholds, System wide

alarms
show environment
System temperature, fan and PEM status
Version 1.0
959
Module 9
System Status and Alarms (continued)

show version Command
To display system and software information, enter the show version
command. The display output includes the following data:
960
Cisco IOS image
System bootstrap and configuration register
Uptime
Route processor configuration data, memory, flash, etc.
Toaster status
Interface data
Version 1.0
Module 9
show version Command
P2R2#sho version
Cisco Internetwork Operating System Software
IOS (tm) 10000 Software (C10K2-P11-M), Version 12.2(16)BX,RELEASE SOFTWARE(f)
TAC Support: http://www.cisco.com/tac
Copyright (c) 1986-2003 by cisco Systems, Inc.
Compiled Thu 03-Jul-03 16:41 by torowe
Image text-base: 0x600109C4, data-base: 0x61C80000
ROM: System Bootstrap, Version 12.0(20020314:211744)[REL-pulsar_sx.ios-rommonE
BOOTLDR: 10000 Software (C10K2-EBOOT-M), Version 12.2(15)BX, RELEASE SOFTWARE)
P2R2 uptime is 3 minutes
System returned to ROM by reload at 09:54:15 UTC Wed Sep 10 2003
System image file is "disk0:c10k2-p11-mz.122-16.BX.bin"
cisco C10005 (PRE2-RP) processor with 946175K/98304K bytes of memory.
R7000 CPU at 500Mhz, Implementation 39, Rev 4.1, 256KB L2, 8192KB L3 Cache
Backplane version 1.0, 5 slot
Last reset from register reset
PXF processor tmc0 is running.
1 Ethernet/IEEE 802.3 interface(s)
2 FastEthernet/IEEE 802.3 interface(s)
2 Gigabit Ethernet/IEEE 802.3 interface(s)
4 ATM network interface(s)
6 Channelized T3 port(s)
2045K bytes of non-volatile configuration memory.
125440K bytes of ATA PCMCIA card at slot 0 (Sector size 512 bytes).
65536K bytes of Flash internal SIMM (Sector size 512KB).
Configuration register is 0x2102
Version 1.0
961
Module 9

show facility-alarm status Command
The show facility-alarm status command displays critical, major, and
minor alarms caused by line errors and chassis components alarms.
Most alarms are caused by one of the following line card error conditions:
Loss of signal
Loss of framing
No carrier
Temperature, fan, power, and secondary cutover alarm conditions also

exist.
962
Temperature alarms use two temperature sensors in the C10K: one at

the air intake and one on the PRE motherboard.
The blower assembly contains multiple fans. If any of them fail, a fan
alarm occurs. If multiple fan failures occur, or if the blower assembly is
removed for several minutes, the PRE goes into an overtemperature
condition.
The power alarm is asserted if one of the power modules is defective.
Secondary cutover can also cause an alarm. If the primary PRE in a

redundant system fails, the other PRE takes over and the alarm is
asserted.
Alarm conditions and the clearing of alarms are logged to the console.
Version 1.0
Module 9
show facility-alarm status Command
P2R2#show facility-alarm status
Thresholds:
Intake minor 40 major 49 critical 67 Core minor 45 major 53 critical 85
System Totals Critical: 15 Major: 18 Minor: 0
Source: T3 1/0/0 Severity: MAJOR ACO: NORMAL Description: Near End detects
Loss Of Signal Failure
Source: T3 1/0/0 Severity: MAJOR ACO: NORMAL Description: Other Failure
Source: T3 1/0/0 Severity: MAJOR ACO: NORMAL Description: Physical Port Link
Down
Source: T3 1/0/1 Severity: MAJOR ACO: NORMAL Description: Physical Port Link
Down
Version 1.0
963
Module 9

show environment Command
The Cisco 10000 router has two power supplies, one fan tray, and two
temperature sensors. To show the current state of all of the environmental
sensors, use the show environment command.
964
Version 1.0
Module 9
show environment Command
P2R2#show environment
Temperature normal: chassis inlet measured at 30C/86F
Temperature normal: chassis core measured at 32C/89F
Fan:
OK
Power Entry Module 1 type AC status:
OK
Version 1.0
965
Module 9
Checking the Data Path

The following are actions that you can perform to identify problems specific
to a data path or interface on the box.
Check inbound traffic on a particular interface.
Check PXF counters by interface.
Are packets being dropped for any reason?
Check the status of the PXF CEF table.
966
Are packets being dropped? If so why?
Check the Output Packet Buffer Queue .
Are good packets being recognized?
Are the entries correct?
Version 1.0
Module 9
Are there problems specific to a data path

or interface?
Check inbound traffic on a particular interface.
Are good packets being recognized?
Check PXF counters by interface.
Are packets being dropped?
If so why?
Check the Output Packet Buffer Queue.
Are packets being dropped for any reason?
Check the status of the PXF CEF table.
Are the entries correct?
Version 1.0
967
Module 9
Checking the Data Path (continued)

show interface Command
The show interface command provides a quick indication of the following:
968
Layer 2 status
Input queue information
Output queue information
Queuing strategy
Inbound and outbound traffic statistics
Version 1.0
Module 9
show interface Command
P2R2#sho int g2/0/0

GigabitEthernet2/0/0 is up, line protocol is up
Hardware is Half-height Gigabit Ethernet MAC Controller, address
is 0005.dc39)
Internet address is 172.16.2.22/24
MTU 1500 bytes, BW 1000000 Kbit, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
Unknown duplex, Unknown Speed, link type is autonegotiation,
media type is SX
output flow-control is XOFF, input flow-control is XOFF
ARP type: ARPA, ARP Timeout 04:00:00
Last input 00:00:01, output 00:00:08, output hang never
Last clearing of "show interface" counters never
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output
drops: 0
Queueing strategy: fifo
Output queue: 0/40 (size/max)
5 minute input rate 0 bits/sec, 0 packets/sec
5 minute output rate 0 bits/sec, 0 packets/sec
306172 packets input, 47829691 bytes, 0 no buffer
Received 8 broadcasts, 0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog, 159851 multicast, 0 pause input
0 input packets with dribble condition detected
491365 packets output, 41279876 bytes, 0 underruns
0 output errors, 0 collisions, 1 interface resets
0 babbles, 0 late collision, 0 deferred
4 lost carrier, 0 no carrier, 0 pause output
0 output buffer failures, 0 output buffers swapped out
Version 1.0
969
Module 9

show pxf interface Command
The show pxf interface <interface> command helps you determine
whether there are errors in traffic from a particular interface. The source
of errors can be extracted by using the details option for the command .
Displaying the details will help you determine the Layer 2 and Layer 3
causes of lost input packets. The details option of this command is
presented next.
The following PXF interface statistics are not associated with an interface
but are at the card level:
970
FBB Rx Cobalt input packet memory stopped accepting packets from

the line card. This counter shouldnt normally increment; if it does
theres a problem.
PXF DMA RE drops Receive error drops are the result of packets
corrupted on the backplane. This counter may increment during OIR or
PRE failover.
Null config drops Null config drops will occur prior to virtual channel
common index (VCCI) assignment when nothing is configured on an
interface (packets will be dropped). This counter should not increment
once the interface is up, but it may increment during OIR.
Version 1.0
Module 9
show pxf interface Command
P2R2# sho pxf interface atm3/0/0

ATM3/0/0 is up, enabled, PXF enabled, IOS encap ATM
(33)
Last clearing of ATM3/0/0 counters: never
Total PXF input errors (pkts/bytes):
0/0
Slot 3/0: FBB Rx: 0x00003000 OCQ debug: 0x00006080, qN_entry_cnt[5:0]: 0
PXF DMA RE drops: 0/0, Null config drops: 0/0
Last clearing of slot 3/0 counters: never
Version 1.0
971
Module 9

show pxf interface detail Command
The show pxf interface <interface> detail command displays packet
drop errors and statistical data. These statistics are accumulated in
column zero of the PXF. The illustration provides a display of the command
with available statistics.
972
Version 1.0
Module 9
show pxf interface detail Command
P2R2# sho pxf interface atm3/0/0 detail

ATM3/0/0 is up, enabled, PXF enabled, IOS encap ATM
Min MTU: 4
Max MTU: 4482
VCCI 0x9D0, ICMP address 0.0.0.0
Last clearing of ATM3/0/0 counters: never
Total PXF input errors (pkts/bytes):
0/0
VCCI undefined:
0/0
in l2 max mtu
0
0
in l2 min mtu
0
0
encap not supported
0
0
mlfr fragament
0
0
mpls not enabled
0
0
ip version
0
0
ip header length
0
0
ip length max
0
0
ip length min
0
0
ip checksum
0
0
fib rpf fail
0
0
acl denied
0
0
ttl
0
0
unreachable
0
0
df multicast
0
0
car input drop
0
0
car output drop
0
0
out l2 max mtu
0
0
out l2 min mtu
0
0
tunnel no match
0
0
(33)
Slot 3/0: FBB Rx: 0x00001000 OCQ debug: 0x00006080,

qN_entry_cnt[5:0]: 0
PXF DMA RE drops: 0/0, Null config drops: 0/0
Last clearing of slot 3/0 counters: never
Version 1.0
973
Module 9

show pxf cpu queue interface Command
The show pxf cpu queue <interface> command provides a view of the
PXF output queues (located in the PXF output packet buffer). Each entry
includes the class ID, class name (the type of queue or a configured queue ),
queue identifier (QID), its length, and whether or not packets have been
dropped.
In the sample output in the illustration, notice the following information
under the column headings ClassID, ClassName , and QID:
A queue named class-default with a QID of 14
A queue named pk-priority with a QID of 2
A queue named label with a QID of 16
By default all interfaces, except the OC-48 POS line card, will have two
queues, class-default with a class ID of 0 and pk-priority with a class
ID of 31. The result is that 30 additional class queues can be configured on
an interface.
The OC-48 interface has an additional queue by default. This queue is the
control queue with a class ID of 30. Additionally, the OC-48 card uses a
buddy queue system so each queue appears twice with two individual
QID numbers.
974
Version 1.0
Module 9
show pxf cpu queue interface Command
P2R2#sho pxf cpu que g2/0/0

VCCI 2511:
VCCI/ClassID ClassName
2511/0 class-default
2511/1
label
2511/31
-
QID
14
16
2
Length/Max Res
0/132
4
0/256
4
0/32
4
Dequeues
0
0
327026
Drops
0
0
0
Legend:
$: Priority Queue
~: RED Queue
P: MLP Pkt Queue
F: MFR Pkt Queue
M: Max Utilization Pkt Queue
Version 1.0
975
Module 9

show pxf cpu queue qid Command
The show pxf cpu queue <QID> command provides detailed information
on the interface queues such as committed information rate (CIR), excess
information rate (EIR), and maximum information r ate (MIR). Also
included are queue drop counts along with the reason. Use this command
to determine why a particular queue is dropping traffic.
The following provide an explanation of dropped packets:
Tail Drops The queue is full and operating in FIFO mode and has
exceeded the maximum threshold value.
Random Drops The WRED is configured, there is congestion on the

interface, and the minimum queue threshold has been exceeded.
Max Threshold The WRED maximum threshold has been exceeded

and all traffic is being dropped.
No Packet Handle The packet is dropped by the Cobalt ASIC because

the ASIC is out of packet handles.
Buffer Low The PXF is r unning low on output buffers and is holding
those available for high-priority traffic.
For conveniences, definitions of the CIR, EIR, and MIR parameters are
given below:
976
Committed information rate (CIR) represents the guaranteed

bandwidth that is assigned via the priority, bandwidth and police
commands.
Excessive information rate (EIR) excess bandwidth, distributed

proportionally to queues based on assigned bandwidth, which occurs
when CIR = EIR. The remaining bandwidth is the bandwidth available
after the priority queue, if present, is given its required bandwidth.
Bandwidth distribution may be overridden by the remaining bandwidth
percent command.
Minimum information rate (MIR) shows the proportion of the

bandwidth that has been assigned by the shape command.
Version 1.0
Module 9
show pxf cpu queue qid Command
P2R2#sho pxf cpu que 16

ID
: 16
CIR (in-use/configured)
: 2/2
EIR (in-use/configured)
: 0/0
MIR (in-use/configured)
: 65535/65535
Link
: 2
Flowbit (period/offset)
: 16384/16384
Burst Size
: 8000 bytes
Actual Bundle FIFO Size
: 0
Bandwidth
: 1000008 Kbps.
Channel
: 0
Packet Descriptor Base
: 0x00001000
ML Index
: 0
Length/Average/Max/Alloc
: 0/0/256/256
Enqueues (packets/octets)
: 0/0
Dequeues (packets/octets)
: 0/0
Drops (tail/random/max_threshold)
: 0/0/0
Drops (no_pkt_handle/buffer_low)
: 0/0
WRED (weight/avg_smaller)
: 0/0
WRED (next qid/drop factor)
: 0/0
WRED (min_threshold/max_threshold/scale/slope):
precedence 0
: 0/0/0/0
precedence 1
: 0/0/0/0
precedence 2
: 0/0/0/0
precedence 3
: 0/0/0/0
precedence 4
: 0/0/0/0
precedence 5
: 0/0/0/0
Version 1.0
977
Module 9

show pxf cpu cef Command
Use the show pxf cpu cef command to check the PXFs FIB table to
determine if the destination route is present.
978
Version 1.0
Module 9
show pxf cpu cef Command
P2R2#sho pxf cpu cef

Shadow 10-10-6-6 Toaster Mtrie:
71 leaves, 3692 leaf bytes, 40 nodes, 37600 node bytes
102 invalidations
168 prefix updates
refcounts: 9377 leaf, 9280 node
Prefix/Length
0.0.0.0/0
52.10.0.0/16
52.10.0.0/32
52.10.10.20/32
52.10.100.22/32
52.10.255.255/32
52.20.0.0/16
52.20.0.0/32
52.20.0.22/32
52.20.0.101/32
52.20.255.255/32
127.0.0.0/8
127.0.0.0/32
127.0.0.10/32
127.0.0.11/32
Refcount
5419
460
4
4
4
4
334
4
4
4
4
2368
4
4
4
Parent
0.0.0.0/0
52.10.0.0/16
52.10.0.0/16
52.10.0.0/16
52.10.0.0/16
0.0.0.0/0
52.20.0.0/16
52.20.0.0/16
52.20.0.0/16
52.20.0.0/16
0.0.0.0/0
127.0.0.0/8
127.0.0.0/8
127.0.0.0/8
Version 1.0
979
Module 9
System-Wide Statistics and Performance

You can use various system-level commands to observe traffic loading and
other conditions that affect the performance of the router. The following
system level checks are explained on the pages that follow:
980
Check overall PXF statistics to determine system loading
Check PXF IP, drop, and diversion statistics
Check the PXF route processor queues
Check overall status of PXF buffer handles
Check the status of the PXF microcode
Version 1.0
Module 9
System Wide Statistics and Performance
Commands to check system level

performance
Check overall PXF statistics to determine
system loading
Check PXF IP, drop, and diversion statistics

Check the PXF route processor queues
Check overall status of PXF buffer handles
Check the status of the PXF microcode
Version 1.0
981
Module 9
System-Wide Statistics and Performance (continued)

show pxf cpu context Command
The show pxf cpu context command provides an indication of the current
load on the PXF in addition to the average load for the past minute,
5 minutes, and hour.
The value of null contexts in the illustration is approximately 6.3 million
contexts per second. That is the maximum rate of the PXF without
considering overhead. Overhead increases with traffic load; therefore, this
number will decline as the traffic increases.
The following are descriptions of the parameters shown in the show pxf
cpu context display output:
982
new_work New packets input to the PXF pipeline
from_ lc These are new packets that have come directly from a line
card.
from_rp New packets into the pipeline that were generated by the
route processor.
from_replay This counter represents traffic such as fragmented or

multicast packets that are sent back to the PXF to have a header
written for each fragment or multicast destination interface.
feed_back Packets requiring additional passes through the pipeline

such as those requiring both input and output QoS. The feed_back
counter is incremented one time for each additional pass that a packet
makes.
Null Contexts that do not contain packet headers. Available PXF

bandwidth.
Actual = (feedback + new) / (feedback + new + null)
Theoretical = (feedback + new) / 6.3mpps
Maximum = (feedback + new + null) / 6.3mpps
Version 1.0
Module 9
show pxf cpu context Command
P2R2#sho pxf cpu context

FP context statistics
count
--------------------------------feed_back
113111
new_work_from_lc
2808661
new_work_from_rp
3162988
new_work_from_replay 0
null_context
10991592472601
FP average context/sec
--------------------feed_back
new_work_from_lc
new_work
new_work_from_replay
null_context
--------------------Total
FP context utilization
--------------------Actual
Theoretical
Maximum
1min
---------0
1
1
0
6315522
---------6315525
1min
---------0
%
0
%
98 %
rate (since last time command was run)

---------0
2
1
0
6315522
---------6315526
5min
---------0
1
1
0
6315522
---------6315525
5min
---------0
%
0
%
98 %
Version 1.0
60min
---------0
1
1
0
6315677
---------6315680
cps
cps
cps
cps
cps
cps
60min
---------0
%
0
%
98 %
983
Module 9

show pxf statistics ip Command
The show pxf statistics ip command provides an overall picture of
system-wide IP statistics by traffic type and action taken.
The following are descriptions of some of these statistics:
984
IP dropped FIB, MFIB, TFIB lookups in Column 1 are not resolved.
Punted Locally addressed packets such as keepalives, link negotiation

(LCP, NCP), glean adjacency (resolve MAC, IP, Frame Relay
addresses), etc. that are diverted to the route processor.
No adjacency CEF has the prefix but cant resolve the address.
No route Router does not contain information for this prefix.
IP unicast RPF A packet has been dropped as the result of a RPF

failure.
ICMP created TTL expired, unreachable, etc.
IP packets fragmented Increment the counter for the first packet

fragment.
Fragments A count of all subsequent packet fragments.
Failed An attempt was made to fragment the packet and it failed.
Version 1.0
Module 9
show pxf statistics ip Command
P2R2# sho pxf statistics ip

Chassis-wide PXF forwarding counts
IP inputs 62851, forwarded 25055, punted 2
IP dropped 36153, no adjacency 0, no route 73960
IP unicast RPF 0, unresolved 0
ICMP created 37807, Unreachable sent 36179, TTL expired sent 0
ICMP echo requests 0, replies sent 0
ICMP checksum errors 0
IP packets fragmented 0, total fragments 0, failed 0
IP don't-fragment 0, multicast don't-fragment 0
IP mcast total 28037, switched 0, punted 28037, failed 0
IP mcast drops 0, RPF 0, input ACL 0, output ACL + taildrops 0
Last clearing of PXF forwarding counters: 3d02h
Version 1.0
985
Module 9

show pxf statistics drop detail Command
The show pxf statistics drop detail command provides information
about PXF input traffic that was not assigned to a particular interface but
that was dropped for the reasons noted. These statistics are for the entire
system.
986
Version 1.0
Module 9
show pxf statistics drop detail Command
P2R2# sho pxf statistics drop detail

PXF input drops:
Unassigned drops (pkts/bytes):
0/0
PXF Unassigned input drop details:
(These input drops are not assigned to a particular PXF interface.)
packets
bytes
generic
0
0
mpls_no_eos
0
0
fib_zero_dest
0
0
fib_drop_null
0
0
fib_icmp_no_adj
36172
3327824
fib_icmp_bcast_dst 0
0
mfib_ttl_0
0
0
mfib_disabled
0
0
mfib_rpf_failed
0
0
mfib_null_oif
0
0
tfib_rp_flag
0
0
tfib_eos_violation 0
0
tfib_nonip_expose
0
0
tfib_label_invalid 0
0
tfib_path_unknown
0
0
tfib_nonip_ttl_exp 0
0
icmp_unrch_interval 331
40190
icmp_on_icmp
1299
166164
icmp_bad_hdr
0
0
icmp_multicast
0
0
icmp_frag
0
0
no_touch
0
0
enq_id_0
0
0
no_pkt_handles
0
0
ipm_replay_full
0
0
re_bit[00]
0
0
inv_resource[00]
0
0
null_config[00]
0
0
Last clearing of drop counters: 3d02h
Version 1.0
987
Module 9

show pxf statistics diversion Command
The show pxf statistics diversion command provides statistics for all
traffic that the PXF has forwarded to the route processor for processing.
The statistics provide the reason why the traffic is diverted or punted.
988
Version 1.0
Module 9
show pxf statistics diversion Command
P2R2# sho pxf statistics diversion

Diversion Cause Stats:
divert
= 0
encap
= 0
clns_isis = 0
clns
= 0
cdp
= 0
cgmp
= 0
arp
= 202
rarp
= 0
mpls_ctl = 0
keepalive = 0
ppp_cntrl = 171118
fr_lmi
= 0
atm ilmi = 0
oam f4
= 0
oam f5 ete= 0
oam f5 seg= 0
mlfr lip = 0
srp topo = 0
srp ips
= 0
ip version= 0
option
= 20000
fib bcast = 2
fib glean = 0
fib dest = 13
tfib flag = 0
tfib ver = 0
tfib opts = 0
mfib 224 = 28094
igmp
= 0
p2p_prune = 0
assert
= 0
null_out = 0
direct
= 0
join_spt = 0
register = 0
no_fast
= 0
local_mem = 0
dat_prune = 0
no_group = 0
nest frag = 0
pbr arp
= 0
ipc_resp = 0
fp ipc
= 0
pppoe disc = 2
atm crl
= 0
fib rp punt = 0
l2tp control = 0
acl punt = 0
Version 1.0
989
Module 9

show pxf cpu queue Command
The system maintains eight queues for communication from the PXF to
route processor. The show pxf cpu queue command provides data on the
status of each of these queues.
The eight queues are used as follows:
990
QID
Traffic Type
256
Default
257
Netflow
258
ACL Log
259
L2 Ctrl
260
Routing
261
Future
262
Keepalive
263
ATM CRL
Version 1.0
Module 9
show pxf cpu queue Command
P2R2# sho pxf cpu que

VCCI/ClassID ClassName
1/0
1/1
1/2
$
1/3
$
1/4
$
1/5
$
1/6
1/7
-
QID
263
262
261
260
259
258
257
256
Length/Max Res
0/1024
1
0/1024
1
0/1024
1
0/1024
1
0/1024
1
0/1024
1
0/1024
1
0/1024
1
Dequeues
311853
0
51
1071287
7
0
7
0
Drops
0
0
0
0
0
0
0
0
Legend:
$: Priority Queue
~: RED Queue
P: MLP Pkt Queue
F: MFR Pkt Queue
M: Max Utilization Pkt Queue
Version 1.0
991
Module 9

show pxf cpu buffers Command
The show pxf cpu buffers command provides an indication about
whether or not the system is running low on output particle buffers. Small
buffers use up to 3 handles for packets up to 768 bytes in size , whereas
large buffers will use up to 36 handles for packets up to 9208 bytes in size .
992
Version 1.0
Module 9
show pxf cpu buffers Command
P2R2#sho pxf cpu buffers

Cobalt2 ttc running.
Calculations could be off by (+/ -) cache sizes.
cache size
small
512
large
128
pool
# handles
available
-------------------------------small
524288
523808
large
32768
32626
Version 1.0
993
Module 9

show pxf microcode Command
The show pxf microcode displays the microcode image that is running in
the PXF processors, tells how long ago it was loaded, and gives the status
of the four PXF Toaster processors.
994
Version 1.0
Module 9
show pxf microcode Command
P2R2#sho pxf microcode

PXF
PXF
PXF
PXF
PXF
complex: 4 Toasters 8 Columns total

processor tmc0 is running.
Loaded microcode: system:pxf/c10k2-11-ucode.6.1.0.0

Version: 6.1.0.0
Release Software created Wed 02-Jul-03 17:03
Signature: dc48471c609871e59fe7cf3befc28e90
Microcode load attempted 1 time(s), latest 2w6d ago
tmc0
tmc1
tmc2
tmc3
FG_PC=0
FG_PC=0
FG_PC=0
FG_PC=0
BG_PC=6
BG_PC=6
BG_PC=6
BG_PC=6
WDog=1024
WDog=1024
WDog=1024
WDog=1024
MinPhase=23
MinPhase=23
MinPhase=23
MinPhase=23
SecPreScalerTimer=11542680
Version 1.0
MSecPreScalerTimer=150
995
Module 9
Summary
In this module, you learned the following:
996
The major elements within the Cisco 10000 router operating system
The Cisco IOS dependencies that exist within the Cisco 10000 router
The major services available when using the PRE-2
The features, functions, and interaction of QoS mechanisms supported

on the Cisco 10000 router
Use Cisco IOS commands to determine system firmware, hardware,

and software
Appropriate show commands for displaying pertinent PXF statistics
Version 1.0
Glossary
Version 1.0
Glossary1
Glossary
Technology Acronyms
AAA
authentication, authorization, and accounting
AAL5
ATM adaptation layer 5
ADSL
asymmetrical digital subscriber line
ATM
Asynchronous Transfer Mode
AToM
Anything over MPLS
AV pair
attribute-value pair
BPDU
bridge protocol data unit
BRAS
Broadband Router Aggregation Server
CAC
Call Admission Control
CAR
committed access rate
CHAP
Challenge Handshake Authentication Protocol
CLP
cell loss priority
CoS
class of service
CPE
customer premise equipment
DHCP
Dynamic Host Control Protocol
DSL
digital subscriber line
DSLAM
digital subscriber line access multiplexer
HDLC
High-Level Data Link Control
ICMP
Internet Control Message Protocol
ILMI
Integrated Local Management Interface
IPCP
IP Control Protocol
ISP
Internet service provider
L2TP
Layer 2 Tunneling Protocol
LAC
L2TP access c oncentrator
LDAP
Lightweight Directory Access Protocol
LCP
Link Control Protocol
LDP
Label Distribution Protocol
LMI
Local Management Interface
LNS
L2TP network server
MLP
Multilink PPP
Glossary2
Version 1.0
Glossary
MLPP
Multilink point-to-point
MPLS
Multi-Protocol Label Switching
NAP
network access provider
NAS
Network access server
NAT
Network Address Translation
NCP
Network Control Protocol
NSP
network service provider
provider
PADI
PPPoE Active Discovery Initiation
PADO
PPPoE Active Discovery Offer
PADR
PPPoE Active Discovery Request
PADS
PPPoE Active Discovery Session-Configuration
PADT
PPPoE Active Discovery Termination
PAP
Password Authentication Protocol
PAT
Port Address Translation
PCR
peak cell rate
PDU
protocol data unit
PE
provider edge
PPP
Point-to-Point Protocol
PPPoA
Point-to-Point Protocol over ATM
PPPoE
Point-to-Point Protocol over Ethernet
PPPoEoE
PPPoE over Ethernet
PPPoEo802.1q
PPPoE over 802.1q
PTA
PPP termination and aggregation
PTA-MD
PPP termination and aggregation multi-domain
PVC
permanent virtual connection
QoS
Quality of Service
RADIUS
Remote Authentication Dial-In User Service
RA-MPLS
Remote Access into MPLS
RBE
Routed Bridge Encapsulation
SAM
Subscriber Access and Management
Version 1.0
Glossary3
Glossary
SAR
segmentation and reassembly
SCR
sustained cell rate
SESM
Subscriber Edge Services Manager
SNMP
Simple Network Management Protocol
SSG
Service Selection Gateway
TACACS
Terminal Access Control Access Control Server
TE
traffic engineering
TDP
Tag Distribution Protocol
UBR
unspecified bit rate
UDP
User Datagram Protocol
VBR-nrt
variable bit rate non real time
VC
virtual connection
VCC
virtual channel connection
VCI
virtual connection identifier
VPDN
Virtual private dial-up network
VPI
virtual path identifier
VPN
irtual Private Network
VRF
VPN routing and forwarding
VSA
vendor specific attribute
xTU-R
xDSL Transmission Unit-remote
Glossary4
Version 1.0
Glossary
Cisco 10000 Series Router Acronyms

ASIC
application specific integrated circuit
ACE
Access Control Entry
ACL
access control list
ASP
Automatic Switching Protection
BPE
backplane Ethernet
CAR
committed access rate
CDP
Cisco DiscoveryProtocol
CEF
Cisco Express Forwarding
CBWFQ
class-based weighted fair queuing
CoS
class of service
DMA
direct memory access
DSCP
differentiated services code point
DBS
dynamic bandwidth selection
EHSA
enhanced high system availability
FCRAM
Fast Cycle RAM
FIB
Forwarding Information Base
FPGA
field programmable gate array
GBIC
Gigabit Interface converter
GRE
generic routing encapsulation
HSRP
Hot Standby Routing Protocol
IPC
IOS interprocess communication
IPE
Inter-PRE Ethernet
LoS
loss of signal
MLFR
Multilink Frame Relay
MMF
multimode fiber
MQC
modular QoS CLI
MSP
Multiplex Section Protection
OIR
online insertion and removal
PEM
power entry module
PoS
Packet over SONET
Version 1.0
Glossary5
Glossary
POST
power-on self test
PQ
priority queuing
PRE
Performance Routing Engine
PXF
Parallel eXpress Forwarding
RPR
route processor redundancy
RPF
Reverse Path Forwarding
Rx
receive
SAR
segmentation and reassembly
SDH
Synchronous Digital Hierarchy
SFP
small form factor pluggable
SMF
single-mode fiber
SONET
Synchronous Optical Network
STM
Synchronous Transport Module
TFIB
tag forwarding information base
Tx
transmit
VCCI
Virtual Channel Common Index
WFQ
weighted fair queuing
WRED
weighted random early detection
Glossary6
Version 1.0
Appendix A
Review Question Answers
Appendix Contents
This appendix contains answers to review questions at the end of each
module.
Version 1.0
A1
Appendix A
Module 1 Broadband Aggregation Architectures

1. List the segments that make up a broadband subscriber network
environment.
Customer Premise Equipment, Network Access Provider, and
_________________________________________________________
Network Service Provider
2. A service provider that provides the access connection to the subscriber
and connects the subscriber to the NSP is characteristic of a
wholesale
_________________________
service.
3. Which of the following is not characteristic of a VC service?
a. NAPs do not need to deal with IP address management.
Answer
b. The NAP determines the users encapsulation method.

c. End-to-end provisioning takes time.
d. It is a wholesale service that a NAP would provide.
e. It does not scale well.
4. Which of the following is a reason that RBE is preferred over strict

RFC 1483 bridging?
a. With RBE, the CPE is in routing mode rather than in bridging
mode.
b. The PC encapsulates Layer 3 data into Ethernet.
Answer
c. RBE is more secure and scalable than RFC1483 bridging.

d. RBE is more suitable for business applications.
5. Which of the following statements are true when comparing PPPoA to

PPPoE?
Answer
a. The CPE functions as a router with PPPoA and as a bridge with

PPPoE.
Answer
b. The PPP session is initiated by the CPE with PPPoA and by the PC
with PPPoE.
c. The CPE is able to run NAT for both methods and conserve IP
addresses.
d. PPPoA functions only with ATM access methods and PPPoE
functions only with Ethernet access methods.
Answer
e. When there are multiple users behind the CPE, PPPoE is more
flexible than PPPoA for selection of multiple services.
6. What is the preferred method for authenticating PPP sessions?

______________________________
RADIUS
A2
Version 1.0
Appendix A
Module 1 Broadband Aggregation Architectures
7. When comparing L2TP to PTA, which of the following identify distinct

advantages of L2TP over PTA? Choose two.
Answer
a. PPP sessions may be terminated at the NSP rather than the NAP.
b. L2TP supports multiple protocols.
c. L2TP shares access to core components.
Answer
d. The access provider only looks at the Layer 2 information.
8. What functionality on a Cisco router do managed LNS and RA-MPLS

Virtual routing and forwarding
make use of? __________________________________________________
9. Which of the following distinguishes RA-MPLS from managed LNS?
a. RA-MPLS supports RBE.
b. RA-MPLS allows use of overlapping IP addresses.
Answer
c. RA-MPLS does not require L2TP.

d. RA-MPLS supports PPPoX.
10. What does SSG enable subscribers to do?

Selectively access different services
________________________________________________________________
Version 1.0
A3
Appendix A
Module 2 RBE and RFC 1483

1. How does the CPE function differently between RFC 1483 bridging and
RBE?
a. The CPE functions as a bridge with RFC 1483 bridging and as a
router with RBE.
b. The CPE performs LLC/SNAP or VC multiplexing with RFC 1483
bridging but not with RBE.
c. The CPE will route IP data and bridge all other data.
Answer
d. The is no difference.
2. What is the functional difference at the aggregation device between
RFC 1483 bridging and RBE?
Answer
a. The aggregator functions as a bridge with RFC 1483 bridging and

as a router with RBE.
b. The aggregator performs LLC/SNAP or VC multiplexing with RFC
1483 bridging but not with RBE.
c. For incoming subscriber data, RBE makes forwarding decisions
based on the frame header, and RFC 1483 Routing forwards packets
based upon the Layer 3 header.
d. The is no difference.
3. List two RFC 1483 encapsulation methods for multiplexing and
transporting datalink and network layer protocols over ATM AAL5.
LLC and SNAP

a. __________________________
VC multiplexing
b. __________________________
4. Which of the following ATM interfaces can be used with RBE?
Answer
a. Numbered point-to-point subinterfaces

b. Numbered multipoint subinterfaces
Answer
c. Unnumbered point-to-point subinterfaces

d. Unnumbered multipoint subi nterfaces
5. What must be added to the aggregation router configuration when
using unnumbered interfaces with statically assigned subscriber host
addresses?
Host routes
_________________________________________________________________
A4
Version 1.0
Appendix A
Module 2 RBE and RFC 1483
6. Which of the following configuration methods is preferred for RBE?

a. Numbered interfaces
b. Numbered interfaces with DHCP
c. Unnumbered interfaces
Answer
d. Unnumbered interfaces with DHCP
7. List four parameters that must be configured under the ATM

subinterface to support unnumbered RBE interfaces.
Ip unnumbered loopback
a. _________________________________
Atm routed-bridge
b. _________________________________
pvc
c. _________________________________
encapsulation
d. _________________________________
8. List four advantages of RBE over RFC 1483 bridging.
a. Better
_________________________________________________________
security for IP hijacking and ARP spoofing
broadcast storms
b. Eliminates
_________________________________________________________
Better scaling and performance
c. _________________________________________________________
d. Supports
_________________________________________________________
SSG
9. List four disadvantages of RBE.
Using numbered interfaces wastes IP addresses
a. _________________________________________________________
interfaces without DHCP requires many host routes

b. Unnumbered
_________________________________________________________
configurations increase boot-up time
c. Large
_________________________________________________________
d. Difficult
_________________________________________________________
CPE management
10. Which of the following is common to both RFC 1483 routing and RBE?
a. RFC 1483 routing supports NAT.
b. The CPE functions as a router with RFC 1483 routing.
Answer
c. RFC 1483 routing uses a LLC and SNAP header.

d. Routing updates may be exchanged between the aggregator and
CPE.
Version 1.0
A5
Appendix A
11. Which of the following configuration methods is preferred for RFC 1483
routing?
a. Numbered interfaces
b. Numbered interfaces with DHCP
Answer
c. Unnumbered interfaces
d. Unnumbered interfaces with DHCP
12. List three parameters that must be configured under the ATM
subinterface to support unnumbered RFC 1483 routing interfaces.
Ip unnumbered loopback
a. _________________________________
pvc
b. _________________________________
encapsulation
c. _________________________________
13. List four advantages of RFC 1483 routing.
a. Well
_________________________________________________________
suited for business customers
become manageable
b. CPE
_________________________________________________________
Multiple PVCs for different traffic types
c. _________________________________________________________
d. CPE
_________________________________________________________
can implement NAT or PAT
14. List four disadvantages of RFC 1483 routing.
Not easy for accounting for subscriber traffic
a. _________________________________________________________
authentication of subscribers limits service selection

b. No
_________________________________________________________
router configuration become more complex
c. CPE
_________________________________________________________
Aggregation router configuration is more complex
d. _________________________________________________________
A6
Version 1.0
Appendix A
Module 3 PPPoA
Module 3 PPPoA
1. What are the two locations that terminate PPPoA sessions?
a. _____________________________________
Network access server (NAS)
Customer premise equipment (CPE)

b. _____________________________________
2. With PPPoA, which device in the network initiates the PPP session?
Choose two.
a. Subscriber host
Answer
b. Subscriber CPE
c. DSLAM
Answer
d. Aggregation router (not in PPP passive mode)

e. NSPs router
3. When using PPPoA with PTA, which two devices terminate the PPP
session?
Subscriber CPE and Aggregation router
_______________________________________________________________
4. When using PPPoA with tunneling, which two devices terminate the
PPP session?
_______________________________________________________________
Subscriber CPE and NSP router
5. Put the following events in the correct order in which they would occur
when PPPoA is used with PTA. Use numbers to indicate the correct
order.
a. The NAPs aggregation device or RADIUS server authenticates the

subscriber.
b. The subscriber CPE initiates the PPP session.
c. The NAPs aggregation device, RADIUS server, or DHCP server

allocates IP address to the CPE.
d. The user data is routed to the service destination.
Version 1.0
A7
Appendix A
when PPPoA is used with tunneling. Use numbers to indicate the
correct order.

subscribers domain name.
b. The subscriber CPE initiates the PPP session.
c. The NSPs aggregation device, RADIUS server, or DHCP server

allocates IP address to the CPE.
d. The PPP session is tunneled from NAP router to NSP router.
e. The NSPs aggregation device or RADIUS server authenticates the

subscribers domain and user names.
f. The User data is routed to the service destination.

7. List the three methods for allocating IP addresses to the subscriber
CPE.
Local pool
a. _________________________________
b. _________________________________
DHCP
RADIUS
c. _________________________________
8. Which of the following is not a characteristic of virtual access
interfaces?
a. Virtual access interfaces are cloned from parameters configured on
a virtual template interface.
Answer
b. Once created, virtual access interfaces are created permanently.

c. With PPPoA, a VC is bound to a virtual access interface.
d. With PPPoA, the virtual access interface is created when the PPP
session is initiated.
9. Which of the following are preferred ways to configure the aggregation
router for PPPoA? Choose two.
Answer
a. Using unnumbered loopback interfaces with virtual template

interfaces
Answer
b. Using multipoint ATM interfaces

c. Using point-to-point ATM interfaces
d. Configuring the username database on the router, especially when
many unique subscriber name are required
A8
Version 1.0
Appendix A
Module 3 PPPoA
10. Which of the following are true statements about PPPoA? Choose four.
Answer
a. Users can be authenticated using PAP or CHAP.
Answer
b. IP addresses can be conserved at the CPE using NAT.
Answer
c. High scaling can be achieved using RADIUS for AAA services.

d. PPPoA is limited to one user host per CPE.
Answer
e. Service providers need to maintain a database of usernames when

PPP sessions are terminated at the aggregation router.
f. Oversubscription is not possible with PPPoA.
Version 1.0
A9
Appendix A
Module 4 PPPoE
1. At what two locations are PPPoE sessions terminated?
a. _____________________________________
Network access server (NAS)
Subscriber host
b. _____________________________________
2. With PPPoE, which device in the network usually initiates the PPP
session?
Answer
a. Subscriber host
b. Subscriber CPE
c. DSLAM
d. Aggregation router
e. NSPs router
3. When using PPPoE with PTA, which two devices terminate the PPP
session?
Subscriber host and Aggregation router
_______________________________________________________________
4. When using PPPoE with tunneling, which two devices terminate the
PPP session?
_______________________________________________________________
Subscriber host and NSP router
5. List the four messages types that are exchanged between the host and
aggregation device during PPPoE discovery.
PADI
a. ______________________
PADO
b. ______________________
PADR
c. ______________________
PADS
d. ______________________
when PPP session is used with PTA after PPPoE discovery is
completed. Use numbers to indicate the correct order.
A10

subscriber.
b. The subscriber host initiates the PPP session.
c. The NAPs aggregation device, RADIUS server, or DHCP server

allocates IP address to the host.
d. The user data is routed to the service destination.
Version 1.0
Appendix A
Module 4 PPPoE
when PPP is used with tunneling after PPPoE discovery is completed.
Use numbers to indicate the correct order.

subscribers domain name.
b. The subscriber host initiates the PPP session.
c. The NSPs aggregation device, RADIUS server, or DHCP server

allocates IP address to the host.
d. The PPP session is tunneled from NAP router to NSP router.
e. The NSPs aggregation device or RADIUS server authenticates the

subscribers domain and user names.
f. The user data is routed to the service destination.

8. List the four methods that IP addresses can be allocated to the
subscriber host.
Local pool
a. _________________________________
b. _________________________________
DHCP
RADIUS
c. _________________________________
9. Which of the following is not a characteristic of virtual access
interfaces?
a. Virtual access interfaces are cloned from parameters configured on
a virtual template interface.
Answer
b. Once created, virtual access interfaces are created permanently.

c. With PPPoE, a session is bound to a virtual access interface.
d. With PPPoE, the virtual access interface is created when the PPP
session is initiated.
10. Which of the following are preferred ways to configure the aggregation
router for PPPoEoA? Choose two.
Answer
a. Using unnumbered loopback interfaces with virtual access

interfaces
Answer
b. Using multipoint ATM interfaces

c. Using point-to-point ATM interfaces
d. Configuring the username database on the router, especially when
many unique subscriber name are required
Version 1.0
A11
Appendix A
11. Which of the following are true statements about PPPoE? Choose four.
Answer
a. Users can be authenticated using PAP or CHAP.
Answer
b. Each subscriber connected to the CPE can be authenticated

individually.
Answer
c. High scaling can be achieved using RADIUS for AAA services.

d. PPPoE is limited to one user host per CPE.
Answer
e. Service providers need to maintain a database of user names when

PPP sessions are terminated at the aggregation router.
f. Oversubscription is not possible with PPPoE.
A12
Version 1.0
Appendix A
Module 5 Cisco Aggregation Optimization Features
Module 5 Cisco Aggregation Optimization Features

1. You may use PVC range with which of the following access methods?
Choose three.
Answer
a. RBE
b. RFC 1483 routing
Answer
Answer
c. PPPoA
d. PPPoEoA
e. PPPoEoE
2. Give the command syntax for creating a PVC range for the following
VCs: 1/1 through 1/127, 2/1 though 2/127, and 3/1 through 3/127.
Range pvc 1/1 3/127
_________________________________________________________________
3. How would you temporarily shut down PVC 2/55 in the range from the
previous question?
_________________________________________________________________
Use the pvc -in-range 2/55 command followed by shutdown
4. What command enables PVCs to be autoprovisioned?
Create on-demand
_________________________________________________________________
5. Using autosense of the encapsulation method permits distinguishing
between which of the following connection types?
a. PPPoA MUX and RBE SNAP
b. PPPoE MUX and RBE SNAP
c. PPPoA MUX and PPPoE MUX
d. PPPoA SNAP and PPPoE SNAP
Answer
e. PPPoA MUX and PPPoE SNAP
6. When using PPPoE profiles, users who do not get their profile from a
global
named BBA group get their profile from the ________________
group.
7. Which of the following are true with respect to using BBA groups?
Choose three.
Answer
Answer
a. BBA groups o vercome the limitations of a single VPDN group.

b. BBA groups allow use of multiple virtual templates.
c. BBA groups may be used concurrently with a VPDN group used for
PPPoE.
d. PPPoA connections get their profile from the VPDN group.
Answer
e. Session limits may be configured on the BBA group.
Version 1.0
A13
Appendix A
Module 7 AAA Services

1. Within the Cisco IOS software, AAA can be configured on which one of
the following?
a. console
b. aux console
c. tty
d. vty lines
Answer
e. all of the above
2. What new file is automatically created every 24 hours to contain

RADIUS log information?
a. clients file
b. users file
c. dictionary file
d. radius.debug
Answer
e. logfile.yymmdd
3. What are the distinct phases that a PPP link undergoes? Choose four.
Answer a. Link Establishment

Answer b. authentication
c. Link Alive
Answer
d. Network Control Protocol
Answer
e. Link Terminate
4. Which of the following statements are true? Choose four
Answer
a. RADIUS is a standards based protocol.
Answer
Answer
b. RADIUS protocol uses UDP and not TCP.
Answer
d. The RADIUS-server key must be the same at both the NAS and the
AAA server.
c. RADIUS is designed to operate in a client/server model.
e. RADIUS is supported only on a UNIX platform.
A14
Version 1.0
Appendix A
Module 7 AAA Services
5. RADIUS vendor-specific attributes (VSAs) are derived from which

IETF attribute?
a. attribute 52
b. attribute 62
Answer
c. attribute 26
d. attribute 25
e. attribute 36
Version 1.0
A15
Appendix A
Module 7 L2TP
1. True or False. L2TP allows the Layer 2 and the PPP endpoints to
reside on different networks.
Answer
a. True
b. False
2. Select all that apply to L2TP tunneling:
a. Supports only registered IP addresses
Answer
b. Separates the Layer 2 and the PPP session endpoints
Answer
c. Allows end user to appear directly connected to remote servers

d. Supports a single tunnel between LAC and LNS
3. True or False. Tunnel identifiers are at each end of the tunnel must be
identical.
a. True
Answer
b. False
4. Which of the following statements apply to the L2TP call setup process?
Choose two.
a. A call request from the user will be processed only if a tunnel
already exists.
Answer
b. Tunnel setup must be completed before an L2TP session can be

initiated.
c. The Start-Control-Connection-Reply message indicates the
completion of tunnel establishment.
d. The LAC must wait for the Incoming-Call-Reply message from the
LNS before answering the incoming call request.
Answer
e. The Incoming-Call-Connected message completes the session setup

process.
5. True or False. Sequence numbers are present on all data messages
passing through the L2TP tunnel.
a. True
Answer
A16
b. False
Version 1.0
Appendix A
Module 7 L2TP
6. True or False. Local PPP authentication requires that a local database

of usernames be setup in the router:
Answer
a. True
b. False
7. Circle the command that is NOT part of the LAC configuration when it
initiates the tunnel:
a. protocol l2tp
b. vpdn-group (group-number)
Answer
c. accept-dialin
d. request-dialin
e. domain (name)
8. A tunnel request is associated with a particular VPDN group based on
which of the following:
a. Destination IP address
b. Virtual template
c. Protocol type
Answer
d. Domain name
e. VPDN search order
9. The peer IP address pool in the LNS is used for which of the following
purposes:
a. Set the tunnel destination IP address.
Answer
b. Respond to an IPCP request from the remote CPE.

c. Assign an IP address to the Ethernet port of the LAC.
d. Set the IP address of the ATM interface.
e. None of the above.
Version 1.0
A17
Appendix A
Module 8 Cisco 10000 Series Router Hardware Overview

1. Which of the following statements is not true about the Cisco 10000
chassis?
a. The chassis has eight slots for line cards.
b. The chassis supports two PRE modules.
Answer
c. ATM line cards should be inserted in slots 1 4.

d. Half-height line cards may be used in any slot.
2. What are the two main sections of the PRE?
Route Processor
a. _____________________________________
b. _____________________________________
3. Which functions are performed by the route processor? Choose three.
a. Chassis management
b. System initialization
c. Route processor redundancy
Answer
d. Packet buffering
4. Which functions are performed by the forwarding processor? Choose
three.
Answer
a. Routing protocol updates

b. IP forwarding
c. Layer 3 features
d. QoS features
5. The _________________________
Iron Bus
is the primary data path between the
PREs and line cards.
6. The backplane bandwidth between the PRE-2 and a line card slot is
Answer
a. 3.2 Gbps
b. 1.6 Gbps
c. 800 Mbps
d. 51.2 Gbps
A18
Version 1.0
Appendix A
Module 8 Cisco 10000 Series Router Hardware Overview
7. Which of the following statements is not true about the PXF?

a. The PXF is made up of 64 CPUs.
b. The CPUs are arranged into eight rows and eight columns.
c. Each column of CPUs has its own dedicated 128 MB column
memory.
Answer
d. Each column of CPUs is internal to Toaster ASICs.

8. What are the four resultant operations of the PXF?
a. Forward
_______________________________
b. Feedback
_______________________________
c. Punt
_______________________________
(diversion)
d. Drop
_______________________________
9. Which of the following statements are true about PRE redundancy?
Choose three.
a. The active PRE is in slot 0A,
Answer
Answer
Answer
b. The standby PRE monitors the state of the active PRE

c. The standby PRE maintains of files with the active PRE
d. The standby PRE initiates cutover upon failure of the active PRE
10. Which of the following functions is common to all line cards? Choose
three.
Answer
a. OIR
b. APS
Answer
c. RPR+
Answer
d. Fail LED
Version 1.0
A19
A20
Appendix A
Version 1.0
Appendix B
Router Starting Configurations
Appendix Contents
This appendix contains examples of the starting configurations for the
routers in the student pods that are used with the lab exercises. The
examples that follow are for the routers in pod 1. The headings give the file
name of the configuration.
This appendix also includes the configurations of the core routers and PC
CPE routers for pod 1.
Version <x.x>
B1
Appendix B
P1R1 Configurations
p1r1-baseline-config
! P1R1 config
version 12.2
no parser cache
service timestamps debug uptime
service timestamps log uptime
no service password-encryption
hostname P1R1
boot system flash slot0:c3640-is-mz.122-8.T.bin
enable password lab
username P1R1 password 0 lab
ces 1/0
framer-type t1
ip subnet-zero
fax interface-type fax-mail
mta receive maximum-recipients 0
interface Ethernet0/0
ip address 52.10.100.11 255.255.0.0
half-duplex
no ip address
shutdown
half-duplex
no ip address
shutdown
half-duplex
no ip address
shutdown
half-duplex
interface ATM1/0
no ip address
no atm ilmi-keepalive
interface FastEthernet2/0
no ip address
shutdown
duplex auto
speed auto
ip classless
ip http server
ip pim bidir-enable
call rsvp-sync
mgcp profile default
dial-peer cor custom
line con 0
line aux 0
line vty 0 4
B2
Version 1.0
Appendix B
P1R1 Configurations
password lab
login
monitor
end
p1r1-rbe-config
! P1R1-rbe-config
version 12.2
no parser cache
hostname P1R1
enable password lab
ces 1/0
framer-type t1
ip subnet-zero
! fax interface-type fax-mail
ip address 52.10.100.11 255.255.0.0
half-duplex
no ip address
shutdown
half-duplex
no ip address
shutdown
half-duplex
no ip address
shutdown
half-duplex
interface ATM1/0
no ip address
interface ATM1/0.132 point-to-point
ip address 192.168.16.2 255.255.255.0
atm route-bridged ip
pvc 1/32
encapsulation aal5snap
ip address dhcp
atm route-bridged ip
pvc 2/32
no ip address
Version 1.0
B3
Appendix B
shutdown
duplex auto
speed auto
ip classless
ip http server
ip pim bidir-enable
call rsvp-sync
line con 0
line aux 0
line vty 0 4
password lab
login
end
p1r1-routing-config
! p1r1-routing-config
version 12.2
no parser cache
hostname P1R1
enable password lab
ces 1/0
framer-type t1
ip subnet-zero
ip address 52.10.100.11 255.255.0.0
half-duplex
no ip address
shutdown
half-duplex
no ip address
shutdown
half-duplex
no ip address
shutdown
half-duplex
interface ATM1/0
no ip address
B4
Version 1.0
Appendix B
P1R1 Configurations
ip address 192.168.18.2 255.255.255.0

pvc 3/32
no ip address
shutdown
duplex auto
speed auto
ip classless
ip route 0.0.0.0 0.0.0.0 ATM1/0.332
ip http server
ip pim bidir-enable
call rsvp-sync
line con 0
line aux 0
line vty 0 4
password lab
login
end
p1r1-pppoa-config
! p1r1-pppoa-config
version 12.2
no parser cache
hostname P1R1
enable password lab
ces 1/0
framer-type t1
ip subnet-zero
ip address 52.10.100.11 255.255.0.0
half-duplex
no ip address
shutdown
half-duplex
no ip address
shutdown
half-duplex
no ip address
Version 1.0
B5
Appendix B
shutdown
half-duplex
interface ATM1/0
no ip address
shutdown
pvc 4/32
encapsulation aal5mux ppp Virtual-Template1
shutdown
pvc 4/33
shutdown
pvc 4/34
shutdown
pvc 4/35
shutdown
pvc 5/32
no ip address
shutdown
duplex auto
speed auto
interface Virtual-Template1
ip address negotiated
ppp chap hostname p1user1
ppp chap password 0 lab
ip classless
ip http server
B6
Version 1.0
Appendix B
P1R1 Configurations
ip pim bidir-enable
call rsvp-sync
line con 0
line aux 0
line vty 0 4
password lab
login
end
p1r1-pppoe-config
! p1r1-pppoe-config
version 12.2
no parser cache
hostname P1R1
enable password lab
ces 1/0
framer-type t1
ip subnet-zero
vpdn enable
vpdn-group PPPoE
accept-dialin
protocol pppoe
virtual-template 6
ip address 52.10.100.11 255.255.0.0
half-duplex
no ip address
shutdown
half-duplex
no ip address
shutdown
half-duplex
no ip address
shutdown
half-duplex
interface ATM1/0
no ip address
Version 1.0
B7
Appendix B
pvc 6/32
protocol pppoe
pvc 6/33
protocol pppoe
pvc 6/34
protocol pppoe
pvc 6/35
protocol pppoe
shutdown
pvc 7/32
protocol pppoe
no ip address
shutdown
duplex auto
speed auto
ip classless
ip http server
ip pim bidir-enable
call rsvp-sync
line con 0
line aux 0
line vty 0 4
password lab
login
end
p1r1-optimization-config
! p1r1-optimization-config
version 12.2
no parser cache
B8
Version 1.0
Appendix B
P1R1 Configurations

hostname P1R1
enable password lab
ces 1/0
framer-type t1
ip subnet-zero
vpdn enable
vpdn-group PPPoE
accept-dialin
protocol pppoe
virtual-template 7
ip address 52.10.100.11 255.255.0.0
half-duplex
no ip address
shutdown
half-duplex
no ip address
shutdown
half-duplex
no ip address
shutdown
half-duplex
interface ATM1/0
no ip address
pvc 4/32
pvc 4/33
pvc 4/34
pvc 4/35
pvc 5/32
pvc 6/32
Version 1.0
B9
Appendix B
protocol pppoe
pvc 6/33
protocol pppoe
pvc 6/34
protocol pppoe
pvc 6/35
protocol pppoe
pvc 7/32
protocol pppoe
no ip address
duplex auto
speed auto
ip address dhcp
no keepalive
no keepalive
ip classless
B10
Version 1.0
Appendix B
P1R1 Configurations
ip http server
ip pim bidir-enable
call rsvp-sync
line con 0
line aux 0
line vty 0 4
password lab
login
end
p1r1-aaa-config
! p1r1-aaa-config
version 12.2
no parser cache
hostname P1R1
enable password lab
ces 1/0
framer-type t1
ip subnet-zero
vpdn enable
vpdn-group PPPoE
accept-dialin
protocol pppoe
virtual-template 7
ip address 52.10.100.11 255.255.0.0
half-duplex
no ip address
shutdown
half-duplex
no ip address
shutdown
half-duplex
no ip address
shutdown
half-duplex
interface ATM1/0
shut
no ip address
Version 1.0
B11
Appendix B
pvc 4/32
pvc 4/33
pvc 4/34
pvc 4/35
pvc 5/32
pvc 6/32
protocol pppoe
pvc 6/33
protocol pppoe
pvc 6/34
protocol pppoe
pvc 6/35
protocol pppoe
pvc 7/32
protocol pppoe
no ip address
duplex auto
speed auto
pppoe enable
ppp chap hostname p1
B12
Version 1.0
Appendix B
P1R1 Configurations

ip address dhcp
no keepalive
no keepalive
ip classless
ip http server
ip pim bidir-enable
call rsvp-sync
line con 0
line aux 0
line vty 0 4
password lab
login
end
p1r1-l2tp-config
! p1r2-l2tp-config
version 12.2
no parser cache
hostname P1R1
enable password lab
ces 1/0
framer-type t1
ip subnet-zero
no ip domain-lookup
vpdn enable
vpdn-group PPPoE
accept-dialin
protocol pppoe
Version 1.0
B13
Appendix B
virtual-template 7
ip address 52.10.100.11 255.255.0.0
half-duplex
no ip address
shutdown
half-duplex
no ip address
shutdown
half-duplex
no ip address
shutdown
half-duplex
interface ATM1/0
no ip address
pvc 4/32
pvc 4/33
pvc 4/34
pvc 4/35
pvc 5/32
pvc 6/32
protocol pppoe
pvc 6/33
protocol pppoe
pvc 6/34
protocol pppoe
pvc 6/35
protocol pppoe
B14
Version 1.0
Appendix B
P1R1 Configurations

pvc 7/32
protocol pppoe
no ip address
duplex auto
speed auto
ppp chap hostname p1user1@isp1.com
no keepalive
no keepalive
ip classless
ip http server
ip pim bidir-enable
call rsvp-sync
line con 0
line aux 0
line vty 0 4
password lab
login
monitor
end
Version 1.0
B15
Appendix B
P1R2 Configurations
! p1r2-baseline-config
version 12.2
no service pad
service timestamps debug datetime msec
service timestamps log datetime msec
hostname P1R2
boot system flash bootflash:c10k2-p11-mz.122-16.BX.bin
logging queue-limit 100
enable password lab
facility-alarm intake-temperature major 49
facility-alarm intake-temperature minor 40
facility-alarm core-temperature major 53
facility-alarm core-temperature minor 45
card 1/0 6cht3-1
card 2/0 1gigethernet-hh-1
card 2/1 8fastethernet-1
card 3/0 4oc3atm-1
card 4/0 1gigethernet-1
ip subnet-zero
no ip domain lookup
mpls ldp logging neighbor-changes
controller T3 1/0/0
controller T3 1/0/1
controller T3 1/0/2
controller T3 1/0/3
controller T3 1/0/4
controller T3 1/0/5
interface Loopback0
ip address 200.0.0.12 255.255.255.255
interface FastEthernet0/0/0
ip address 52.10.100.12 255.255.0.0
speed 100
full-duplex
interface GigabitEthernet2/0/0
ip address 172.16.0.12 255.255.255.0
no ip address
no ip address
no ip address
no ip address
no ip address
B16
Version 1.0
Appendix B
P1R2 Configurations
no ip address
no ip address
ip address 52.20.0.12 255.255.0.0
interface ATM3/0/0
no ip address
interface ATM3/0/1
no ip address
interface ATM3/0/2
no ip address
interface ATM3/0/3
no ip address
no ip address
negotiation auto
router ospf 1
log-adjacency-changes
redistribute connected
network 172.16.0.0 0.0.0.255 area 0.0.0.0
ip classless
no ip http server
line con 0
line aux 0
line vty 0 4
password lab
login
end
p1r2-optimization-config
! p1r2-optimization-config
version 12.2
no service pad
hostname P1R2
enable password lab
username p1user1 password 0 lab
Version 1.0
B17
Appendix B

card 1/0 6cht3-1
card 3/0 4oc3atm-1
ip subnet-zero
no ip domain lookup
vpdn enable
vpdn-group PPPoE
accept-dialin
protocol pppoe
virtual-template 4
pppoe limit per-vc 2
controller T3 1/0/0
controller T3 1/0/1
controller T3 1/0/2
controller T3 1/0/3
controller T3 1/0/4
controller T3 1/0/5
interface Loopback0
ip address 200.0.0.12 255.255.255.255
interface Loopback4
ip address 192.168.19.1 255.255.255.0
interface Loopback5
ip address 192.168.20.1 255.255.255.0
interface Loopback6
ip address 192.168.21.1 255.255.255.0
interface Loopback7
ip address 192.168.22.1 255.255.255.0
ip address 52.10.100.12 255.255.0.0
speed 100
full-duplex
ip address 172.16.0.12 255.255.255.0
no ip address
no ip address
no ip address
no ip address
B18
Version 1.0
Appendix B
P1R2 Configurations
no ip address
no ip address
no ip address
ip address 52.20.0.12 255.255.0.0
interface ATM3/0/0
no ip address
interface ATM3/0/0.432 multipoint
atm pppatm passive
pvc 4/32
pvc 4/33
pvc 4/34
pvc 4/35
atm pppatm passive
pvc 5/32
pvc 12/32
pvc 12/33
pvc 12/34
pvc 6/32
protocol pppoe
pvc 6/33
protocol pppoe
pvc 6/34
protocol pppoe
pvc 6/35
protocol pppoe
pvc 7/32
protocol pppoe
pvc 13/32
protocol pppoe
pvc 13/33
Version 1.0
B19
Appendix B
protocol pppoe
pvc 13/34
protocol pppoe
interface ATM3/0/1
no ip address
interface ATM3/0/2
no ip address
interface ATM3/0/3
no ip address
no ip address
negotiation auto
ip unnumbered Loopback4
peer default ip address pool PPPoAPTApool
ppp authentication chap
peer default ip address pool PPPoAPTApool2
peer default ip address pool PPPoEPTApool
ppp mtu adaptive
peer default ip address pool PPPoEPTApool2
ppp mtu adaptive
router ospf 1
network 172.16.0.0 0.0.0.255 area 0.0.0.0
ip local pool PPPoEPTApool 192.168.21.2 192.168.21.254
ip local pool PPPoEPTApool2 192.168.22.2 192.168.22.254
ip local pool PPPoAPTApool 192.168.19.2 192.168.19.254
ip local pool PPPoAPTApool2 192.168.20.2 192.168.20.254
ip classless
no ip http server
line con 0
line aux 0
line vty 0 4
password lab
login
end
B20
Version 1.0
Appendix B
P1R2 Configurations
p1r2-aaa-config
! p1r2-aaa-config
version 12.2
no service pad
hostname P1R2
enable password lab
card 1/0 6cht3-1
card 3/0 4oc3atm-1
ip subnet-zero
no ip domain lookup
vpdn enable
vpdn-group PPPoE
accept-dialin
protocol pppoe
virtual-template 4
pppoe limit per-vc 2
controller T3 1/0/0
controller T3 1/0/1
controller T3 1/0/2
controller T3 1/0/3
controller T3 1/0/4
controller T3 1/0/5
interface Loopback0
ip address 200.0.0.12 255.255.255.255
interface Loopback4
ip address 192.168.19.1 255.255.255.0
interface Loopback5
ip address 192.168.20.1 255.255.255.0
interface Loopback6
ip address 192.168.21.1 255.255.255.0
Version 1.0
B21
Appendix B
interface Loopback7
ip address 192.168.22.1 255.255.255.0
ip address 52.10.100.12 255.255.0.0
speed 100
full-duplex
ip address 172.16.0.12 255.255.255.0
no ip address
no ip address
no ip address
no ip address
no ip address
no ip address
no ip address
ip address 52.20.0.12 255.255.0.0
interface ATM3/0/0
no ip address
shut
atm pppatm passive
pvc 4/32
pvc 4/33
pvc 4/34
pvc 4/35
shut
atm pppatm passive
pvc 5/32
pvc 12/32
pvc 12/33
pvc 12/34
pvc 6/32
B22
Version 1.0
Appendix B
P1R2 Configurations
protocol pppoe
pvc 6/33
protocol pppoe
pvc 6/34
protocol pppoe
pvc 6/35
protocol pppoe
pvc 7/32
protocol pppoe
pvc 13/32
protocol pppoe
pvc 13/33
protocol pppoe
pvc 13/34
protocol pppoe
interface ATM3/0/1
no ip address
interface ATM3/0/2
no ip address
interface ATM3/0/3
no ip address
no ip address
negotiation auto
ppp mtu adaptive
ppp mtu adaptive
Version 1.0
B23
Appendix B

router ospf 1
network 172.16.0.0 0.0.0.255 area 0.0.0.0
ip classless
no ip http server
line con 0
line aux 0
line vty 0 4
password lab
login
end
p1r2-l2tp-config
! p1r2-l2tp-config
version 12.2
no service pad
hostname P1R2
enable password lab
card 1/0 6cht3-1
card 3/0 4oc3atm-1
ip subnet-zero
no ip domain lookup
controller T3 1/0/0
B24
Version 1.0
Appendix B
P1R2 Configurations
controller T3 1/0/1
controller T3 1/0/2
controller T3 1/0/3
controller T3 1/0/4
controller T3 1/0/5
bba-group pppoe global
virtual-template 3
sessions per-vc limit 2
bba-group pppoe extranet
virtual-template 4
sessions per-vc limit 5
vc-class atm pppoa
ubr 1000
encapsulation aal5autoppp Virtual-Template1
create on-demand
vc-class atm pppoe
ubr 2000
create on-demand
interface Loopback0
ip address 200.0.0.12 255.255.255.255
interface Loopback4
ip address 192.168.19.1 255.255.255.0
interface Loopback5
ip address 192.168.20.1 255.255.255.0
interface Loopback6
ip address 192.168.21.1 255.255.255.0
interface Loopback7
ip address 192.168.22.1 255.255.255.0
ip address 52.10.100.12 255.255.0.0
speed 100
full-duplex
ip address 172.16.0.12 255.255.255.0
no ip address
no ip address
no ip address
no ip address
no ip address
no ip address
no ip address
ip address 52.20.0.12 255.255.0.0
interface ATM3/0/0
Version 1.0
B25
Appendix B
no ip address
atm pppatm passive
range pvc 4/32 4/131
class-range pppoa
atm pppatm passive
pvc 5/32
range pvc 12/32 12/131
create on-demand
range pvc 6/32 6/131
class-range pppoe
pvc 7/32
protocol pppoe group extranet
range pvc 13/32 13/131
encapsulation aal5autoppp Virtual-Template2 group extranet
create on-demand
interface ATM3/0/1
no ip address
interface ATM3/0/2
no ip address
interface ATM3/0/3
no ip address
no ip address
negotiation auto
ppp mtu adaptive
ppp mtu adaptive
B26
Version 1.0
Appendix B
P1R2 Configurations

router ospf 1
network 172.16.0.0 0.0.0.255 area 0.0.0.0
ip classless
no ip http server
line con 0
line aux 0
line vty 0 4
password lab
login
end
p1r2-pxf-config
! p1r2-pxf-config
version 12.2
no service pad
hostname P1R2
enable password lab
card 1/0 6cht3-1
card 3/0 4oc3atm-1
ip subnet-zero
no ip domain lookup
controller T3 1/0/0
Version 1.0
B27
Appendix B
controller T3 1/0/1
controller T3 1/0/2
controller T3 1/0/3
controller T3 1/0/4
controller T3 1/0/5
interface Loopback0
ip address 200.0.0.12 255.255.255.255
interface Loopback5
ip address 192.168.20.1 255.255.255.0
ip address 52.10.100.12 255.255.0.0
speed 100
full-duplex
ip address 172.16.0.12 255.255.255.0
no ip address
no ip address
no ip address
no ip address
no ip address
no ip address
no ip address
ip address 52.20.0.12 255.255.0.0
interface ATM3/0/0
no ip address
atm pppatm passive
pvc 12/33
interface ATM3/0/1
no ip address
interface ATM3/0/2
no ip address
interface ATM3/0/3
no ip address
no ip address
negotiation auto
B28
Version 1.0
Appendix B
P1R2 Configurations

router ospf 1
network 172.16.0.0 0.0.0.255 area 0.0.0.0
ip classless
no ip http server
line con 0
line aux 0
line vty 0 4
password lab
login
end
Version 1.0
B29
Appendix B
P1R3 Configuration
! p1r3-baseline-config
version 12.3
no service pad
hostname P1R3
boot-start-marker
boot system flash disk0:c7400-is-mz.123-3.bin
boot-end-marker
enable password lab
no aaa new-model
ip subnet-zero
ip cef
no ip domain lookup
no voice hpi capture buffer
no voice hpi capture destination
interface Loopback0
ip address 200.0.0.13 255.255.255.0
interface GigabitEthernet0/0
ip address 172.16.1.13 255.255.255.0
duplex full
speed 1000
media-type gbic
negotiation auto
no cdp enable
ip address 200.1.1.13 255.255.255.0
duplex half
speed 1000
media-type gbic
negotiation auto
no cdp enable
ip address 52.10.100.13 255.255.0.0
duplex half
interface Group-Async0
physical-layer async
no ip address
router ospf 1
network 172.16.0.0 0.0.255.255 area 0.0.0.0
network 200.0.0.0 0.0.0.255 area 0.0.0.0
network 200.1.0.0 0.0.255.255 area 4
ip classless
no ip http server
B30
Version 1.0
Appendix B
P1R3 Configuration
no cdp advertise-v2
gatekeeper
shutdown
line con 0
transport preferred all
transport output all
line aux 0
line vty 0 4
password lab
login
transport input all
end
Version 1.0
B31
Appendix B
Core Routers Configurations

cr1-baseline-config
! CR1 config
version 12.2
no service pad
hostname CR1
boot system flash slot0:c7200-is-mz.122-15.B.bin
enable password lab
username CR1 password 0 lab
ip subnet-zero
ip cef
interface Loopback0
ip address 200.0.0.1 255.255.255.0
ip address 52.10.100.1 255.255.255.0
no ip route-cache
no ip mroute-cache
duplex half
media-type mii
ip address 172.16.0.1 255.255.255.0
negotiation auto
no cdp enable
ip address 172.16.2.1 255.255.255.0
negotiation auto
no cdp enable
ip address 172.16.1.1 255.255.255.0
negotiation auto
no cdp enable
ip address 172.16.3.1 255.255.255.0
negotiation auto
no cdp enable
router ospf 1
network 172.16.0.0 0.0.255.255 area 0.0.0.0
ip classless
no ip http server
call rsvp-sync
B32
Version 1.0
Appendix B
gatekeeper
shutdown
line con 0
stopbits 1
line aux 0
stopbits 1
line vty 0 4
password lab
login
end
cr2-baseline-config
! CR2 config
version 12.2
no service pad
hostname CR2
boot system flash slot0:c7200-is-mz.122-15.B.bin
enable password lab
ip subnet-zero
ip cef
!
interface Loopback0
ip address 200.0.0.2 255.255.255.0
ip address 52.10.100.2 255.255.255.0
no ip route-cache
no ip mroute-cache
duplex half
media-type mii
ip address 172.16.4.2 255.255.255.0
negotiation auto
no cdp enable
ip address 172.16.6.2 255.255.255.0
negotiation auto
no cdp enable
ip address 172.16.5.2 255.255.255.0
negotiation auto
no cdp enable
ip address 172.16.7.2 255.255.255.0
negotiation auto
Version 1.0
B33
Appendix B
no cdp enable
router ospf 1
network 172.16.0.0 0.0.255.255 area 0.0.0.0
ip classless
no ip http server
call rsvp-sync
gatekeeper
shutdown
!
line con 0
stopbits 1
line aux 0
stopbits 1
line vty 0 4
password lab
login
end
cr3-baseline-config
!.CR3 config
version 12.2
no service pad
hostname CR3
boot system flash disk0:c7200-is-mz.122-15.B.bin
enable password lab
ip subnet-zero
ip cef
no ip domain lookup
ip address 52.10.100.3 255.255.0.0
no ip route-cache
no ip mroute-cache
duplex half
ip address 200.1.1.3 255.255.255.0
negotiation auto
no cdp enable
ip address 200.1.2.3 255.255.255.0
negotiation auto
no cdp enable
B34
Version 1.0
Appendix B
ip address 200.2.1.3 255.255.255.0
negotiation auto
no cdp enable
ip address 200.2.2.3 255.255.255.0
negotiation auto
no cdp enable
ip address 52.30.0.3 255.255.255.0
duplex auto
speed auto
no ip address
shutdown
duplex auto
speed auto
ip address 200.1.3.3 255.255.255.0
duplex auto
speed auto
ip address 200.2.3.3 255.255.255.0
duplex auto
speed auto
router ospf 1
network 52.30.0.0 0.0.255.255 area 4
network 200.0.0.0 0.0.0.255 area 4
network 200.1.0.0 0.0.255.255 area 4
network 200.2.0.0 0.0.255.255 area 4
ip classless
no ip http server
call rsvp-sync
gatekeeper
shutdown
line con 0
stopbits 1
line aux 0
stopbits 1
line vty 0 4
password lab
login
end
Version 1.0
B35
Appendix B
PC CPE Configurations
p1cpe1-baseline-config
! p1cpe1-config
version 12.2
no service pad
hostname P1CPE1
enable password lab
ip subnet-zero
no ip routing
no ip domain lookup
no ip dhcp conflict logging
interface Ethernet0
no ip address
no ip route-cache
no ip mroute-cache
bridge-group 1
hold-queue 100 out
interface ATM0
no ip address
no ip route-cache
no ip mroute-cache
pvc 1/32
bundle-enable
dsl operating-mode auto
bridge-group 1
hold-queue 224 in
ip classless
no ip http server
bridge 1 protocol ieee
banner motd ^CThis CPE is configured for RBE^C
line con 0
stopbits 1
line vty 0 4
password lab
login
scheduler max-task-time 5000
end
B36
Version 1.0
Appendix B
PC CPE Configurations
!p1cpe2-config
version 12.2
no service pad
hostname P1CPE2
logging queue -limit 100
enable password lab
ip subnet-zero
ip dhcp excluded-address 10.0.0.1
ip dhcp pool PPPoA
network 10.0.0.0 255.255.255.0
interface Ethernet0
ip address 10.0.0.1 255.255.255.0
ip nat inside
hold-queue 100 out
interface ATM0
no ip address
ip nat outside
no atm ilmi -keepalive
pvc 1/32
encapsulation aal5mux ppp dialer
dialer pool-member 1
hold-queue 224 in
interface Dialer1
ip nat outside
encapsulation ppp
dialer pool 1
no cdp enable
ip nat inside source list 1 interface Dialer1 overload
ip classless
ip route 0.0.0.0 0.0.0.0 Dialer1
no ip http server
access-list 1 permit 10.0.0.0 0.0.0.255
banner motd ^C This CPE is configured for PPPoA ^C
line con 0
stopbits 1
line vty 0 4
password lab
login
Version 1.0
B37
Appendix B

end
!p1cpe3-config
version 12.2
no service pad
hostname P1CPE3
enable password lab
ip subnet-zero
no ip routing
interface Ethernet0
no ip address
no ip route-cache
no ip mroute-cache
bridge-group 1
hold-queue 100 out
interface ATM0
no ip address
no ip route-cache
no ip mroute-cache
pvc 1/32
bundle-enable
bridge-group 1
hold-queue 224 in
ip classless
no ip http server
bridge 1 protocol ieee
banner motd ^C This router is configured for PPPoE ^C
line con 0
stopbits 1
line vty 0 4
password lab
login
end
B38
Version 1.0

StudentGuide Implementing Broadband Aggregation On Cisco10k Vol2

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

StudentGuide Implementing Broadband Aggregation On Cisco10k Vol2

Transféré par

Droits d'auteur :

Formats disponibles

BBAGG

Book Title, Revision xx.x: Student Guide

Cisco System Engineers (SEs)

System Integrators (SIs)

Interconnecting Cisco Network Devices (ICND) or equivalent

Campus ATM (CATM) or equivalent experience

Basic DSL End To End Architecture either video on demand or

2003 Cisco Systems, Inc.

Comprehensive network support is available from Cisco Systems

Implementing Broadband Aggregation

2003 Cisco Systems, Inc.

Implementing Broadband Aggregation

Course Introduction and Objectives

Compare and contrast the various broadband aggregation

Explain how PPPoA and PPPoE work, along with descriptions of

2003 Cisco Systems, Inc.

Explain and configure various methods for optimizing subscriber

Explain AAA services available on Cisco routers and RADIUS

Explain how L2TP works, describe its typical architecture and

Identify and describe system modules and services on the Cisco

Implementing Broadband Aggregation

Course Introduction and Objectives........................................................................ ix

Module 1 Broadband Aggregation Architectures ..........................................11

Module 2 RBE and RFC 1483 Routing...............................................................21

2003 Cisco Systems, Inc.

How Does RFC 1483 Routing Work? ........................................................................ 226

Module 3 PPPoA .....................................................................................................31

Module 5 Cisco Aggregation Optimization Features ....................................51

Implementing Broadband Aggregation

Optimization Features Introduction .......................................................................... 52

Module 6 AAA Services.........................................................................................61

Module 7 L2TP .........................................................................................................71

2003 Cisco Systems, Inc.

L2TP Configuration Overview ................................................................................. 724

Module 8 Cisco 10000 Series Router Hardware Overview ..........................81

Implementing Broadband Aggregation

Module 9 Cisco 10000 Series Router Software Overview............................91

Appendix A Review Question Answers........................................................... A1

2003 Cisco Systems, Inc.

Appendix B Router Starting Configurations ..................................................B1

Implementing Broadband Aggregation

Describe how the Cisco 10000 router is used in typical broadband

Describe the Cisco 10000 router chassis components

Identify Cisco 10000 router functional components, interconnections,

Describe PRE-2 architecture and operation, including the route

Trace the flow of a packet through the PRE-2

Describe the Cisco 10000 router high-availability hardware and

2003 Cisco Systems, Inc.

Cisco 10000 Series Router Hardware Overview

Cisco 10000 Series Router Introduction

Leased-line edge aggregation

This training module focuses on the broadband aggregation utilization of

Implementing Broadband Aggregation

Cisco 10000 Series Router Introduction

Cisco 10000 Series Router Introduction

Edge service aggregation support

2003 Cisco Systems, Inc.

Cisco 10000 Series Router Hardware Overview

Broadband Aggregation Deployment Scenarios

Subscriber Connection Termination

Route Bridge Encapsulation (RBE)

PPP over ATM (PPPoA)

PPP over Ethernet (PPPoE), including PPPoEoA, PPPoEoE, and