TGN (Traffic GeNerator)

User Manual

Pagent Release 4.8

Cisco Company Confidential

Document Revision 1.0

Last Modified: 9/18/08
 CONTENTS

Chapter 1 Getting Started 1

Defining and Sending TGN Packets 1
Defining and Sending TGN Packet Flows 3
Defining and Sending TGN Packet Sequences 4
Defining and Sending TGN Packets on a Mixed Interface 5
Defining and Sending TGN Packets on a Subinterface 5
Creating Packets for SRE 5
Additional TGN Commands 5
Using the Router Console or vty Port 7
TGN Command Prompt Modes 7
Using Flow Mode 8
Using SRE Mode 9
Using TCL Scripts 9
IOS File System 10
TFTP 10
RCP 10
Flash 11
FTP 11

Chapter 2 Command Reference 1

<1-4294967295> Selecting a Traffic Stream by Number 1
<interface_name> Selecting an Interface 1
L2-.... Updating the Datalink Header Definition 1
IOS-Dependent Datalink Header Update Commands 2
HEX Datalink Header Update Commands 2
Ethernet ARPA Encap Datalink Header Field Update Commands 2
Ethernet SNAP Encap Datalink Header Field Update Commands 2
Ethernet SAP Encap Datalink Header Field Update Commands 2
Ethernet Novell-Ether Encap Datalink Header Field Update Commands 2
Token Ring SNAP Encap Datalink Header Field Update Commands 3
Token Ring SAP Encap Datalink Header Field Update Commands 3
FDDI SNAP Encap Datalink Header Field Update Commands 3
FDDI SAP Encap Datalink Header Field Update Commands 3
Serial HDLC Datalink Header Field Update Commands 3
L3-.... Updating the Network Header Definition 4
IP Network Header Field Update Commands 4
ARP Network Header Field Update Commands 4
IPX Network Header Field Update Commands 5
AppleTalk Phase 2 Network Header Field Update Commands 5
AppleTalk Phase 1 Network Header Field Update Commands 5
AARP (AppleTalk ARP) Network Header Field Update Commands 6
CLNS Network Header Field Update Commands 6
DECnet Network Header Field Update Commands 6

Cisco Company Confidential Contents 3

 L4-.... Updating the Transport Header Definition 7
TCP Transport Header Field Update Commands 7
UDP Transport Header Field Update Commands 7
ICMP Transport Header Field Update Commands 7
IGMP Transport Header Field Update Commands 9
layer Replacing the Template for a Specific Layer 9
IPv6 Layer Header Field Update Commands 10
ICMPv6 Layer Header Field Update Commands 10
add Adding a Traffic Stream 10
all Updating Multiple Traffic Streams 13
bit-rate - Setting the transmission Rate in bits per second 15
broadcast Broadcast Mode 15
Effect of Broadcast Mode and all Commands on Flow Traffic Streams 15
burst Sending Traffic Stream in Bursts 16
clear Clearing Configurations or Counters 16
close-logfile Closing an Open IFS Log File 17
data Setting Data in a Data Array 17
data-length Setting the Data Array Length 17
datalink Specifying the Datalink Header Encapsulation 18
delayed-start Delaying Start of Packet Generation 18
delete Deleting One or Several Traffic Streams 19
field Adding and Updating Configurable Fields 20
end Exiting the TGN Command Prompt 20
expand Expanding a Traffic Stream into Multiple Copies 20
field Adding and Updating Configurable Fields 21
fill-pattern Defining Data Pattern to Fill Packet 24
filter PKTS-FILTER Command Prompt Mode 25
flow Adding and Updating Packet Flows 25
fragmentation Configuring IP Fragmentation 26
insert-at Inserting a Traffic Stream 27
interval Setting the Interval Between Sending Packets 27
isl-crc-added Adding CRC to ISL Packets 27
layer Replacing the Template for a Specific Layer 28
length Setting Packet Length 28
load-config Loading a Configuration from IFS 29
max-bit-rate Setting Interface Bandwidth Control 30

4 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 mixed-interface Defining Traffic Streams on a Mixed Interface 30
name Assigning a Name to a Traffic Stream 30
on/off Activating or Deactivating a Traffic Stream 31
open-logfile Opening an IFS Log File 31
ordered-traffic Setting Ordered-Traffic Scheduling 32
output-mode Setting the Output Mode 32
prompt Setting Command Prompt Format 34
pkts PKTS Command Prompt 34
rate Setting the Packet Send Rate 35
repeat Resending Packets Repeatedly 35
replace - Selectively replacing IP Address and TCP/UDP Port Number 36
save-config Saving a Configuration to IFS 36
secondary Selecting a SECONDARY Processor for Transmission 37
select Selecting a Traffic Stream by Name 37
send Sending Packets 38
sequence Adding and Updating Packet Sequences 38
show Displaying Traffic Stream and Summary Information 38
show Displaying a Traffic Stream or Flow Member 40
show aarp Displaying AARP Header Information 41
show aarp-responder Displaying AARP Responders 41
show all Displaying Summary of Traffic Streams or Flow Members 42
show appletalk Displaying AppleTalk Header Information 42
show arp Displaying ARP Header Information 43
show arp-responder Displaying ARP Responders 43
show burst Displaying Burst Configurations 44
show clns Displaying CLNS Header Information 44
show clns-hello-generator Displaying CLNS Hello-generators 45
show config Displaying Traffic Stream Configuration Commands 45
show debug Displaying Debugging Information for Program Developers 46
show decnet Displaying DECnet Header Information 46
show decnet-hello Displaying DECnet Hello-Generators 46
show flow Displaying Summary of Packet Flows 47
show global Displaying Global Parameters 47
show fragments-sent Displaying Number of Fragments Sent 48
show icmp Displaying ICMP Header Information 48
show igmp Displaying IGMP Header Information 48
show interface Displaying Interface Status 49
show interface config Displaying Interface Configurations 49
show interface max-bit-rate Displaying Maximum Bit Rates of Interfaces 50
show interface tcl-output Displaying Interface Info in TCL-Friendly Format 50
show ip Displaying IP Header Information 50
show ipx Displaying IPX Header Information 51
show mac Displaying MAC Addresses 51

Cisco Company Confidential Contents 5

 show name Displaying Traffic Stream Names and Delayed-Start Information 51
show output-mode Displaying Output Mode Information 52
show packet Displaying Packet Sent by Traffic Stream 52
show packet fragments Displaying IP Packet Fragments 53
show pagent-format Displaying a Packet in Pagent Format 54
show program-status Displaying Current Program Status 54
show rate Displaying Traffic Stream Rates 55
show secondary Displaying Activity Status of SECONDARY Processors 56
show send Displaying Summary of Send Process 56
show sequence Displaying Summary of Packet Sequences 57
show tcp Displaying TCP Header Information 58
show traffic-stream Displaying a Traffic Stream by Name or Number 58
show udp Displaying UDP Header Information 59
sre Defining Traffic Streams for TGN or SRE 59
start/stop Starting and Stopping Traffic Generation 59
variability Defining the Variability in Packet Intervals 60
verbose - Configuring for Activity Messages 60
wait-to-release Sierra Wait-to-Release Paktype 61
write Writing Information to an IFS Log File 61

Chapter 3 Defining Header Field Values 1

Decimal and Hex Fields 1
MAC Address Fields 2
IP Address Fields 2
IPv6 Address Fields 3
Nested Increments 3
Automatic Setting of Length and Checksum Fields 4
Token Ring RIF 4
CLNS Area Fields 5

Chapter 4 ARP-Responder and Hello-Generator Commands 1

IP ARP Responder 1
AppleTalk ARP Responder 2
VINES Hello-Generator 2
CLNS Hello-Generator 2
DECnet Hello-Generator 3

Chapter 5 Using TGN and PKTS Timestamps to Measure Latency 1

Creating a TGN Traffic Stream with a Timestamp 1
Creating a PKTS Selective Filter with a Timestamp 1
Displaying Timestamp Information 1

6 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Example of Using Timestamps 2

Appendix A Examples of Traffic Streams 1

Datalink Traffic Streams 1
Example of Unknown Datalink Header with IP and TCP Headers 1
Example of Ethernet with ARPA Encapsulation Traffic Stream 2
Example of Ethernet with SNAP Encapsulation Traffic Stream 2
Example of Ethernet with SAP Encapsulation Traffic Stream 3
Example of Ethernet with Novell-Ether Encapsulation Traffic Stream 3
Example of Token Ring with SNAP Encapsulation Traffic Stream 4
Example of Token Ring with SAP Encapsulation Traffic Stream 4
Example of FDDI with SNAP Encapsulation Traffic Stream 5
Example of FDDI with SAP Encapsulation Traffic Stream 5
Example of Serial HDLC Traffic Streams 6
Network Protocol Traffic Stream 6
Example of IP Traffic Stream 6
Example of IPX Traffic Stream 7
Example of AppleTalk Phase 2 Traffic Stream 7
Example of AppleTalk Phase 1 Traffic Stream 8
Example of VINES Traffic Stream 9
Example of CLNS Traffic Stream 9
Example of DECnet Traffic Stream 10
Example of XNS Traffic Stream 11
Example of ARP (IP) Traffic Stream 11
Example of AARP (AppleTalk ARP) Traffic Stream 12
IP Transport Protocol Traffic Streams 13
Example of TCP Traffic Stream 13
Example of UDP Traffic Stream 14
Example of ICMP Traffic Stream 15
Example of IGMP Traffic Stream 16
IPv6 Traffic Streams 17
Example of IPv6 Header with Routing Header Extension 17
Example of TCP over IPv6 Traffic Stream 18
Example of UDP over IPv6 Traffic Stream 19
Example of ICMPv6 Echo Request Message Traffic Stream 20
ARP Responder and Hello-Generator Traffic Streams 21
Example of IP ARP Responder 21
Example of AARP (AppleTalk ARP) Responder 21
Example of VINES Hello-Generator 21
Example of CLNS Hello-Generator 21
Example of DECnet Hello-Generator 21

Appendix B Explanation of Fields in Headers 1

Explanation of IP Header Fields 1
Explanation of TCP Header Fields 3
Explanation of UDP Header Fields 4
Explanation of ICMP Header Fields 4

Cisco Company Confidential Contents 7

 Explanation of IGMP Header Fields 5
Explanation of IPX Header Fields 6
Explanation of XNS Header Fields 7
Explanation of Vines Header Fields 9
Explanation of AppleTalk Header Fields 10
Explanation of DECnet Header Fields 11
Explanation of CLNS Header Fields 12
Explanation of IP ARP and Appletalk ARP Header Fields 14

8 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 CHAP TER 1

Getting Started
The Pagent Traffic GeNerator (TGN) is an IOS-based program in the Pagent test tool set. It defines
and sends packets on any combination of supported interfaces on a router.
TGN has predefined templates for specific packet types. Packet definitions can also be imported
from the PKTS program capture buffer.
TGN also provides IP ARP and AppleTalk ARP responders, in addition to VINES, DECnet, and
CLNS end-node hello generators, so that the routers under test can forward packets.

Defining and Sending TGN Packets

The following list is an overview of what is involved in using TGN to define and send packets. For
more information on a particular procedure, refer to the references provided.
1 Select and load the appropriate Pagent image onto your router.

2 Select the interface you want to send packets on. Make sure the interface is not shut. Use show
interface to see the current status of the interfaces.
<interface_name> Selecting an Interface (page 2-1)
show interface Displaying Interface Status (page 2-49)
3 Create traffic streams using the add or insert-at commands. Each traffic stream can be thought
of as a separate process to create specific types of packets and send the packets in a specific way.
You can add as many traffic streams as the router memory allows. You can create traffic streams
for any number of supported interfaces.
add Adding a Traffic Stream (page 2-10)
insert-at Inserting a Traffic Stream (page 2-27)
You can create traffic streams based on packets in the PKTS program capture buffer. Both add
and insert-at support this.
You can also create traffic streams by cloning existing ones, using the add or insert-at
commands.
4 Use the TGN command prompt to get the current status of the program.

TGN Command Prompt Modes (page 1-7)

5 Select a traffic stream for update by entering its number.

<1-4294967295> Selecting a Traffic Stream by Number (page 2-1)

6 Update information in the traffic stream header fields. Decimal, hex, mac address, and ip address
fields are supported. With a few exceptions, almost all header fields can be set to be constant,

Cisco Company Confidential Getting Started 1-1

Defining and Sending TGN Packets

incrementing, or random. The L2 commands update datalink header fields. The L3 commands
update network header fields. The L4 commands update transport header fields.
Decimal and Hex Fields (page 3-1)
MAC Address Fields (page 3-2)
IP Address Fields (page 3-2)
L2-.... Updating the Datalink Header Definition (page 2-1)
L3-.... Updating the Network Header Definition (page 2-4)
L4-.... Updating the Transport Header Definition (page 2-7)
7 You can nest incrementing header fields, which means that one field increments only when the
incrementing field it links to has gone through its entire range. This ensures that all combinations
of the incrementing field values are generated.
Nested Increments (page 3-3)
8 You can set header length fields and IP checksum fields automatically to their correct values.
Automatic Setting of Length and Checksum Fields (page 3-4)
9 You can define in hex the data that comes after the headers.

data Setting Data in a Data Array (page 2-17)

data-length Setting the Data Array Length (page 2-17)
10 Complete a packet to its requested length with a fill pattern.

fill-pattern Defining Data Pattern to Fill Packet (page 2-24)

11 Use the length command to define the traffic stream packet lengths, which can be set
automatically, or they can be set to constant, incrementing, or random.
length Setting Packet Length (page 2-28)
12 You can define the traffic stream send rate in packets per second using the rate command, or in
bits per second using bit-rate command, or in milliseconds between packets using the interval
command.
rate Setting the Packet Send Rate (page 2-35)
bit-rate - Setting the transmission Rate in bits per second (page 2-15)
interval Setting the Interval Between Sending Packets (page 2-27)
13 Select one of the following output-modes: process (slowest), fast (default), dedicated (fastest),
or optimal (this hardware-specific superfast mode is available on some processors, but has
limitations on packet definitions).
ordered-traffic Setting Ordered-Traffic Scheduling (page 2-32)
14 When a traffic stream is started, the first packet transmission is delayed by a random value within
the packet transmit interval. If there are multiple traffic streams, this prevents all packets from
being sent out in bursts. You can use the delay-start command to define in milliseconds,
microseconds, or nanoseconds how long the traffic stream waits after the start command before
sending its first packet.
datalink Specifying the Datalink Header Encapsulation (page 2-18)
15 If you are creating an ISL packet, you need to use the isl-crc-added command to complete the
packet with a CRC over the encapsulated packet.
isl-crc-added Adding CRC to ISL Packets (page 2-27)
16 To temporarily turn off a traffic stream, use the off command.

on/off Activating or Deactivating a Traffic Stream (page 2-31)

1-2 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Defining and Sending TGN Packet Flows

17 Use the start and stop commands to start and stop traffic generation. The s command toggles
between the two states. The start send command causes traffic streams with defined send
amounts to send the specified number of packets.
start/stop Starting and Stopping Traffic Generation (page 2-59)
send Sending Packets (page 2-38)
18 Some IOS hardware is implemented with a PRIMARY processor and multiple SECONDARY
processors, for example the RSP (PRIMARY processor) with VIPs (SECONDARY processors).
On this hardware, either the PRIMARY processor or the SECONDARY processor (this is the
default) can transmit the packets.
secondary Selecting a SECONDARY Processor for Transmission (page 2-37)
show secondary Displaying Activity Status of SECONDARY Processors (page 2-56)
19 There are many summary commands that give an overview of traffic stream configurations. Use
the show command to review the current traffic stream configuration. The show packet
command displays what a packet looks like, as defined by a traffic stream configuration. The
show pagent-format displays a packet in a format that can be input to the classic Pagent
program.
show Displaying Traffic Stream and Summary Information (page 2-38)
show packet Displaying Packet Sent by Traffic Stream (page 2-52)
show pagent-format Displaying a Packet in Pagent Format (page 2-54)
20 Use the all option to update all traffic steams on an interface or a subset of them.

all Updating Multiple Traffic Streams (page 2-13)

21 Use the broadcast mode to update and review all traffic streams on all interfaces with a single
command.
bit-rate - Setting the transmission Rate in bits per second (page 2-15)

Defining and Sending TGN Packet Flows

TGN packet flows are special traffic streams comprised of packets that must be transmitted in the
order configured. Each flow has members. The following summarizes the steps involved in defining
and sending packet flows. For more information on a particular procedure, refer to the references
provided.
1 Use the add flow command to add a traffic stream containing a flow of packets.

add Adding a Traffic Stream (page 2-10)

2 Use the flow command to add and configure flow members.

flow Adding and Updating Packet Flows (page 2-25)

You can use flow mode or flow commands to add and configure flow members. The flow mode
prompt and flow commands are available only when the currently selected traffic stream is a flow
traffic stream. To get into flow mode, select a flow traffic stream and enter the flow command.
<1-4294967295> Selecting a Traffic Stream by Number (page 2-1)
Using Flow Mode (page 1-8)
3 Use the start and start send commands to start traffic generation.
Use the stop and s commands to stop or toggle the traffic generation.
start/stop Starting and Stopping Traffic Generation (page 2-59)

Cisco Company Confidential Getting Started 1-3

Defining and Sending TGN Packet Sequences

4 In flow mode, use the show command to display individual flow members or a summary of flow
packets.
show Displaying Traffic Stream and Summary Information (page 2-38)
In TGN mode, use the show flow to display a summary of packet flows.
show flow Displaying Summary of Packet Flows (page 2-47)

Defining and Sending TGN Packet Sequences

Note Packet flows are designed to replace TGN packet sequences, so use TGN packet flows instead
of packet sequences. (Packet sequences will not be developed further and are being maintained for
backward compatibility only.) See Defining and Sending TGN Packet Flows, (page 1-3).

There are a number of advantages to packet flows over sequences. Packet flows offer an unequal
intermember interval (which can be random), and you can specify a delayed start for the flow. You
can configure and view flow members separately as a group, since each flow maintains it own list
of members. Sequence items are part of a traffic stream list.

Packet sequences are special traffic streams comprised of packets that must be transmitted in the
order configured. The following summarizes the steps involved in defining and sending packet
sequences. For more information on a particular procedure, refer to the references provided.
1 Define a regular traffic stream that contains definitions of packets.

Defining and Sending TGN Packets (page 1-1)

2 Use the add sequence command to add a traffic stream containing a sequence of packets.

add Adding a Traffic Stream (page 2-10)

3 Use the sequence add command to build a list of references to traffic streams with packet
definitions. Use the sequence insert-at command to add a reference into an existing sequence.
With the sequence command, you can modify the sequence list by removing or disabling specific
references. Use the sequence interval command to specify the interval between consecutive
packets in the sequence.
sequence Adding and Updating Packet Sequences (page 2-38)
4 Use the start and start send commands to start traffic generation. If a sequence list references a
traffic stream, it is deactivated and cannot be used by another sequence list. Active traffic streams
not used by a sequence list are scheduled and transmitted as usual.
The scheduling information (for example, rate and send) of the first packet in the sequence is
applied to the entire sequence. Use the stop and s commands to stop or toggle the traffic
generation.
start/stop Starting and Stopping Traffic Generation (page 2-59)
5 Use the show sequence command to display a summary of packet sequences.

show sequence Displaying Summary of Packet Sequences (page 2-57)

1-4 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Defining and Sending TGN Packets on a Mixed Interface

Defining and Sending TGN Packets on a Mixed Interface

In TGN mixed interface mode, traffic streams across different interfaces are organized in a single
list. Currently, this mode only supports process and fast-send output mode. The following
summarizes the steps involved in defining and sending packets on a mixed interface. For more
information on a particular procedure, refer to the references provided.
1 Turn on mixed interface mode.

mixed-interface Defining Traffic Streams on a Mixed Interface (page 2-30)

2 Select the interface you want to send packets on.

3 Define and send regular traffic streams that contain definitions of packets.

Defining and Sending TGN Packets (page 1-1)

Traffic streams defined in mixed interface mode can only be sent using mixed interface mode. The
TGN command prompt indicates whether mixed interface mode is on by adding an X to the
application name (TGN-X).
Mixed interface mode is particularly useful when scheduling traffic streams that are interdependent.
The default scheduling for mixed interface is ordered traffic (see ordered-traffic Setting
Ordered-Traffic Scheduling (page 2-32)).

Defining and Sending TGN Packets on a Subinterface

By default, the datalink header is assembled with user-defined header field values. For a subset of
built-in packet templates (currently only IP-based protocol header templates), you can use the
datalink command to automatically assemble the datalink header according to the interface or
subinterface configuration on the router.
1 Define regular traffic streams that contain definitions of packets.

Defining and Sending TGN Packets (page 1-1)

2 Set the datalink header, IOS-dependent encapsulation with the subinterface name.

datalink Specifying the Datalink Header Encapsulation (page 2-18)

Creating Packets for SRE

TGN allows you to create packets that can be used by SRE (Stimulus Response Engine). To define
packets for SRE use, use the sre on command to select the mode that allows definition of packets
for SRE use. You must assign SRE packets a name, since that is how SRE accesses the packet
definition.
sre Defining Traffic Streams for TGN or SRE (page 2-59)
name Assigning a Name to a Traffic Stream (page 2-30)

Additional TGN Commands

The following commands provide additional flexibility and customization when using TGN.
TGN allows you to define your own configurable fields in a traffic stream. These fields are given
a name, can be placed anywhere in the packet, can define decimal, hex, or IP address data, and
the data can be constant, incrementing, or random. You can almost define your own headers.
field Adding and Updating Configurable Fields (page 2-21)

Cisco Company Confidential Getting Started 1-5

Additional TGN Commands

The timestamp is a special configurable field type that is updated just before the packet is
transmitted. You can use it in combination with the PKTS program for network latency
measurements.
field Adding and Updating Configurable Fields (page 2-21)
Using TGN and PKTS Timestamps to Measure Latency (page 5-1)
For every show command that displays information on the console, there is an equivalent write
command to write the same information to the IOS file system (IFS) log file.
write Writing Information to an IFS Log File (page 2-61)
open-logfile Opening an IFS Log File (page 2-31)
close-logfile Closing an Open IFS Log File (page 2-17)
The save command saves the current TGN configuration to an IFS file. The load command loads
a saved configuration from the IFS.
replace - Selectively replacing IP Address and TCP/UDP Port Number (page 2-36)
load-config Loading a Configuration from IFS (page 2-29)
Use the add sniffer-file command to create traffic streams by reading in a sniffer file via IFS.
add Adding a Traffic Stream (page 2-10)
Routers need end stations to forward packets to. TGN supports the definition of ARP responders
and hello generators to act like destination stations.
ARP-Responder and Hello-Generator Commands (page 4-1)
ARP Responder and Hello-Generator Traffic Streams (page A-21)
By default, traffic steams send packets continuously. Use the burst commands to send packets
in user-defined bursts.
burst Sending Traffic Stream in Bursts (page 2-16)
Use the delete command to delete one or more traffic streams.
delete Deleting One or Several Traffic Streams (page 2-19)
Use the clear all command to delete all traffic stream configurations on all interfaces.
clear Clearing Configurations or Counters (page 2-16)
Use the expand command to make multiple copies of a traffic stream, with the new copies having
constant fields and lengths.
expand Expanding a Traffic Stream into Multiple Copies (page 2-20)
Use the repeat command to send multiple copies of a packet in a fast burst.
repeat Resending Packets Repeatedly (page 2-35)
Use the variability command to maintain the basic packet rate but introduce variability in the
time intervals between packets.
variability Defining the Variability in Packet Intervals (page 2-60)
Use the ordered-traffic command to toggle between ordered traffic scheduling and the default
style of scheduling each traffic stream independently of one another. This feature is configured
on a per interface basis.
ordered-traffic Setting Ordered-Traffic Scheduling (page 2-32)
Use the verbose command to control the generation of activity messages and the verbose
logging-to command to control where the activity messages will appear.

1-6 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Using the Router Console or vty Port

verbose - Configuring for Activity Messages (page 2-60)

Using the Router Console or vty Port

TGN works through either the console port connection or a vty port. Do not use both ports
concurrently or switch between the two when either is at a Pagent program command prompt.

TGN Command Prompt Modes

When TGN mode is entered, the router command prompt changes. An example of the TGN
command line prompt is:
hostname#tgn
hostnam(TGN:OFF,Et0/0/0:none)#

Note that the hostname is shortened. The command prompt option field (in parenthesis) is used
extensively in TGN to report current status and location in the program. TGN allows 20 characters
for the option field. Since IOS limits the hostname plus option field to 27 characters, TGN limits the
hostname display to seven characters so that 20 characters are available for the option field. If the
option field exceeds 20 characters, only the last 20 are displayed because it contains the most
immediate information. The full hostname is restored when TGN is exited.
The option field shows the following (the information in the parenthesis refers to the example
above):
Program (TGN)

Note When TGN is in a special mode, such as SRE or mixed interface, a suffix is added to TGN.
For example: TGN-SRE.

Whether traffic generation is ON or OFF (OFF)

Which interface is currently selected (ethernet0/0/0)
How many traffic streams have been configured on the interface and which traffic stream is
currently selected (none)
Once traffic streams are created on the interface, the prompt changes to the following:
hostnam(TGN:ON,Et0/0/0:2/3)#

This example shows that traffic is being generated, that three traffic streams have been created on
the interface, and that traffic stream 2 of 3 is currently selected.
When traffic streams are sending their requested number of packets (start send is active), the
command prompt displays the following:
hostnam(TGN:SEND,Et0/0/0:2/3)#

When broadcast mode is selected, a command can be applied to all traffic streams on all interfaces,
instead of just a single selected traffic stream on a specific interface.
The TGN prompt in broadcast mode is:
hostnam(TGN:OFF,Broadcast)#

If a wait-to-release time is set on a 4500 or 4700, the command prompt indicates whether the wait
period is on after a stop command. For example:

Cisco Company Confidential Getting Started 1-7

TGN Command Prompt Modes

hostnam(TGN:ON,Et0/0/0:2/3)#stop
hostnam(TGN:WAIT,Et0/0/0:2/3)#
hostnam(TGN:WAIT,Et0/0/0:2/3)#
hostnam(TGN:OFF,Et0/0/0:2/3)#

The TGN prompt command changes the format of the command prompt to a static format, with a
full hostname and PAGENT in the option field. This format can be useful for test automation
scripts.
hostname(PAGENT)#

Using TCL Scripts (page 1-9)

prompt Setting Command Prompt Format (page 2-34)

Using Flow Mode

When TGN flow mode is entered, the router command prompt changes. An example of the flow
command line prompt is:
hostname(TGN:OFF,Et0:2/3)#flow
hostnam(FF,Et2/3,FLOW:NONE)#

Note that the prompt has changed. Flow mode information is appended to the normal TGN prompt.
As explained above, if the option field exceeds 20 characters, only the last 20 are displayed. It returns
to the normal TGN prompt when you exit flow mode.
In the above example, the information pertaining to flow mode starts with the characters FLOW.
It shows the following (the information in the parenthesis refers to the example above):
Flow mode (FLOW)
How many members have been configured on the interface and which member is currently
selected (NONE)
Once members are added to the flow, the prompt changes to the following:
hostnam(ON,Et0:2/3,FLOW:4/6)#

This example shows: traffic is being generated; three traffic streams have been created on the
interface Et0; traffic stream 2 of 3 is currently selected, and that it is a flow traffic stream; and six
members are in the flow, and member 4 is currently selected.
When the command start send is used, the prompt changes to:
hostnam(ND,Et0:2/3,FLOW:4/6)#

Note On some platforms, the flow prompt might look like this:

hostnam(,Et0:2/3/2,FLOW:2/3)#

This prompt does not display if the traffic generation is on or off (due to the 20 character limitation
mentioned above). In such cases, you can verify the program status with the show program-status
command.

1-8 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Using SRE Mode

Using SRE Mode

TGN also has an SRE mode in which traffic streams are created to be used as packet definitions for
the SRE program. The command prompt replaces TGN with TGN-SRE when in SRE mode. For
example:
c4700-p(TGN:OFF,Et1:2/3)#sre on
c4700-p(TGN-SRE,OFF,Et1:none)#add ip
c4700-p(TGN-SRE,OFF,Et1:1/1)#sre off
c4700-p(TGN:OFF,Et1:2/3)#

On a RSP this can look like this:

c7513a-(TGN:OFF,Et8/0:3/3)#sre on
c7513a-(N-SRE:OFF,Et8/0:none)#add ip
c7513a-(GN-SRE:OFF,Et8/0:1/1)#sre off
c7513a-(TGN:OFF,Et8/0:3/3)#

Using TCL Scripts

You can control TGN with an ATS (Automated Test System) TCL script using CSCCON (ATS
command set to control an IOS router), but you must enter the commands as if from the router exec
command prompt and not from the TGN command prompt. CSCCON cannot process TGN
command prompt options.
The examples in this manual show using TGN from the TGN command prompt, but any TGN
command can be executed from the router exec prompt by preceding the command with TGN. A
TGN command entered from the router exec can be used by TCL scripts.
The following examples both execute the following sequence of commands, but one enters the
commands at the TGN command prompt, and the other at the router exec prompt.
Selects the ethernet1 interface
Creates an IP traffic steam
Sets output rate to 10000 pps
Sets packet length to 60
Starts traffic generation
Stops traffic generation
Displays rate information
Using TGN command prompt From the router exec prompt
=========================== =========================
hostname#TGN
hostnam(TGN:OFF,Et0:none)#ethernet1 hostname#tgn ethernet1
hostnam(TGN:OFF,Et1:none)#add ip hostname#tgn add ip
hostnam(TGN:OFF,Et1:1/1)#rate 10000 hostname#tgn rate 10000
hostnam(TGN:OFF,Et1:1/1)length 60 hostname#tgn length 60
hostnam(TGN:OFF,Et1:1/1)#start hostname#tgn start
hostnam(TGN:ON,Et1:1/1)#
... send traffic for a while ...
hostnam(TGN:ON:,Et1:1/1)#stop hostname#tgn stop
hostnam(TGN:OFF,Et1:1/1)#show rate hostname#tgn show rate
hostnam(TGN:OFF,Et1:1/1)#q
hostname#

Cisco Company Confidential Getting Started 1-9

IOS File System

You can use the tgn show program-status command in a TCL script to display the information
available in the TGN command prompt options. See show program-status Displaying Current
Program Status (page 2-54).
There are also show commands that display TGN-generated data in a TCL-friendly format using the
tcl-output option. With TCL-friendly format, it is easy to extract data from the output text because
it follows a unique keyword and is not row- and column-position dependent, which can change with
Pagent releases.
The following display commands support the tcl-output option:
show Displaying a Traffic Stream or Flow Member (page 2-40)
show interface tcl-output Displaying Interface Info in TCL-Friendly Format (page 2-50)
Test automation scripts can use the prompt command to set a static command prompt format so that
scripts can enter commands as if from the NQR program command prompt instead of the IOS exec.
TGN Command Prompt Modes (page 1-7)
prompt Setting Command Prompt Format (page 2-34)

IOS File System

The TGN program has commands to save TGN configurations to the IOS File System (IFS), to load
a saved configuration from IFS, to log information to an IFS file, and to create a traffic stream based
on a sniffer file. The following sections provide information on the various file systems. See the
sections listed below for more information on the specific TGN commands that work with IFS.
replace - Selectively replacing IP Address and TCP/UDP Port Number (page 2-36)
load-config Loading a Configuration from IFS (page 2-29)
open-logfile Opening an IFS Log File (page 2-31)
close-logfile Closing an Open IFS Log File (page 2-17)
add Adding a Traffic Stream (page 2-10)

TFTP
When using TFTP for file transfers, note the following:
The TFTP session closes if there is more than 10 seconds of inactivity. This is not a problem
when saving or loading a configuration file, but it makes TFTP awkward for logging. There
cannot be more then 10 seconds between the completion of writing the information of one write
command to completing the entry of the next write command.
You must configure the Pagent router so that there is an IP network path from the router to the
TFTP server. You should be able to ping the TFTP server.
The file name can include directory names, but the directory path must be relative to the /tftpboot
directory on the TFTP server.
On most TFTP servers, the file to be written must exist and have world write permissions. Use
the UNIX touch and chmod commands to create the file and assign access permissions.
You can specify a TFTP server host name instead of an address if your Pagent router has been
configured with the appropriate ip host ... alias command.

RCP
When using RCP for file transfers, note the following:

1-10 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Flash

The TFTP session closes if there is more than 15 seconds of inactivity. This is not a problem
when saving or loading a configuration file, but it makes TFTP awkward for logging. There
cannot be more then 15 seconds between the completion of writing the information of one write
command to completing the entry of the next write command.
When RCP opens a file, it needs to know the file length. When TGN saves a configuration, the
program can determine the length of the configuration file but the program does not know what
the length of a log file will be. For this reason, when a RCP log file is opened, it prompts the user
for the file length. This prompt occurs even when a complete URL is entered, which makes RCP
logging unusable for scripts.
If the user writes out more information than the specified file length, the file will close when the
file size is reached and additional output data will be dropped. If the file is closed before the file
size is reached, the TGN program writes out spaces to complete the file to the specified length.
In the Cisco testing environment, RCP files are written relative to the users home directory.
The user must enter a user id in the URL. You can use the router configuration command:
ip rcmd remote-username myuserid

If this is not entered, the default will be the router hostname when using IFS prompts. If this
command is configured, the prompt default will be myuserid.
You must add the Pagent routers hostname to the .rhosts file in the myuserid home directory. The
router hostname must be preceded with a + and space. When TGNs command prompt is
active, it shortens hostnames to seven characters, so if the router hostname is more than seven
characters long, you need to enter both the full hostname and the shortened seven-character
hostname.
Example of an .rhosts file:
+ c7513a-pagent
+ c7513a-
+ c4700-pagent
+ c4700-p
dirt.cisco.com
yorkie.cisco.com
autons-dev-server1.cisco.com

Flash
Flash is available only if the router supports it.
Flash can keep a file open indefinitely. This is an advantage for logging, compared to TFTP and RCP,
which close after about 10 to 15 seconds of inactivity.

FTP
IFS logs onto FTP as anonymous. Within the Cisco testing environment, FTP servers are rarely
configured to accept an anonymous login. The man pages state, The anonymous account is
inherently dangerous and should be avoided when possible.
FTP has not been tested and is not recommended.

Cisco Company Confidential Getting Started 1-11

IOS File System

1-12 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 CHAP TER 2

Command Reference
This chapter lists the TGN commands and how to define and update header information.

<1-4294967295> Selecting a Traffic Stream by Number

<1-4294967295>
After you have selected an interface, you can select an existing traffic stream on the interface for
update or review by entering its number.

<interface_name> Selecting an Interface

Ethernet0
et1
Fddi1/0
Serial8/1/3
Tunnel0
You select an interface by entering its name in IOS format.
For the Tunnel interface, only PROCESS output mode is supported. All traffic streams must be
datalink ios-dependent and have the repeat equal to 1. Currently, only IP-based protocol header
templates are supported on Tunnel interfaces.

L2-.... Updating the Datalink Header Definition

L2-....
The L2 commands update the definition of the datalink header.
The L2-encapsulation command changes the LAN media encapsulation. The encap can be
ARPA, SNAP, SAP, or NOVELL-ETHER. The program only accepts encapsulations that are
valid for the media and network protocol.
L2-encapsulation LAN encapsulation

All other L2 commands are used to set the value in a header field. The commands that are
available depend on the media and the encapsulation.

Cisco Company Confidential Command Reference 2-1

L2-.... Updating the Datalink Header Definition

IOS-Dependent Datalink Header Update Commands

The IOS-dependent datalink header is created when the datalink command is available for the
protocol template of the traffic stream and the commands ios-dependent option is specified.
L2-arp-for Update IP address for datalink ARP

datalink Specifying the Datalink Header Encapsulation (page 2-18)

HEX Datalink Header Update Commands

The hex datalink header is created when the TGN program does not recognize the interface media.
L2-data Unknown datalink header hex data
L2-data-length Length of unknown datalink header

For an example, see Example of Unknown Datalink Header with IP and TCP Headers (page A-1).

Ethernet ARPA Encap Datalink Header Field Update Commands

L2-dest-addr Update Destination MAC address field
L2-protocol Update Protocol identification field
L2-src-addr Update Source MAC address field

For an example, see Example of Ethernet with ARPA Encapsulation Traffic Stream (page A-2).

Ethernet SNAP Encap Datalink Header Field Update Commands

L2-control Update Control field after DSAP and SSAP
L2-dest-addr Update Destination MAC address field
L2-ether-length Update Ethernet 802.3 length field
L2-protocol Update Protocol identification field
L2-snap-oui Update SNAP header OUI field
L2-src-addr Update Source MAC address field
L2-ssap Update SSAP address field

For an example, see Example of Ethernet with SNAP Encapsulation Traffic Stream (page A-2).

Ethernet SAP Encap Datalink Header Field Update Commands

L2-control Update Control field after DSAP and SSAP
L2-dest-addr Update Destination MAC address field
L2-dsap Update DSAP address field
L2-ether-length Update Ethernet 802.3 length field
L2-src-addr Update Source MAC address field
L2-ssap Update SSAP address field

For an example, see Example of Ethernet with SAP Encapsulation Traffic Stream (page A-3).

Ethernet Novell-Ether Encap Datalink Header Field Update Commands

L2-dest-addr Update Destination MAC address field
L2-ether-length Update Ethernet 802.3 length field
L2-src-addr Update Source MAC address field

For an example, see Example of Ethernet with Novell-Ether Encapsulation Traffic Stream
(page A-3).

2-2 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Token Ring SNAP Encap Datalink Header Field Update Commands

Token Ring SNAP Encap Datalink Header Field Update Commands

L2-access-control Update Access-control field
L2-control Update Control field after DSAP and SSAP
L2-dest-addr Update Destination MAC address field
L2-dsap Update DSAP address field
L2-frame-control Update Token ring frame-control field
L2-protocol Update Protocol identification field
L2-rif Update Token ring Routing Information Field
L2-snap-oui Update SNAP header OUI field
L2-src-addr Update Source MAC address field
L2-ssap Update SSAP address field

For an example, see Example of Token Ring with SNAP Encapsulation Traffic Stream (page A-4).

Token Ring SAP Encap Datalink Header Field Update Commands

L2-access-control Update Access-control field
L2-control Update Control field after DSAP and SSAP
L2-dest-addr Update Destination MAC address field
L2-dsap Update DSAP address field
L2-frame-control Update Token ring frame-control field
L2-rif Update Token ring Routing Information Field
L2-src-addr Update Source MAC address field
L2-ssap Update SSAP address field

For an example, see Example of Token Ring with SAP Encapsulation Traffic Stream (page A-4).

FDDI SNAP Encap Datalink Header Field Update Commands

L2-access-control Update Access-control field
L2-control Update Control field after DSAP and SSAP
L2-dest-addr Update Destination MAC address field
L2-dsap Update DSAP address field
L2-protocol Update Protocol identification field
L2-snap-oui Update SNAP header OUI field
L2-src-addr Update Source MAC address field
L2-ssap Update SSAP address field

For an example, see Example of FDDI with SNAP Encapsulation Traffic Stream (page A-5).

FDDI SAP Encap Datalink Header Field Update Commands

L2-access-control Update Access-control field
L2-control Update Control field after DSAP and SSAP
L2-dest-addr Update Destination MAC address field
L2-dsap Update DSAP address field
L2-src-addr Update Source MAC address field
L2-ssap Update SSAP address field

For an example, see Example of FDDI with SAP Encapsulation Traffic Stream (page A-5).

Serial HDLC Datalink Header Field Update Commands

L2-flags Update HDLC flags field
L2-protocol Update HDLC protocol identification field

For an example, see Example of Serial HDLC Traffic Streams (page A-6).

Cisco Company Confidential Command Reference 2-3

L3-.... Updating the Network Header Definition

L3-.... Updating the Network Header Definition

L3-....
The L3 commands update the value of fields in network header definitions. Each network
protocol has a different set of commands.

IP Network Header Field Update Commands

L3-checksum Update IP header checksum field
L3-dest-addr Update IP destination address field
L3-fragmentation Update IP fragmentation + flags field
L3-header-length Update IP version field
L3-id Update IP ID field
L3-length Update IP length field
L3-option-data IP option data
L3-option-length IP option length
L3-protocol Update IP transport protocol field
L3-src-addr Update IP source address field
L3-tos Update IP type-of-service field
L3-ttl Update IP time-to-live field
L3-version Update IP version field

The default configuration for L3-header-length is auto. If the header length is configured to a
constant value > 4, TGN adds 4-byte ip-options (not shown in the config) to the outgoing packet. If
L3-header-length is set to incrementing/random, TGN generates erroneous packets if there is an
L4-header.
show ip Displaying IP Header Information (page 2-50)
Example of IP Traffic Stream (page A-6)
Explanation of IP Header Fields (page B-1)

ARP Network Header Field Update Commands

L3-hardware Update ARP hardware field
L3-hardware-length Update ARP hardware address length field
L3-operation Update ARP operation field
L3-protocol Update ARP protocol field
L3-protocol-length Update ARP protocol address length field
L3-sender-haddr Update ARP sender mac address field
L3-sender-paddr Update ARP sender IP address field
L3-target-haddr Update ARP target mac address field
L3-target-paddr Update ARP target IP address field

show arp Displaying ARP Header Information (page 2-43)

Example of ARP (IP) Traffic Stream (page A-11)
Explanation of IP ARP and Appletalk ARP Header Fields (page B-14)

2-4 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 IPX Network Header Field Update Commands

IPX Network Header Field Update Commands

L3-checksum Update IPX checksum field
L3-dest-host Update IPX destination host mac address field
L3-dest-net Update IPX destination network field
L3-dest-socket Update IPX destination socket field
L3-length Update IPX length field
L3-packet-type Update IPX packet type field
L3-src-host Update IPX source host mac address field
L3-src-net Update IPX source network field
L3-src-socket Update IPX source socket field
L3-transport-control Update IPX transport control field

show ipx Displaying IPX Header Information (page 2-51)

Example of IPX Traffic Stream (page A-7)
Explanation of IPX Header Fields (page B-6)

AppleTalk Phase 2 Network Header Field Update Commands

L3-checksum Update APPLETALK checksum field
L3-ddp-type Update APPLETALK DDP type field
L3-dest-net Update APPLETALK destination network
L3-dest-node Update APPLETALK destination node field
L3-dest-socket Update APPLETALK destination socket field
L3-hopcount Update APPLETALK hopcount field
L3-length Update APPLETALK length field
L3-phase APPLETALK phase 1 or 2
L3-src-net Update APPLETALK source network field
L3-src-node Update APPLETALK source node field
L3-src-socket Update APPLETALK source socket field

show appletalk Displaying AppleTalk Header Information (page 2-42)

Example of AppleTalk Phase 2 Traffic Stream (page A-7)
Explanation of AppleTalk Header Fields (page B-10)

AppleTalk Phase 1 Network Header Field Update Commands

L3-checksum Update APPLETALK checksum field
L3-ddp-type Update APPLETALK DDP type field
L3-dest-net Update APPLETALK destination network
L3-dest-node Update APPLETALK destination node field
L3-dest-socket Update APPLETALK destination socket field
L3-hopcount Update APPLETALK hopcount field
L3-length Update APPLETALK length field
L3-llap-dest-node APPLETALK phase 1 LLAP destination node
L3-llap-src-node APPLETALK phase 1 LLAP source node
L3-llap-type APPLETALK phase 1 LLAP type
L3-phase APPLETALK phase 1 or 2
L3-src-net Update APPLETALK source network field
L3-src-node Update APPLETALK source node field
L3-src-socket Update APPLETALK source socket field

show appletalk Displaying AppleTalk Header Information (page 2-42)

Example of AppleTalk Phase 1 Traffic Stream (page A-8)
Explanation of AppleTalk Header Fields (page B-10)

Cisco Company Confidential Command Reference 2-5

L3-.... Updating the Network Header Definition

AARP (AppleTalk ARP) Network Header Field Update Commands

L3-hardware Update AARP hardware field
L3-hardware-length Update AARP hardware address length field
L3-operation Update AARP operation field
L3-protocol Update AARP protocol field
L3-protocol-length Update AARP protocol address length field
L3-sender-haddr Update AARP sender mac address field
L3-sender-network Update AARP sender network address field
L3-sender-node Update AARP sender node address field
L3-target-haddr Update AARP target mac address field
L3-target-network Update AARP target network address field
L3-target-node Update AARP target node address field

show aarp Displaying AARP Header Information (page 2-41)

Example of AARP (AppleTalk ARP) Traffic Stream (page A-12)
Explanation of IP ARP and Appletalk ARP Header Fields (page B-14)

CLNS Network Header Field Update Commands

L3-checksum Update CLNS checksum field
L3-dest-area Update CLNS destination area field
L3-dest-host Update CLNS destination host mac address.
L3-dest-len Update CLNS destination length field
L3-dest-protocol Update CLNS destination protocol field
L3-flags Update CLNS flags field
L3-header-length Update CLNS header-length field
L3-id Update CLNS id field
L3-lifetime Update CLNS lifetime field
L3-option-length CLNS option length
L3-segment-length Update CLNS segment length field
L3-src-area Update CLNS source area field
L3-src-host Update CLNS source host mac address.
L3-src-len Update CLNS source length field
L3-src-protocol Update CLNS source protocol field
L3-version Update CLNS version field

CLNS Area Fields (page 3-5)

show clns Displaying CLNS Header Information (page 2-44)
Example of CLNS Traffic Stream (page A-9)
Explanation of CLNS Header Fields (page B-12)

DECnet Network Header Field Update Commands

L3-dest-area Update DECNET destination area field
L3-dest-node Update DECNET destination node field
L3-flags Update DECNET flags field
L3-length Update DECNET length field
L3-next-level2 Update DECNET next level field
L3-protocol Update DECNET protocol field
L3-service Update DECNET service field
L3-src-area Update DECNET source area field
L3-src-node Update DECNET source node field
L3-visit-count Update DECNET visit count field

show decnet Displaying DECnet Header Information (page 2-46)

Example of DECnet Traffic Stream (page A-10)
Explanation of DECnet Header Fields (page B-11)

2-6 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 L4-.... Updating the Transport Header Definition

L4-.... Updating the Transport Header Definition

L4-....
The L4 commands update the value of fields in transport header definitions. Each transport
protocol has a different set of commands.

TCP Transport Header Field Update Commands

L4-acknowledge Update TCP acknowledge number field
L4-checksum Update TCP header checksum field
L4-dest-port Update TCP destination port field
L4-flags Update TCP flags field
L4-header-length Update TCP header-length field
L4-option-data TCP option data
L4-option-length TCP option length
L4-sequence Update TCP sequence number field
L4-src-port Update TCP source port field
L4-urgent Update TCP urgent pointer field
L4-window Update TCP window field

show tcp Displaying TCP Header Information (page 2-58)

Example of TCP Traffic Stream (page A-13)
Explanation of TCP Header Fields (page B-3)

UDP Transport Header Field Update Commands

L4-checksum Update UDP checksum field
L4-dest-port Update UDP destination port field
L4-length Update UDP length field
L4-src-port Update UDP source port field

show udp Displaying UDP Header Information (page 2-59)

Example of UDP Traffic Stream (page A-14)
Explanation of UDP Header Fields (page B-4)

ICMP Transport Header Field Update Commands

This is a simplified definition of the ICMP header. These L4- fields only define the first 8 bytes of
an ICMP header. The type, code, and checksum fields are the first 4 bytes of all ICMP headers. The
next 4 bytes are represented by the option field, which has a different meaning for different ICMP
types. The rest of the ICMP data has to be supplied in the data array.
L4-checksum Update ICMP checksum field
L4-code Update ICMP code field
L4-option Update ICMP option field
L4-type Update ICMP type field

show icmp Displaying ICMP Header Information (page 2-48)

Example of ICMP Traffic Stream (page A-15)
Explanation of ICMP Header Fields (page B-4)
Here are some ICMP header definitions:
Destination Unreachable Message
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type | Code | Checksum |

Cisco Company Confidential Command Reference 2-7

L4-.... Updating the Transport Header Definition

+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| unused |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Internet Header + 64 bits of Original Data Datagram |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

Time Exceeded Message

0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type | Code | Checksum |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| unused |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Internet Header + 64 bits of Original Data Datagram |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

Parameter Problem Message

0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type | Code | Checksum |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Pointer | unused |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Internet Header + 64 bits of Original Data Datagram |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

Source Quench Message

0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type | Code | Checksum |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| unused |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Internet Header + 64 bits of Original Data Datagram |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

Redirect Message
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type | Code | Checksum |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Gateway Internet Address |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Internet Header + 64 bits of Original Data Datagram |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

Echo or Echo Reply Message

0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type | Code | Checksum |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Identifier | Sequence Number |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Data ...
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

Timestamp or Timestamp Reply Message

0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

2-8 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 IGMP Transport Header Field Update Commands

| Type | Code | Checksum |

+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Identifier | Sequence Number |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Originate Timestamp |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Receive Timestamp |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Transmit Timestamp |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

Information Request or Information Reply Message

0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type | Code | Checksum |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Identifier | Sequence Number |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

ICMP Router Advertisement Message

0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type | Code | Checksum |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Num Addrs |Addr Entry Size| Lifetime |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Router Address[1] |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Preference Level[1] |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Router Address[2] |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Preference Level[2] |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

ICMP Router Solicitation Message

0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type | Code | Checksum |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Reserved |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

IGMP Transport Header Field Update Commands

L4-checksum Update IGMP checksum field
L4-group-address Update IGMP group address field
L4-type Update IGMP type field
L4-version Update IGMP version field

show igmp Displaying IGMP Header Information (page 2-48)

Example of IGMP Traffic Stream (page A-16)
Explanation of IGMP Header Fields (page B-5)

layer Replacing the Template for a Specific Layer

layer layer-number layer-template

Cisco Company Confidential Command Reference 2-9

add Adding a Traffic Stream

Replaces the template of the numbered layer with the specified layer template. The layer template
options are:
EthernetEthernet layer template
HDLCHDLC layer template
PPPPoint to Point Protocol layer template
dataHEX string layer template
icmpv6ICMPv6 layer template
ipv6IPv6 layer template
rtpReal Time Protocol layer template
layer layer-number undo
Undoes the template of the numbered layer and reinitializes the layer according to the packet
template.
layer max-level number
Sets the maximum number of layers that can be configured. The minimum is 4, and the maximum
is 255. The default is 4.

Note For more information on using header templates available through the layer command, see the
Layer Templates document on the Pagent Web page.

IPv6 Layer Header Field Update Commands

L3-version 6
L3-traffic-class 0
L3-flow-label 0x0
L3-payload-length auto
L3-next-header auto
L3-hop-limit 64
L3-src-addr ::
L3-dest-addr ::
L3-header total 0 modules

Please read RFC 2460 for an explanation of IPv6 header fields.

Example of IPv6 Header with Routing Header Extension (page A-17)

ICMPv6 Layer Header Field Update Commands

L4-type auto
L4-code 0
L4-checksum auto
L4-message is data
L4-message length 0 bytes

Please read RFC 2463 for explanation of ICMPv6 header fields.

Example of ICMPv6 Echo Request Message Traffic Stream (page A-20)

add Adding a Traffic Stream

The add command takes the following arguments. Each format is described in more detail below.
add template [timestamp]

2-10 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 add Adding a Traffic Stream

Creates a traffic stream based on a template.

add pkts-packet [pkt# [to to-pkt#]] [filter-with {active-display-filters | pkts-filter-name}] [tag]
[timestamp]
Creates a traffic stream based on a packet in a PKTS capture buffer.
add interface ts-name-or-number
Creates a traffic stream by cloning an existing traffic stream.
add mixed-interface [primary | secondary slot-number] ts-name-or-number
Creates a traffic stream by cloning an existing traffic stream from a mixed interface.
add {arp | aarp} responder
Creates an IP or AppleTalk ARP responder.
add {decnet | clns} hello-generator
Creates a DECnet or Connectionless Network Service (CLNS) hello-generator.
add sniffer-file url
Creates a traffic stream based on sniffer file.
add flow
Creates a traffic stream containing a packet flow.
add sequence
Creates a traffic stream containing a sequence of packets.
Any format of the add command can use the following options:
k4700-p(TGN:OFF,Et0:none)#add ?
Async Async interface
BVI Bridge-Group Virtual Interface
CDMA-Ix CDMA Ix interface
CTunnel CTunnel interface
Dialer Dialer interface
Ethernet IEEE 802.3
Group-Async Async Group interface
Lex Lex interface
Loopback Loopback interface
MFR Multilink Frame Relay bundle interface
Multilink Multilink-group interface
Null Null interface
Serial Serial
Tunnel Tunnel interface
Vif PGM Multicast Host interface
Virtual Virtual interface
Virtual-PPP Virtual PPP interface
Virtual-Template Virtual Template interface
Virtual-TokenRing Virtual TokenRing
aarp Appletalk ARP traffic stream or responder
appletalk Appletalk traffic stream
arp IP ARP traffic stream or responder.
cdp CDP traffic stream
clns CLNS traffic stream
datalink Datalink traffic stream.
decnet DECnet traffic stream
flow flow
hex HEX only traffic stream
icmp ICMP traffic stream
icmpv6 ICMPv6 traffic stream

Cisco Company Confidential Command Reference 2-11

add Adding a Traffic Stream

igmp IGMP traffic stream

ip IP traffic stream
ipv6 IPv6 traffic stream
ipx IPX traffic stream
mixed-interface Mixed interface
pkts-packet Get a packet from the PKTS capture buffer.
sequence sequence
sniffer-file Get packets from a sniffer file.
tcp TCP traffic stream
template compiled template
udp UDP traffic stream

add template [timestamp]

Creates a traffic stream that allows you to define a packet on an interface and define how the
packet is to be sent out the interface. The options for template are:
hexNo headers or fields and just hex data.
datalinkDatalink header only.
ip, arp, appletalk, aarp, ipx, clns, decnet Datalink and network headers.
tcp, udp, icmp, igmpDatalink, IP network, and transport headers.
ipv6, icmpv6, cdpSpecial templates to create the respective filters. They translate to the
following set of commands:
add ipv6
add ip
layer 3 ipv6

add icmpv6
add icmp
layer 4 icmpv6

add cdp
add ip
layer 3 cdp

If the timestamp keyword is included, a timestamp configurable field is added to the data array
of the packet. The TGN program writes a timestamp in this field just before transmitting the
packet. For more information on timestamps, see field type timestamp in field Adding and
Updating Configurable Fields (page 2-21).
add pkts-packet [pkt# [to to-pkt#]] [filter-with {active-display-filters | pkts-filter-name}] [tag]
[timestamp]
Creates a traffic stream based on an existing packet in the PKTS program capture buffer. pkt# is
the number of the packet in the currently active capture buffer in the PKTS program, or a range
of packets.
Use the filter-with option to select packets by applying the active display filters to the specified
range, or to apply a specific filter.
The tag option selects the tagged packets in the PKTS buffer in the specified range.
TGN picks a best match template for the packet. Any additional packet data beyond the template
headers is copied into the data array. For IP, TCP, UDP, ICMP and IGMP headers, the length and
checksum fields are set to auto in the new traffic streams.
If the packet from PKTS is for a different media than the TGN interface, you must update and
correct the datalink header.

2-12 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 all Updating Multiple Traffic Streams

When a packet is captured with subinterface (ios-dependent) capture, PKTS does not know the
L2-header for the packet. Hence, it is not copied into the traffic stream.
If the timestamp keyword is included, a timestamp configurable field is added to the data array
of the packet. The TGN program writes a timestamp in this field just before transmitting the
packet. For more information on timestamps, see field type timestamp in field Adding and
Updating Configurable Fields (page 2-21).
add interface ts-name-or-number
Creates a traffic steam by cloning an existing traffic stream. The traffic stream to be cloned is
identified by the interface it is configured on and its name or number.
If the cloned interfaces are on different media than the original, the datalink header cannot be
duplicated, and you must define the datalink fields in the cloned traffic stream (see L2-....
Updating the Datalink Header Definition (page 2-1)).
add mixed-interface [primary | secondary slot-number] ts-name-or-number
Creates a traffic stream by cloning an existing traffic stream from a mixed interface.
add {arp | aarp} responder
Creates a process that responds to ARP requests for a specific IP or AppleTalk address with an
ARP response. This allows a router under test to fill its ARP cache, so it can forward a packet
onto a local network.
add {decnet | clns} hello-generator
Creates a process that sends end-node hello packets every 30 seconds to simulate an active station
for the DECnet and CLNS protocols. A router needs to receive end-node hellos for these
protocols, so it knows which interface a station is on and the stations MAC address.
add sniffer-file url
Adds a batch of traffic streams from an .ENC sniffer file or a .pcap/.cap libpcap file or a .cap
netxray file to the currently selected interface, regardless of the media type. The file is read in
using the IOS File System (IFS). The use of url is similar to loading a configuration from IFS
(see load-config Loading a Configuration from IFS (page 2-29)).
add sequence
Creates a traffic stream containing a sequence of packets. Each packet in the sequence is defined
with a separate traffic stream. The sequence is defined with a list of references to traffic streams
with packet definitions (see sequence Adding and Updating Packet Sequences (page 2-38)).
add flow
Creates a traffic stream containing a sequence of packets. Each packet (member) in the flow is
defined with a separate traffic stream (see flow Adding and Updating Packet Flows
(page 2-25)).
clear Clearing Configurations or Counters (page 2-16)
delete Deleting One or Several Traffic Streams (page 2-19)
expand Expanding a Traffic Stream into Multiple Copies (page 2-20)
insert-at Inserting a Traffic Stream (page 2-27)

all Updating Multiple Traffic Streams

all [template] [from ts-name-or-number [to ts-name-or-number]]

Cisco Company Confidential Command Reference 2-13

all Updating Multiple Traffic Streams

Preceding a command that updates a traffic stream configuration with the keyword all causes the
command to be applied to all traffic streams.
In non-broadcast mode, the command is applied to all TGN traffic streams on the selected
interface.
In broadcast mode, all is implied, and the command is applied to all TGN traffic streams on all
interfaces (see bit-rate - Setting the transmission Rate in bits per second (page 2-15)). For how
broadcast mode works on flow traffic streams, see Effect of Broadcast Mode and all Commands
on Flow Traffic Streams (page 2-15).
In both cases, commands that update header field definitions only apply to traffic streams of the
same template as the currently selected traffic stream. For example, if the currently selected
traffic stream is a TCP template, the command updates TCP traffic streams only.
template
Limits the command to only traffic streams of a specific template. Valid templates are:
datalink
ip
arp
arp responder
tcp
udp
icmp
igmp
appletalk
aarp
aarp responder
ipx
decnet
decnet hello-generator
clns
clns hello-generator
from ts-name-or-number [to ts-name-or-number]
Limits the command to a specific range of traffic streams. You can identify the traffic streams by
name or number. If identifying by name, you must enter the full exact name.
If you are in broadcast mode, both traffic stream names must refer to traffic streams on the same
interface. The command is applied to the selected range on all interfaces.
If in non-broadcast mode, traffic stream names must be on the currently selected interface.
Examples
all length 1000
Sets all traffic streams to send packets of 1000 byte length.
all ip L3-dest-addr 100.1.1.1
Sets all traffic streams with an ip template (note that this excludes tcp, udp, icmp, and igmp
templates) to have a destination IP address of 100.1.1.1.
all from 5 to 10 rate 3000
Sets all traffic streams from number 5 to 10 to send packets at 3000 packets per second (pps).
all tcp from 10 L4-dest-port 68

2-14 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 bit-rate - Setting the transmission Rate in bits per second

Sets all traffic streams with a tcp template, starting with number 10 to the last traffic stream on
the interface, to have a TCP destination port address of 68.

bit-rate - Setting the transmission Rate in bits per second

bit-rate bits-per-second
Sets the traffic transmission rate. This command is useful when high precision is required.
ordered-traffic Setting Ordered-Traffic Scheduling (page 2-32)
rate Setting the Packet Send Rate (page 2-35)
interval Setting the Interval Between Sending Packets (page 2-27)
variability Defining the Variability in Packet Intervals (page 2-60)
show rate Displaying Traffic Stream Rates (page 2-55)

broadcast Broadcast Mode

broadcast [on | off]
The broadcast command puts the TGN program in broadcast mode. This mode is reflected in
the command prompts in that the selected interface and traffic stream number are replaced by the
word Broadcast. Broadcast mode allows all traffic streams (or a subset with the all modifier)
on all interfaces to be reviewed and updated.
In broadcast mode, the all command is implicit, since the purpose of the mode is to affect all
traffic streams, but you have to include the all keyword to use the template or traffic stream
range options (see all Updating Multiple Traffic Streams (page 2-13)).
In broadcast mode, commands that update header field definitions only apply to traffic streams
of the same template as the currently selected traffic stream. For example, if the currently
selected traffic stream is a TCP template, the command updates TCP traffic streams only.
You can use either broadcast or broadcast on to go to broadcast mode, and tgn or broadcast
off to turn broadcast mode off.

Effect of Broadcast Mode and all Commands on Flow Traffic Streams

The following occurs when you are in flow mode and using broadcast mode and the all commands:
If a flow traffic stream matches the specified criteria, the broadcast mode commands and all
commands that are applicable to flow traffic streams (delayed-start, rate, interval, off, and on
commands) update the flow traffic stream but do not affect its members.
If a flow traffic stream matches the specified criteria, the broadcast mode commands and all
commands that are not applicable to flow traffic streams but are applicable to flow members (that
is, the send command), do not affect the flow traffic stream but update all its members.
If a flow traffic stream matches the specified criteria, the broadcast mode commands and all
commands that update header field definitions update all its members.
For example, if 4 is a flow traffic stream, all the following commands update all members of flow
traffic stream 4:

Cisco Company Confidential Command Reference 2-15

burst Sending Traffic Stream in Bursts

k4700-p(TGN:OFF,Et0:2/8)#all from 2 to 6 send 8

k4700-p(TGN:OFF,Et0:2/8)#all from 2 to 6 L3-des 2.3.4.5
k4700-p(TGN:OFF,Et0:2/8)#br on
k4700-p(TGN:OFF,Broadcast)#all from 2 to 6 L3-des 2.3.4.5
k4700-p(TGN:OFF,Broadcast)#all from 2 to 6 send 8

In the following example, the off, on, and delayed-start commands (which are all commands and
broadcast mode commands) update flow traffic stream 4, but not its members. However, the repeat
command does not affect the flow or the flow members, because the command is not available for
flow traffic streams or flow members. The last command in the example updates all members of the
flow with tcp templates.
k4700-p(TGN:OFF,Et0:2/8)#all from 2 to 6 off
k4700-p(TGN:OFF,Et0:2/8)#all from 2 to 6 delayed-start 4
k4700-p(TGN:OFF,Et0:2/8)#br on
k4700-p(TGN:OFF,Broadcast)#all from 2 to 6 on
k4700-p(TGN:OFF,Broadcast)#all from 2 to 6 delayed-start 4
k4700-p(TGN:OFF,Broadcast)#all from 2 to 6 repeat 4
k4700-p(TGN:OFF,Et0:2/8)#br on
k4700-p(TGN:OFF,Et0:2/8)#all from 2 to 6 repeat 4
k4700-p(TGN:OFF,Et0:2/8)#all tcp from 2 to 6 send 20

burst Sending Traffic Stream in Bursts

burst {on | off}
burst duration on n1 [to n2]
burst duration off n1 [to n2]
TGN can send a traffic stream continuously or in bursts. With the command burst off, the traffic
stream is sent continuously. The command burst on causes the traffic stream to be sent in bursts.
The burst duration on and burst duration off commands determine how long, in milliseconds,
the burst will be on and off. If only n1 is specified, the burst is on or off for the specified amount
of time. If n1 and n2 are entered (n2 must be greater than n1), the duration is random within the
time range specified by n1 and n2.
For example:
rate 1000
burst on
burst duration on 1000 to 10000
burst duration off 5000 to 10000

This combination of commands causes the traffic stream to send packets at 1000 pps for one
second, wait for 5 to 10 seconds, send packets for one second, wait for 5 to 10 seconds, and so on.
show burst Displaying Burst Configurations (page 2-44)

clear Clearing Configurations or Counters

clear {all | config | count}
These commands clear configurations, counters, or log files. They affect all traffic streams in
both flow and non-flow mode, including all flow members.
clear all
clear config
Both of these two commands delete all TGN traffic streams configured on all interfaces.
delete Deleting One or Several Traffic Streams (page 2-19)

2-16 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 close-logfile Closing an Open IFS Log File

clear count
Traffic streams keep a count of the number of packets they have sent. This sets the send count for
all traffic streams to zero.

close-logfile Closing an Open IFS Log File

close-logfile
Closes an IFS log file that was opened with the open-logfile command.
IOS File System (page 1-10)
open-logfile Opening an IFS Log File (page 2-31)
write Writing Information to an IFS Log File (page 2-61)

data Setting Data in a Data Array

data starting-byte-offset hex-data-string
Creates and updates a traffic stream data array. The default is that a data array does not appear
when the traffic stream configuration is displayed.
starting-byte-offset
Specifies where the data being defined starts in the data array.
hex-data-string
Specifies the string of hex numbers to put into the data array.
For example, to create a data array and put data into it:
data 24 "24 11"

This example creates a data array 26 bytes long, writes the hex number 24 into location 24, and the
hex number 11 into location 25. Since this is creating the array, the locations from 0 to 23 are set to
zero.
If the example was updating an existing data array of at least 26 bytes length, only locations 24 and
25 would be updated.
If we display the traffic stream configuration, we will see the following:
!
data-length 26
! 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
data 0 "00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00"
data 20 "00 00 00 00 24 11"
!

data-length Setting the Data Array Length (page 2-17)

data-length Setting the Data Array Length

data-length length
Sets how much of the data array to use when a packet is created. length is the data array length
in bytes. If length is set to less than the data in the data array, only that reduced portion of the data
array is used. If it is set to greater than the data array, the size of the data array is increased and
padded with zeros.

Cisco Company Confidential Command Reference 2-17

datalink Specifying the Datalink Header Encapsulation

If the data command is used to add data to the data array and it increases the size of the array,
data-length is updated to the new data array size.
data Setting Data in a Data Array (page 2-17)

datalink Specifying the Datalink Header Encapsulation

datalink user-defined [hex]
datalink ios-dependent [subinterface-name]
Specifies how the datalink header is assembled. By default, it is assembled with user-defined
header field values.
The hex option configures the datalink header as a hex string, which can be configured using the
commands L2-data and L2-data-length. This allows a packet with any datalink header to be
transmitted over an interface (see Example of Unknown Datalink Header with IP and TCP
Headers (page A-1)).
For a subset of built-in packet templates (currently only IP-based protocol header templates), the
datalink command provides an additional option of assembling the datalink header through the
Cisco IOS media stack.
When a traffic stream is configured with the ios-dependent option in combination with a
subinterface name, the datalink protocol header is encapsulated according to the subinterface
configuration on the router. This makes media without built-in protocol header templates and/or
subinterfaces more compatible with the application.
This feature is not supported on distributed CPUs.
L2-.... Updating the Datalink Header Definition (page 2-1)
Defining and Sending TGN Packets on a Subinterface (page 1-5)
IOS-Dependent Datalink Header Update Commands (page 2-2)

delayed-start Delaying Start of Packet Generation

delayed-start {random | n {milliseconds | microseconds | nanoseconds}}
Defines how long a traffic stream waits after a start command before sending its first packet.
This command is per traffic stream.
delayed-start random
Sets the delay to be a random value within the traffic stream packet interval (this is the default
mode). For example, if the packet rate is 10 pps, packet interval is 100 milliseconds, and the start
delay is random from 0 to 100 milliseconds.
This assumes TGN is used as a network exerciser, and the traffic streams are not starting all at
once.
delayed-start n {milliseconds | microseconds | nanoseconds}
Allows the tester to define the delay. The delay can be defined in milliseconds, microseconds, or
nanoseconds, whichever is most convenient. For backward compatibility with previous releases,
milliseconds is assumed if the time interval is left off.
TGN uses an IOS clock that has .23 nanosecond resolution, but the real limitation here is in the
time it take the software to read the clock. Following are times between back-to-back reads of
the clock on specific platforms.
First read

2-18 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 delete Deleting One or Several Traffic Streams

platform Clock routine not in cache Clock routine in cache

-------------------------------------------------------------------
2500 29.9 microseconds 29.9 microseconds
4000 20.9 20.9
4700 6.1 2.1
7200 2.3 1.3
RSP4 2.3 1.4
VIP-40 5.2 1.9

show name Displaying Traffic Stream Names and Delayed-Start Information (page 2-51)

delete Deleting One or Several Traffic Streams

delete traffic-stream [template] [from n1 to n2]
Deletes specific traffic streams. Use the clear all command to delete all traffic streams (see clear
Clearing Configurations or Counters (page 2-16)).
Deleting a flow also deletes all members of the flow.
delete traffic-stream
Deletes the currently selected traffic stream.
delete traffic-stream template
Deletes traffic streams that match the template name. For example, delete traffic-stream ipx
deletes all IPX traffic streams on the currently selected interface if TGN is in non-broadcast
mode. It deletes all IPX traffic streams from all interfaces if in broadcast mode.
delete traffic-stream from n1 to n2
Deletes traffic streams that fall within the range of traffic stream numbers. For example, delete
traffic-stream from 5 to 10 deletes all traffic streams from number 5 to 10 on the currently
selected interface if TGN is in non-broadcast mode. It deletes all traffic streams from 5 to 10 on
all interfaces if in broadcast mode.
delete traffic-stream template from n1 to n2
Deletes traffic streams that match the template name and fall within the range of traffic stream
numbers. For example, delete traffic-stream ipx from 5 to 10 deletes all IPX traffic streams
from number 5 to 10 on the currently selected interface if TGN is in non-broadcast mode. It
deletes all IPX traffic streams from 5 to 10 on all interfaces if in broadcast mode.
This command helps send out complex traffic more quickly. Updating a packet with changing
length and data can cause length fields to be updated and checksums to be recalculated, which
reduces the maximum send rate. If the memory is available, you can create multiple traffic
streams that do not need to be updated so that they can be sent at a faster rate.
expand n
Expands the currently selected traffic stream into the specified number of traffic streams.
expand imix [packet-mix packet_length_list]
Expands the currently selected traffic stream by creating a set of IMIX traffic streams. If
packet-mix is specified, the traffic stream is expanded into the specified number of streams with
the specified lengths in packet_length_list.
If packet-mix is not specified, the traffic stream is expanded into 12 traffic streams of the
following lengths in this order:
64 64 570 64 64 570 64 1518 570 64 64 570

Cisco Company Confidential Command Reference 2-19

field Adding and Updating Configurable Fields

You can use this command with a flow member to send out the IMIX traffic in order.

Note The expand command makes it easy to create lots of traffic streams, but each traffic stream
takes up router memory. If you use up all the free memory, the router crashes. The number of traffic
streams you can safely create depends on the amount of router free memory (use show memory at
exec) and the size and complexity of the packet definitions. Leave a couple of megabytes of memory
free for router processes and stacks.

flow Adding and Updating Packet Flows (page 2-25)

field Adding and Updating Configurable Fields

field add field-name
field insert-at [field# | field-name] name
field [field# | select field-name]
field [field# | select field-name] delete
field [field# | select field-name] name field-name
field [field# | select field-name] type {ip | n {decimal | hex | bcd} bytes | timestamp | ascii}
field [field# | select field-name] start-at sign-post offset bytes
field [field# | select field-name] data {number | ip-address}
field [field# | select field-name] data [increment | random] {number | ip-address} to
{number | ip-address} [no-reset]
field [field# | select field-name] data [iterate-thru [num-values n] start-index i values CSV
[nest-over field-name] [no-reset]]
field [field# | select field-name] data ascii-string
These commands are used to create, delete, and maintain configurable fields in a traffic stream
packet definition. Configurable fields can be used to augment the field definitions supplied by
the templates. They define where a field starts in a packet, how long it is, the format of the data,
and what the data in the field is, whether constant, incrementing, or random.
When you use the [field# | select field-name] option, the specified field becomes the current field.
If you do not specify a field number or field name, the command is applied to the current field
(the field last accessed or modified).

end Exiting the TGN Command Prompt

end
quit
Either of these two commands can be used to exit from the TGN program command prompt and
return to the router exec command prompt. The TGN program continues to run; only the
command prompt changes.
If you exit the TGN command prompt when in flow mode, you will be in flow mode when you
reenter the TGN prompt. The same occurs with the pkts and filter commands.

expand Expanding a Traffic Stream into Multiple Copies

expand {n | imix [packet-mix packet_length_list]}

2-20 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 field Adding and Updating Configurable Fields

Makes copies of the currently selected traffic stream. In flow mode or when using the flow
command, the expand command makes copies of the currently selected flow member. If the
original traffic stream has incrementing, random, or iterating fields, the new traffic streams have
constant values in the fields, but are incremented, iterated, or random for each additional traffic
stream.
This command helps send out complex traffic more quickly. Updating a packet with changing
length and data can cause length fields to be updated and checksums to be recalculated, which
reduces the maximum send rate. If the memory is available, you can create multiple traffic
streams that do not need to be updated so that they can be sent at a faster rate.
expand n
Expands the currently selected traffic stream into the specified number of traffic streams.
expand imix [packet-mix packet_length_list]
Expands the currently selected traffic stream by creating a set of IMIX traffic streams. If
packet-mix is specified, the traffic stream is expanded into the specified number of streams with
the specified lengths in packet_length_list.
If packet-mix is not specified, the traffic stream is expanded into 12 traffic streams of the
following lengths in this order:
64 64 570 64 64 570 64 1518 570 64 64 570
You can use this command with a flow member to send out the IMIX traffic in order.

Note The expand command makes it easy to create lots of traffic streams, but each traffic stream
takes up router memory. If you use up all the free memory, the router crashes. The number of traffic
streams you can safely create depends on the amount of router free memory (use show memory at
exec) and the size and complexity of the packet definitions. Leave a couple of megabytes of memory
free for router processes and stacks.

flow Adding and Updating Packet Flows (page 2-25)

field Adding and Updating Configurable Fields

field add field-name
field insert-at {field# | field-name} name
field {field# | select field-name}
field [field# | select field-name] delete
field [field# | select field-name] name field-name
field [field# | select field-name] type {ip | n {decimal | hex} bytes | timestamp}
field [field# | select field-name] start-at sign-post offset bytes
field [field# | select field-name] data {number | ip-address}
field [field# | select field-name] data [increment | random] {number | ip-address} to
{number | ip-address} [no-reset]
field [field# | select field-name] data [iterate-thru [num-values n] start-index i values CSV
[nest-over field-name] [no-reset]]
These commands are used to create, delete, and maintain configurable fields in a traffic stream
packet definition. Configurable fields can be used to augment the field definitions supplied by
the templates. They define where a field starts in a packet, how long it is, the format of the data,
and what the data in the field is, whether constant, incrementing, or random.

Cisco Company Confidential Command Reference 2-21

field Adding and Updating Configurable Fields

When you use the [field# | select field-name] option, the specified field becomes the current field.
If you do not specify a field number or field name, the command is applied to the current field
(the field last accessed or modified).
field add field-name
Adds a configurable field, which becomes the current field. You must give the field a name. The
limit is 20 alpha-numeric characters; spaces are allowed.
field insert-at [field# | field-name] name
Creates a configurable field that is inserted in front of an existing field, identified by its number
or name. The new field becomes the current field. You must give the new field a name. The limit
is 20 alpha-numeric characters; spaces are allowed.
field [field# | select field-name] delete
Deletes an existing configurable field, identified by its number or name. The next lower field
becomes the current field.
field [field# | select field-name] name field-name
Changes the name assigned to a field.
field [field# | select field-name] type ip
field [field# | select field-name] type n {decimal | hex | bcd} bytes
field [field# | select field-name] type timestamp
field [field# | select field-name] type ascii
These commands define the field type. It can be an IP address field, a hex number, decimal field,
BCD field from 1 to 4 bytes long, or an 8 byte timestamp or an ASCII string.
For hex and decimal fields, there is no difference in data entry. Decimal or hex (with leading 0x)
values can be input into either type. This only affects how the field data is displayed.
For ASCII fields, the data is entered in the form of an ASCII string.
For BCD fields, the value entered is treated as BCD data. For example, these commands
configure the first byte of the field as 8 and the second byte as 9:
field type 2 bcd
field data 89

Timestamp field data cannot be entered. The TGN program updates this field with the IOS
timestamp before the packet is transmitted. This occurs before any transport checksums are
calculated, so that the timestamp can be added into a valid TCP or UDP packet. Turn off transport
checksumming if the checksum is not important to the test. For more information, see Using
TGN and PKTS Timestamps to Measure Latency (page 5-1).
field [field# | select field-name] start-at sign-post offset bytes
Defines where the field starts in the packet relative to well-known locations or sign posts, and
what the positive offset is from the sign post. Valid arguments for sign-post are:
packet-start
mac-address-start
dsap-address-start
network-start
transport-start
data-array-start
packet-end

2-22 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 field Adding and Updating Configurable Fields

Note For packet-end, the offset has be entered as a positive integer but is used as negative offset
from the end of packet.

Example of configuring a signature of 0x11223344 on the last 4 bytes of traffic

stream packets
tgn field add signature
tgn field type 4 hex
tgn field start-at packet-end offset 4
tgn field data 0x11223344

field [field# | select field-name] data {number | ip-address}

field [field# | select field-name] data [increment | random] {number | ip-address} to
{number | ip-address} [no-reset]
field [field# | select field-name] data [iterate-thru [num-values n] start-index i values CSV
[nest-over field-name] [no-reset]]
Puts data into the configurable field. The input data has to match what the field is configured for,
whether an IP address or number (decimal or hex). The data can be entered as a constant value
(the default), or incremented or random between a specified range. This cannot be used for a
timestamp field.
When a field is configured as iterate-thru, the set of values must be specified in as comma
separated values (CSV) with no space in between. The format depends on the type of the field.
The following formats are supported for CSV:
Decimal fieldsDecimal or hex format.
IP address fieldsMust be specified in Dotted Decimal Notation. For example, 127.1.3.10
MAC address fieldsMust be specified in x.x.x format.

Note If there are spaces between the comma separated values, they are ignored, and the entire
set of values must be enclosed in quotes. For example: L4-dest-port start-index 0 values 1, 2, 3,
4.

All incrementing and random fields are reset when traffic generation is started, unless the
no-reset option is specified.
Examples
field add internal-nets
Adds a configurable field named internal-nets.
field 1 start-at data-array-start offset 0
field select internal-nets start-at data-array-start offset 0
field start-at data-array-start offset 0
All three commands do the same thing: they start the field internal-nets (field number 1) at the
beginning of the data array.
field 1 data 25
Assigns the field a value of 25.
field 1 type 2 decimal bytes
field select internal-nets type 2 decimal bytes
field type 2 decimal bytes

Cisco Company Confidential Command Reference 2-23

fill-pattern Defining Data Pattern to Fill Packet

All three commands do the same thing: they make the field 2 bytes long, with its data displayed
in decimal.
The above commands result in a configurable field that displays as follows:
field 1
field name "internal-nets"
field type 2 decimal bytes
field start-at data-array-start offset 0
field data 25
In the following example, assume that there are already five configurable fields, and we want to
insert a field into the sequence at number 3.
field insert-at 3 server-address
Inserts a new configurable field named server-address in front of the current configurable field
3. The new field becomes field number 3.
field 3 start-at data-array-start offset 10
Starts the field server-address at data array byte 10.
field 3 type ip
Specifies that the field will have IP address data.
field data random 100.200.1.1 to 100.200.1.255
Specifies that the IP address data is random from 100.200.1.1 to 100.200.1.255.
The above commands result in a configurable field that displays as follows:
field 3
field name "server address"
field type ip
field start-at data-array-start offset 10
field data random 100.200.1.1 to 100.200.1.255

Iterating fields:
field type ip
field iterate-thru start-index 0 values 1.1.1.1,2.2.2.2,3.3.3.3

field type decimal

field iterate-thru start-index 3 values 1,0x100,50

ASCII field:
field type ascii
field data "string-value"

fill-pattern Defining Data Pattern to Fill Packet

fill-pattern {start-byte increment-by | random [with-update]}
When a traffic stream creates a packet to send, it first uses any L2, L3, and L4 headers configured
by the user. It then adds any data array information. If the packet is to be longer than the headers
and data array, it uses a fill pattern to create the remaining bytes of the packet.
The random option fills the bytes with randomly generated data.
If with-update is specified, the random data is updated for every packet sent out. Otherwise, the
random data is generated only once and used for all the packets in the traffic stream.

2-24 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 filter PKTS-FILTER Command Prompt Mode

The fill pattern is not random. It is defined by a starting byte value and an increment value that
all subsequent bytes are incremented by. By default, start-byte is 0x0, and increment-by is 0x01.

filter PKTS-FILTER Command Prompt Mode

filter
Switches to the PKTS program PKTS-FILTER command prompt mode to define PKTS selective
filters. See also pkts PKTS Command Prompt (page 2-34).

flow Adding and Updating Packet Flows

The following commands, when preceded by the keyword flow, allow you to add and update flows
from the TGN prompt. You can also use these commands without the flow keyword from flow mode
(see Using Flow Mode (page 1-8)). To exit flow mode and return to tgn mode, use the tgn command.
While in flow mode, you can use the pkts, filter, and end commands, which exit the TGN command
prompt. When the TGN prompt is exited from flow mode, you will be in flow mode when you
reenter the TGN command prompt.
end Exiting the TGN Command Prompt (page 2-20)
filter PKTS-FILTER Command Prompt Mode (page 2-25)
pkts PKTS Command Prompt (page 2-34)
The add and insert-at commands add or insert new members into the flow. They are similar to the
add (add Adding a Traffic Stream (page 2-10)) and insert-at (insert-at Inserting a Traffic Stream
(page 2-27)) commands.
add template [timestamp]
add pkts-packet pkt# [to pkt#] [timestamp]
add interface ts-name-or-number
add mixed-interface [primary | secondary slot-number] ts-name-or-number
add sniffer-file url
insert-at ts-name-or-number template [timestamp]
insert-at ts-name-or-number pkts-packet pkt# [timestamp]
insert-at ts-name-or-number interface ts-name-or-number
{member# | select name-or-number}Selects a flow member for update or review.
name character-stringAssigns a name to a member, limited to 39 characters.
all [template] [from member-name-or-number [to member-name-or-number]]Configures a
group of flow members.
delete [member-name-or-number]Deletes the specified member. If no member is specified, it
deletes the currently selected member.
interval millisecondsSpecifies the interval between consecutive members in the packet flow.
If the interval is zero, those two members are sent consecutively.
The following commands behave the same as those at the normal TGN command prompt:
start | start send | s | stop (see start/stop Starting and Stopping Traffic Generation (page 2-59))
clear [all | config | count] (see clear Clearing Configurations or Counters (page 2-16))
on | off (see on/off Activating or Deactivating a Traffic Stream (page 2-31))
expand n (see expand Expanding a Traffic Stream into Multiple Copies (page 2-20))
show (see show Displaying Traffic Stream and Summary Information (page 2-38))
write (see write Writing Information to an IFS Log File (page 2-61))

Cisco Company Confidential Command Reference 2-25

fragmentation Configuring IP Fragmentation

fragmentation Configuring IP Fragmentation

fragmentation {enable [mtu {auto | number_of_bytes}] | disable}
Enables or disables fragmentation. If fragmentation is enabled, it is only active in process output
mode. This command does not have any effect in fast, dedicated, and optimal send modes. If
fragmentation is enabled and mtu is specified as auto, the MTU of the outgoing interface is used.
fragmentation {option-data offset hex-string | option-length length}
Configures IP options as a hex string that is copied into fragments (except the first fragment).
The IP options configured by the commands L3-option-length and L3-option-data are copied
into the first fragment.
When fragmentation is enabled, the large packet is first fully constructed, that is, all the
incrementing, iterating, or random fields (including user-configured fields created with the field
command) are updated. The resulting IP datagram is fragmented if the packet length is greater
than the MTU specified. IP fragmentation is performed as specified by RFC 791.
The following fields in the IP header are copied unmodified into each of the fragments from the
original packet:
version, TOS, Identification, TTL, Protocol, Source Address, Destination Address
For each fragment, flags (3-bits) are set as follows:
Reserved bit, Dont Fragment bitSet to 0
More Fragments bitSet to 0 for the last fragment and 1 for other fragments
Fragment offset is calculated and updated for each fragment. For the flags and fragment offset
fields, the user-configured value is ignored in the fragments.
The fields Header-length, Total-length, and Header-checksum in the IP headers of the fragments
are updated as per the configurations in the original packet definition. If a field is configured as
auto, the field in each fragment is calculated and updated. If the field is configured as constant,
incrementing, or random, the same value is copied into each of the fragments.
fragmentation drop-fragments {enable [mode {random | constant {last | num}}] | disable}
Enables fragment dropping. By default, this is disabled. When drop-fragments is enabled,
the default mode is random. When mode is set to random, TGN randomly picks the
fragment to be dropped.
When mode is set to constant last, TGN drops the last fragment. If there is only one
fragment, it is dropped, and no fragment is sent.
When mode is set to constant num, if the length is constant and num is set to greater than
the number of resulting fragments, no fragments are dropped.
Use the show command to display the fragmentation-related configuration of a traffic stream.
When fragmentation is enabled, you can configure packets of length 65535 bytes (max-length of
an IP packet). This can be done with the length, data-length, or data commands.
output-mode Setting the Output Mode (page 2-32)
show Displaying a Traffic Stream or Flow Member (page 2-40)
show fragments-sent Displaying Number of Fragments Sent (page 2-48)
show packet fragments Displaying IP Packet Fragments (page 2-53)
length Setting Packet Length, (page 2-28)
data-length Setting the Data Array Length, (page 2-17)

2-26 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 insert-at Inserting a Traffic Stream

insert-at Inserting a Traffic Stream

This command is identical to the add command, except that insert-at inserts the new traffic stream
in front of an existing traffic stream rather than just adding it at the end.
The insert-at command takes the following arguments. For more details on implementing these
commands, see add Adding a Traffic Stream (page 2-10).
insert-at ts-name-or-number template [timestamp]
Creates a traffic stream based on a template.
insert-at ts-name-or-number pkts-packet pkt# [timestamp]
Creates a traffic stream based on a packet in PKTS capture buffer.
insert-at ts-name-or-number interface ts-name-or-number
Creates a traffic stream by cloning an existing traffic stream. The traffic stream to be cloned is
identified by the interface it is on and by its name or number.
insert-at ts-name-or-number {arp | aarp} responder
Creates an IP or AppleTalk ARP responder.
insert-at ts-name-or-number {decnet | clns} hello-generator
Creates a DECnet or CLNS hello-generator.

interval Setting the Interval Between Sending Packets

interval milliseconds
Sets the traffic stream sending rate. This command, which is an alternate to the rate command,
is useful when specifying slow send rates.
If the interface is configured for ordered traffic scheduling, the interval represents the time
between the scheduled departure of the traffic stream and the next traffic stream on the interface
list.
ordered-traffic Setting Ordered-Traffic Scheduling (page 2-32)
rate Setting the Packet Send Rate (page 2-35)
bit-rate - Setting the transmission Rate in bits per second (page 2-15)
variability Defining the Variability in Packet Intervals (page 2-60)
show rate Displaying Traffic Stream Rates (page 2-55)

isl-crc-added Adding CRC to ISL Packets

isl-crc-added [off | hardware | software]
This command only appears on IOS interfaces that support ISL.
An ISL packet must have a CRC or FCS that is calculated over the encapsulated packet. It is
placed after the encapsulated packet data.
The default is off. This must be set to off if the traffic stream does not define an ISL packet.
isl-crc-added hardware
Invokes the interface hardware to calculate and add the ISL CRC. This mode allows ISL packets
to be generated fast, with no impact on NQR.

Cisco Company Confidential Command Reference 2-27

layer Replacing the Template for a Specific Layer

The limitation is that the hardware also updates most of the ISL datalink header with valid
information. This helps create a valid ISL packet, but overwrites the datalink header data set in
the NQR traffic stream definition.
If you are using datalink ios-dependent isl-subinterface, you must activate this mode if you are
using output-mode fast or output-mode dedicated. This mode does not work with
output-mode optimal. See datalink Specifying the Datalink Header Encapsulation (page 2-18)
and output-mode Setting the Output Mode (page 2-32).
isl-crc-added software
Uses a software routine in Pagent to calculate and add the ISL CRC. The value of this mode is
that the datalink header defined by NQR is not changed by the transmission hardware, but there
is a performance impact.
This mode does not work on a 7500 VIP. If you need to use this mode on a 7500 VIP interface,
you must use the RP (primary processor) and not the VIP (secondary processor) to transmit the
packets. See secondary Selecting a SECONDARY Processor for Transmission (page 2-37).
This mode does not work with output-mode optimal.

layer Replacing the Template for a Specific Layer

See layer Replacing the Template for a Specific Layer (page 2-9).

length Setting Packet Length

length auto
length packet-length
length increment min-length to max-length [by inc-by]
length random min-length to max-length [by inc-by]
length iterate-thru start-index i [num-values n] values CSV
These commands set the length of a traffic streams packets.
length auto
Sets the packet length to the length of the L2, L3 and L4 headers (depending on the template)
plus the data array. The fill pattern is not added to the packet.
length packet-length
Sets the packet to a constant byte length.
The length can be less than the data array and the headers (I hope you know what you're doing).
If the length is greater than the headers and data array, the fill pattern is used to define the
additional bytes in the packet.
length increment min-length to max-length [by inc-by]
Increments a traffic streams packet from min-length to max-length. By default, the increment is
by 1 byte to max-length and restarts at min-length. To specify another amount, use the inc-by
option.
length random min-length to max-length [by inc-by]
Causes a traffic streams packet length to be random from min-length to max-length. By default,
the random length can be any value from min-length to max-length. The inc-by option causes the
packet length to be min-length plus multiples of inc-by, instead of multiples of 1.

2-28 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 load-config Loading a Configuration from IFS

length iterate-thru start-index i [num-values n] values CSV

Causes a traffic streams packet length to take the specified values. The values must be specified
as decimal or hex.

Note On Ethernet interfaces, if length is specified as > 1514 bytes, TGN automatically truncates the
packet to 1514 bytes without warning. An exception to this behavior is only when fragmentation is
enabled.

fragmentation Configuring IP Fragmentation, (page 2-26)

load-config Loading a Configuration from IFS

load-config url [append]
Loads a TGN traffic stream configuration file from IFS. It first deletes all existing traffic streams
on all interfaces unless the append option is configured. It then reads in and executes the
commands in the requested configuration file to create new traffic streams.
The traffic stream configuration file was created with the save-config command.
Examples
If you enter load-config ?, the program displays which file systems are available on the router.
c7513a-(TGN:Et0/0/0:none)#load-config ?
bootflash: Load config from bootflash:
disk0: Load config from disk0:
disk1: Load config from disk1:
flash: Load config from flash:
null: Load config from null:
nvram: Load config from nvram:
pram: Load config from pram:
rcp: Load config from rcp:
slavebootflash: Load config from slavebootflash:
slavenvram: Load config from slavenvram:
slaveslot0: Load config from slaveslot0:
slaveslot1: Load config from slaveslot1:
slot0: Load config from slot0:
slot1: Load config from slot1:
system: Load config from system:
tftp: Load config from tftp:

If you enter just the file system name, the program prompts you for the remaining information. For
example, you want to load a TGN configuration from TFTP server 192.1.1.2, and read the file
/tftpboot/tgn/test/traffic1.
c7513a-(TGN:Et0/0/0:2 of 2)#load tftp
Address or name of remote host []? 192.1.1.2
IFS filename []? tgn/test/traffic1
Please wait until 'Load Complete' message.

c7513a-(TGN:Et0/0/0:none)#
Loading tgn/test/traffic1 from 192.1.1.2 (via Ethernet0/0/0): !
[OK - 2360/4096 bytes]

Load Complete.

Cisco Company Confidential Command Reference 2-29

max-bit-rate Setting Interface Bandwidth Control

If the complete URL is entered, the program does not prompt for more information. In a TCL script,
you must use the complete URL, because CSCCON does not know how to respond to TGN IFS
prompts.
The following example shows using the command from the router exec with a complete URL.
c7513a-pagent#tgn load tftp://192.1.1.2/tgn/test/traffic1
Please wait until 'Load Complete' message.

c7513a-pagent#
Loading tgn/test/traffic1 from 192.1.1.2 (via Ethernet0/0/0): !
[OK - 2360/4096 bytes]

Load Complete.

If you need help in creating a URL, first open a file using the IFS prompts. You can then use the
show global command to see the complete URL.
replace - Selectively replacing IP Address and TCP/UDP Port Number (page 2-36)
IOS File System (page 1-10)
show global Displaying Global Parameters (page 2-47)

max-bit-rate Setting Interface Bandwidth Control

max-bit-rate {bits-per-second | off}
Specifies the maximum bits-per-second rate allowed on the currently selected interface or, under
broadcast mode, all interfaces. It calculates an interval based on the rate and size of the packet
just transmitted by TGN. The next scheduled packet on the same interface is transmitted after the
interval has elapsed.
Use the keyword off to turn the feature off.
show interface config Displaying Interface Configurations (page 2-49)

mixed-interface Defining Traffic Streams on a Mixed Interface

mixed-interface [on | off]
By default, TGN traffic streams are organized in a set of per-interface lists. In mixed interface
mode, traffic streams are organized in a single list instead. TGN only sends the traffic streams
defined in the mode under which the traffic generation command is issued.
on
TGN traffic streams across different interfaces (but processed by the same CPU) are organized
into a single list.
off
TGN traffic streams are organized into multiple lists based on the interface that the traffic stream
is associated with.
show global Displaying Global Parameters (page 2-47)

name Assigning a Name to a Traffic Stream

name character-string
Assigns a name to a traffic stream. The name is limited to 39 characters.

2-30 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 on/off Activating or Deactivating a Traffic Stream

In TGN mode, the primary purpose of this is to make it easy for a test script to select a traffic
stream for updating.
In Stimulus Response Engine (SRE) mode, the name is required, because SRE accesses packet
definitions by name.
select Selecting a Traffic Stream by Name (page 2-37)
show name Displaying Traffic Stream Names and Delayed-Start Information (page 2-51)

on/off Activating or Deactivating a Traffic Stream

Every traffic stream can be set to be active or inactive.
on
When a traffic stream is on, it sends packets when TGN traffic generation is started. ARP
responders respond to ARP requests, and a hello-generator sends hello packets every 30 seconds.
off
Entering off on a traffic stream deactivates it, and it does not send out packets. An ARP responder
does not respond to ARP requests, and a hello-generator does not send out hello packets.

open-logfile Opening an IFS Log File

open-logfile url
Opens an IFS log file before write commands are used to write data to the log file.
Most IOS file systems close an inactive file after a short idle time. After the file is opened, use
the write commands to log the information, with less than ten seconds delay between write
requests, and then close the log file with close-logfile.
If long-term logging is required, use the Pagent Remote Access Method (PRAM) file system. A
PRAM log file can be kept open for hours or days.
Examples
If you enter open-logfile ?, the program displays which file systems are available on the router.
c7513a-(TGN:Et0/0/0:2 of 2)#open ?
bootflash: Write log file to bootflash:
disk0: Write log file to disk0:
disk1: Write log file to disk1:
flash: Write log file to flash:
lex: Write log file to lex:
null: Write log file to null:
nvram: Write log file to nvram:
pram: Write log file to pram:
rcp: Write log file to rcp:
slavebootflash: Write log file to slavebootflash:
slavenvram: Write log file to slavenvram:
slaveslot0: Write log file to slaveslot0:
slaveslot1: Write log file to slaveslot1:
slot0: Write log file to slot0:
slot1: Write log file to slot1:
system: Write log file to system:
tftp: Write log file to tftp:

Cisco Company Confidential Command Reference 2-31

ordered-traffic Setting Ordered-Traffic Scheduling

If you enter just the file system name, the program prompts you for the remaining information. For
example, you want to log to TFTP server 192.1.1.2 and the file /tftpboot/tgn/test/log1.
c7513a-(TGN:Et0/0/0:2 of 2)#open tftp
Address or name of remote host []? 192.1.1.2
IFS filename []? tgn/test/log1
!

Now enter the write commands.

If the complete URL is entered, the program does not prompt for more information. In a TCL script,
you must use the complete URL, because CSCCON does not know how to respond to TGN IFS
prompts.
The following example shows using the command from the router exec with a complete URL.
c7513a-pagent#tgn open tftp://192.1.1.2/tgn/test/log1
!

If you need help in creating a URL, first open a file using the IFS prompts. You can then use the
show global command to see the complete URL.
write Writing Information to an IFS Log File (page 2-61)
IOS File System (page 1-10)
close-logfile Closing an Open IFS Log File (page 2-17)

ordered-traffic Setting Ordered-Traffic Scheduling

ordered-traffic [on | off]
Specifies whether ordered-traffic scheduling is on or off. The default is independent scheduling
(off). With ordered-traffic scheduling, traffic streams on the currently selected interface or, under
broadcast mode, all interfaces, are sent in the order of the traffic stream number.
Currently, ordered-traffic scheduling is only supported for process and fast-send output modes.
show interface config Displaying Interface Configurations (page 2-49)
output-mode Setting the Output Mode (page 2-32)

output-mode Setting the Output Mode

output-mode [all | primary | secondary {all | n}] {process | fast | dedicated | optimal}
Sets the output mode (packet generation mode) on each processor. The TGN program has four
different modes of sending packets: process, fast, dedicated, and optimal. The optimal mode is
available only on selected processors.
[all | primary | secondary {all | n}]
This option is available only when there are PRIMARY and SECONDARY processors.
If you use all, the output mode is set on the PRIMARY and all SECONDARY processors. This
is the default if all is not entered.
If you use primary, the output mode is set on only the PRIMARY processor.
If you use secondary {all | n}, the output mode is set on all SECONDARY processors, or only
the SECONDARY processor in slot n.
{process | fast | dedicated | optimal}
Selects the output mode.

2-32 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 output-mode Setting the Output Mode

process
In this mode, every time a traffic stream needs to send out a packet, it allocates a paktype
structure, copies in the packet headers, data array, and fill pattern, updates any incrementing
or random fields, updates any length or checksum fields, sends the packet, and then releases
the paktype.

Note For non-IOS programmers, IOS handles all incoming and outgoing packets through a data
structure called paktype, along with memory allocated through the paktype, to hold the packet.

The primary advantage of process mode is that traffic streams can be added, inserted,
deleted, and updated while traffic is being generated. This cannot be done in fast and
dedicated modes.
You can increase the output levels of this mode significantly using the repeat command.
repeat Resending Packets Repeatedly (page 2-35)
fast
This is the default output mode. In this mode, when traffic generation is started, a paktype
structure is allocated to every active traffic stream and the packet headers, data array, and fill
pattern are copied in. The paktype is not released until traffic generation is stopped.
When it is time for a traffic stream to send out a packet, fast mode updates incrementing,
random, length, and checksum fields, if needed, and sends the packet out.
Fast mode is faster that process mode, since it does not need to repeatedly allocate and delete
paktypes. In fast mode, the TGN program regularly releases to IOS, so that operating system,
router processes, and other test programs can run.
Traffic stream packets cannot be created, deleted, or updated while traffic is being output.
dedicated
Dedicated mode is like fast mode, except it does not release to the operating system until
traffic generation is stopped. In this mode, operating system, routing processes, and other test
programs do not get processing cycles.
When this mode is started, it posts the following message:
You have started traffic generation in dedicated output-mode.
TGN will go into a send loop that locks out all other processes.
Enter control-6 or shift-control-6 to stop traffic generation.

This mode is significantly faster than fast mode.

optimal
This mode is available only on some processors. Unlike the other output modes, this makes
use of specific capabilities of the hardware to send packets at higher rates.
In most cases, optimal mode will have limitations that the other modes do not. The
limitations can include the inability to change packet data or lengths, not support repeat, and
limits on packets lengths and the number of traffic streams it can support.

Cisco Company Confidential Command Reference 2-33

prompt Setting Command Prompt Format

When this mode is selected, the program posts a message indicating what limitations that
implementation has. For example:
c7513a-(TGN:OFF,Et0/0/0:1/1)#output-mode secondary 3 optimal
SECONDARY processor 3 optimal send:
This is a special VIP optimal send packet generation mode that sacrifices the
ability to update packet lengths, incrementing and random fields and bursting
for faster packet generation.

c7513a-(TGN:OFF,Et0/0/0:1/1)#output-mode primary optimal

PRIMARY processor optimal send:
This is a special RSP MEMD optimal send packet generation mode that sacrifices
the ability to update packet lengths, incrementing and random fields and
bursting for faster packet generation. The number of traffic streams is
limited by the amount of MEMD (about 700).

show global Displaying Global Parameters (page 2-47)

show output-mode Displaying Output Mode Information (page 2-52)

prompt Setting Command Prompt Format

prompt [static | dynamic]
Sets the format of the command prompt for all Pagent programs. Setting the format of the
command prompt for this program also sets it for all other Pagent programs, so you only have to
set it once if several programs are used.
By default, the command prompt format is set for dynamic, that is, it is constantly being updated
to give the tester the programs current status. In dynamic mode, the IOS hostname in the
command prompt is limited to seven characters.
The static mode is for test automation scripts. With static mode, the option section of the
command prompt displays PAGENT and the hostname is kept at its full length.
TGN Command Prompt Modes (page 1-7)
Using TCL Scripts (page 1-9)

pkts PKTS Command Prompt

pkts
Switches immediately to the PKTS program command prompt. See also filter PKTS-FILTER
Command Prompt Mode (page 2-25).
Note:
We have support to configure all the pkts commands directly from TGN prompt rather than
configuring by going into PKTS prompt.
Below is the snapshot view for more information
Router(TGN:OFF,Et0/0:none)#pkts ?
<1-4294967295> Select a captured packet by number.
Async Async interface
BVI Bridge-Group Virtual Interface
CDMA-Ix CDMA Ix interface
CTunnel CTunnel interface
2-34 TGN (Traffic GeNerator) User Manual Cisco Company Confidential
 rate Setting the Packet Send Rate

Dialer Dialer interface

Ethernet IEEE 802.3
Group-Async Async Group interface
Lex Lex interface
Loopback Loopback interface
MFR Multilink Frame Relay bundle interface
Multilink Multilink-group interface
Null Null interface
Serial Serial
Tunnel Tunnel interface
Vif PGM Multicast Host interface
Virtual-PPP Virtual PPP interface
Virtual-Template Virtual Template interface
Virtual-TokenRing Virtual TokenRing
add-capture-buffer Create packet capture buffer in router memory.
assign Configure how to decode UDP port numbers.
clear Clear interface configs, filters.
--More--

rate Setting the Packet Send Rate

rate packets-per-second
Sets the rate that a traffic stream sends packets. To define a slow rate, use the interval command.
To define rate in bits per second, use the bit-rate command.
interval Setting the Interval Between Sending Packets (page 2-27)
bit-rate - Setting the transmission Rate in bits per second (page 2-15)
variability Defining the Variability in Packet Intervals (page 2-60)
show rate Displaying Traffic Stream Rates (page 2-55)

repeat Resending Packets Repeatedly

repeat #packets [with-update | no-update]
Determines how many packets a traffic stream puts repeatedly on the output queue as fast as
possible before looking for the next traffic stream ready to send packets. The default is 1, which
means no looping.
with-update
Updates traffic streams with incrementing, iterating, or random packet lengths or fields with
each packet sent. By default, a traffic stream is defined with no update on repeat.
no-update

Cisco Company Confidential Command Reference 2-35

replace - Selectively replacing IP Address and TCP/UDP Port Number

Does not update packets while in the repeat loop (the default).

Caution Do not use a repeat greater than 1 on the 4500 and 4700 routers if using process output
mode, because it does not release I/O packet memory. After I/O memory is used up, all you see is
error messages and trace-backs. This is not a problem that affects router operation; it affects only
programs like this that try to send the same packet repeatedly in a tight loop.

replace - Selectively replacing IP Address and TCP/UDP Port Number

replace [ L3-ipv4-addr | L3-ipv6-addr | tcp-port | udp-port ] search-val by [ replace-val | range
min-val to max-val ]
Searches all the streams on the selected interface for search-val in source/destination and if found
replaces them by replace-val or the range (min-val to max-val) as the case may be. For Example:
replace L3-ipv4-addr 10.1.1.1 by 20.1.1.1 searches all the streams on the selected interface for
10.1.1.1 in Source or destination ip and if found, replaces it by 20.1.1.1
replace L3-ipv4-addr 1000 by range 2000 to 2010 searches all the streams on the selected
interface for 1000 in TCP Source or Dest port and if found, replaces it with the range 2000 to
2010.

save-config Saving a Configuration to IFS

save-config url
Saves the current configuration of all TGN traffic streams to IFS. The saved configuration can
be loaded later with the load-config command.
Examples
If you enter save-config ?, the program displays which file systems are available on the router.
c7513a-(TGN:Et0/0/0:2 of 2)#save ?
bootflash: Save config to bootflash:
disk0: Save config to disk0:
disk1: Save config to disk1:
flash: Save config to flash:
lex: Save config to lex:
null: Save config to null:
nvram: Save config to nvram:
pram: Save config to pram:
rcp: Save config to rcp:
slavebootflash: Save config to slavebootflash:
slavenvram: Save config to slavenvram:
slaveslot0: Save config to slaveslot0:
slaveslot1: Save config to slaveslot1:
slot0: Save config to slot0:
slot1: Save config to slot1:
system: Save config to system:
tftp: Save config to tftp:

If you enter just the file system name, the program prompts you for the remaining information. For
example, you want to save the configuration to TFTP server 192.1.1.2 and the file
/tftpboot/tgn/test/traffic1.
c7513a-(TGN:Et0/0/0:2 of 2)#save tftp
Address or name of remote host []? 192.1.1.2
IFS filename []? tgn/test/traffic1
!!

2-36 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 secondary Selecting a SECONDARY Processor for Transmission

Save complete.

If the complete URL is entered, the program does not prompt for more information. In a TCL script,
you must use the complete URL, because CSCCON does not know how to respond to TGN IFS
prompts.
The following example shows using the command from the router exec with a complete URL.
c7513a-pagent#tgn save tftp://192.1.1.2/tgn/test/traffic1
!!
Save complete.

If you need help in creating a URL, first open a file using the IFS prompts. You can then use the
show global command to see the complete URL.
load-config Loading a Configuration from IFS (page 2-29)
IOS File System (page 1-10)

secondary Selecting a SECONDARY Processor for Transmission

secondary {all | [slot] n} {on | off}
Specifies whether the traffic stream on the PRIMARY or the SECONDARY processor generates
packets. This command is only available if there are SECONDARY processors.
Most routers (2500, 4000, 4500, 7200) have only one processorthe PRIMARY processor. The
newer, high-end platforms support multiple processors, each running IOS. These platforms have
one PRIMARY processor and several SECONDARY processors.
For example, on the 75XX, the RSP is the PRIMARY processor, and the VIPs are the
SECONDARY processors. On the GSR, the Route Processor is the PRIMARY processor, and
the line cards are the SECONDARY processors.
When TGN creates traffic streams for an interface on a SECONDARY processor, the traffic
stream is created and maintained on both the PRIMARY and SECONDARY processors.
{all | [slot] n}
Applies to either all the SECONDARY processors or a single SECONDARY processor in
slot n. The word slot is not required.
{on | off}
on sets the SECONDARY processor to transmit packets. By default, the SECONDARY
processor is ON or active.
off sets the PRIMARY processor to transmit packets.
show secondary Displaying Activity Status of SECONDARY Processors (page 2-56)

select Selecting a Traffic Stream by Name

select character-string
Selects a traffic stream by the name assigned to the traffic stream. The command searches all
traffic streams on all interfaces. It selects the first traffic stream that is a complete match. You
must enter the complete name assigned to the traffic stream.
The primary purpose of this command is to make it easy for a test script to select an existing
traffic stream.

Cisco Company Confidential Command Reference 2-37

send Sending Packets

name Assigning a Name to a Traffic Stream (page 2-30)

send Sending Packets

send number-of-packets
Configures a traffic stream to send exactly the requested number of packets when the start send
command is entered.
show send Displaying Summary of Send Process (page 2-56)
start/stop Starting and Stopping Traffic Generation (page 2-59)

sequence Adding and Updating Packet Sequences

Note Packet flows are designed to replace TGN packet sequences. (Packet sequences will not be
developed further and are being maintained for backward compatibility only). You are strongly
encouraged to use TGN flows instead of TGN sequences (see flow Adding and Updating Packet
Flows (page 2-25)).

There are a number of advantages to packet flows over sequences. Packet flows offer an unequal
intermember interval (which can be random), and you can specify a delayed start for the flow. You
can configure and view flow members separately as a group, since each flow maintains it own list
of members. Sequence items are part of a traffic stream list.

sequence add {name | number}

Adds a packet sequence reference. Each packet in a packet sequence is defined as a traffic stream
and incorporated using a packet sequence reference. You can specify a traffic stream reference
by name or number.
sequence insert-at {name | number}
Inserts a packet sequence reference in front of the specified traffic stream reference. You can
specify the traffic stream reference either by name or number.
sequence reference# delete
Deletes a packet sequence reference.
sequence reference# enable
Enables a packet sequence reference.
sequence reference# disable
Disables a packet sequence reference, but does not permanently remove it from the sequence list.
sequence interval milliseconds
Specifies the interval, in milliseconds, between consecutive packets in the packet sequence. If the
interval is zero, all other traffic streams on the same interface are blocked during each iteration
of the packet sequence transmission.

show Displaying Traffic Stream and Summary Information

show option [template] [from ts-name-or-number [to ts-name-or-number]] [extended]

2-38 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 show Displaying Traffic Stream and Summary Information

The show commands display information on the console.

For every show command, there is an equivalent write command that displays the same
information on the console but also writes it to an IFS log file (see write Writing Information
to an IFS Log File (page 2-61)).
When the show and write commands are used in flow mode or with the flow command, they
display information about the members of the currently selected flow (see flow Adding and
Updating Packet Flows (page 2-25)).
You can use the following options singly or together with a show command to select specific
traffic streams to display.
template
Limits the display to traffic streams of a specific template. Valid templates are:
datalink
ip
arp
arp responder
tcp
udp
icmp
igmp
appletalk
aarp
aarp responder
ipx
decnet
decnet hello-generator
clns
clns hello-generator
from ts-name-or-number [to ts-name-or-number]
Selects traffic streams from a specified range. You can identify traffic streams either by name or
number. If identifying by name, you must enter the full exact name.
extended
Extends a summary to display information on all flow members in the specified range. You can
use this option with all commands that display a summary of traffic streams.
For example, if 3 and 6 are flow traffic streams, the following is displayed when you use the
extended option:
k4700-p(TGN:OFF,Et0:6/9)#show all extended

Summary of traffic streams on Ethernet0

ts# interface template interval/rate repeat state packet.length inc_by
1 Et0 TCP 10 1 on auto 54
2 Et0 UDP 10 1 on auto 42
3 Et0 Flow 10 on

Fl-3.1 Et0 APPLE 345678 1 on auto 35

Fl-3.2 Et0 DECNET 345678 1 on auto 37
Fl-3.3 Et0 Datalink 345678 1 on auto 14

4 Et0 Datalink 10 1 on auto 14

5 Et0 APPLE 10 1 on auto 35
6 Et0 Flow 10 on

Cisco Company Confidential Command Reference 2-39

show Displaying Traffic Stream and Summary Information

Fl-6.1 Et0 Datalink 0 1 on auto 14

Fl-6.2 Et0 APPLE 0 1 on auto 35

7 Et0 Datalink 10 1 on auto 14

8 Et0 APPLE 10 1 on auto 35
9 Et0 TCP 10 1 on auto 54

In the above output, FL in the ts# column indicate flow members by their flow number. For example,
FL-3.1 indicates member 1 in flow 3; FL-6.2 indicates member 2 in flow 6. Flow members do not
have a rate; they only have an interval to the next member.

show Displaying a Traffic Stream or Flow Member

show [tcl-output]
show n [tcl-output]
These commands display the configuration of a traffic stream or flow member. show displays the
configuration of the currently selected traffic stream or flow member. show n displays the
configuration of the traffic stream number specified on the currently selected interface or the
specified flow member.
See show traffic-stream Displaying a Traffic Stream by Name or Number (page 2-58) to
display a traffic stream by name.
The following example shows the output of an ICMP traffic stream on Ethernet:
k4700-p(TGN:OFF,Et0:3/3)#show

Traffic stream 3 of 3, ICMP, Ethernet0 (up)

name ""
on
rate 10
variability 0
send 0
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
!
datalink user-defined
length 1000
fragmentation enable mtu auto
fragmentation option-length 0
!
L2-encapsulation arpa
L2-dest-addr 0000.0000.0000
L2-src-addr 0060.3E58.1E1A
L2-protocol 0x0800
!
L3-version 4
L3-header-length auto
L3-tos 0x00
L3-length auto
L3-id 0x0000
L3-fragmentation 0x0000
L3-ttl 60
L3-protocol 1
L3-checksum auto
L3-src-addr 0.0.0.0
L3-dest-addr 0.0.0.0

2-40 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 show aarp Displaying AARP Header Information

L3-option-length 0
!
L4-type 0
L4-code 0
L4-checksum auto
L4-option 0x00000000
!
data-length 0
!
fill-pattern 0x00 0x01

If the tcl-output option is used, the configuration information to identify the traffic stream and
non-configuration information from the show rate and show send commands is displayed in a
TCL-friendly format.With TCL-friendly format, data is easy to extract from the output text because
it follows a unique keyword and is not row- and column-position dependent, which can change with
Pagent releases.
For example:
c7200-p(TGN:OFF,Et1/0:1/1)#sh tcl

interface Ethernet1/0
traffic-stream-number 1
name ip-test1
measured-rate 25779.917
packets-sent 368201
left-to-send 0

show aarp Displaying AARP Header Information

show aarp [selection-options]
Displays a summary of Appletalk ARP header field configurations. For the selection options, see
show Displaying Traffic Stream and Summary Information (page 2-38).
For example:
k4700-p(TGN:OFF,Et0:35/39)#sh aarp

Summary of AARP traffic streams on Ethernet0

sender sender target target
ts# operation mac_address net.node mac_address net.node
6 1 0000.0000.0000 0.0 0000.0000.0000 0.0
24 1 0000.0000.0000 0.0 0000.0000.0000 0.0
35 1 0000.0000.0000 0.0 0000.0000.0000 0.0

AARP (AppleTalk ARP) Network Header Field Update Commands (page 2-6)
Example of AARP (AppleTalk ARP) Traffic Stream (page A-12)
Explanation of IP ARP and Appletalk ARP Header Fields (page B-14)

show aarp-responder Displaying AARP Responders

show aarp responder [selection-options]
Displays a summary of Apple ARP responder configurations. For the selection options, see show
Displaying Traffic Stream and Summary Information (page 2-38).
For example:
tools75(TGN:OFF,Et2/2:54/54)#sh aarp resp
Summary of AARP Responder traffic streams on Ethernet2/2
ts# appletalk-address mac_address

Cisco Company Confidential Command Reference 2-41

show Displaying Traffic Stream and Summary Information

16 0.0 0011.2222.3333
17 0.0 0011.2222.3333
18 0.0 0011.2222.3333

AppleTalk ARP Responder (page 4-2)

Example of AARP (AppleTalk ARP) Responder (page A-21)

show all Displaying Summary of Traffic Streams or Flow Members

show all [selection-options]
Displays a summary of basic configuration variables common to all sending traffic streams or all
members of a flow. The information displayed is different depending whether the program is in
TGN or SRE mode. For the selection options, see show Displaying Traffic Stream and
Summary Information (page 2-38).
For example, in TGN mode the output looks like this:
tools75(TGN:OFF,Et2/2:54/54)#sh all
Summary of traffic streams on Ethernet2/2
ts# template interval/rate repeat state packet.length inc_by
1 IP 10 1 on auto 34
2 IP 10 1 on auto 34
3 IP 10 1 on auto 34
4 ARP 10 1 on auto 42
5 ARP 10 1 on auto 42
6 ARP 10 1 on auto 42
7 ARP Responder on
8 ARP Responder on
9 ARP Responder on
10 APPLE 10 1 on auto 35
11 APPLE 10 1 on auto 35
12 APPLE 10 1 on auto 35
13 AARP 10 1 on auto 50
14 AARP 10 1 on auto 50
15 AARP 10 1 on auto 50
16 AARP Responder on
17 AARP Responder on
18 AARP Responder on
19 TCP 10 1 on auto 54
20 TCP 10 1 on auto 54
21 TCP 10 1 on auto 54

In SRE mode, the output looks like this:

c7513a-(TGN-SRE:OFF,Et0/0/0:5/5)#sh all

Summary of SRE traffic streams on Ethernet0/0/0

ts# template sre.name packet.length
1 IP auto 34
2 IP auto 34
3 IP auto 34
4 TCP auto 54
5 UDP auto 42

sre Defining Traffic Streams for TGN or SRE (page 2-59)

show appletalk Displaying AppleTalk Header Information

show appletalk [selection-options]

2-42 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 show arp Displaying ARP Header Information

Displays a summary of AppleTalk header field configurations. For the selection options, see
show Displaying Traffic Stream and Summary Information (page 2-38).
For example:
k4700-p(TGN:OFF,Et0:28/42)#sh appl

Summary of Appletalk traffic streams on Ethernet0

hop- <---destination--> <-----source----->
ts# count length chksm network node skt network node skt type
6 0 13 0000 0 0 0 0 0 0 0
14 0 13 0000 0 0 0 0 0 0 0
28 0 13 0000 0 0 0 0 0 0 0

AppleTalk Phase 1 Network Header Field Update Commands (page 2-5)

Example of AppleTalk Phase 2 Traffic Stream (page A-7)
Explanation of AppleTalk Header Fields (page B-10)

show arp Displaying ARP Header Information

show arp [selection-options]
Shows a summary of IP ARP header field configurations. For the selection options, see show
Displaying Traffic Stream and Summary Information (page 2-38).
For example:
k4700-p(TGN:OFF,Et0:37/37)#sh arp

Summary of ARP traffic streams on Ethernet0

oper- sender sender target target
ts# ation mac_address ip_address mac_address ip_address
8 1 0000.0000.0000 0.0.0.0 0000.0000.0000 0.0.0.0
20 1 0000.0000.0000 0.0.0.0 0000.0000.0000 0.0.0.0
37 1 0000.0000.0000 0.0.0.0 0000.0000.0000 0.0.0.0

ARP Network Header Field Update Commands (page 2-4)

Example of ARP (IP) Traffic Stream (page A-11)
Explanation of IP ARP and Appletalk ARP Header Fields (page B-14)

show arp-responder Displaying ARP Responders

show arp responder [selection-options]
Shows a summary of IP ARP responder configurations. For the selection options, see show
Displaying Traffic Stream and Summary Information (page 2-38).
For example:
tools75(TGN:OFF,Et2/2:54/54)#sh arp resp
Summary of ARP Responder traffic streams on Ethernet2/2
ts# ip-address mac_address
7 0.0.0.0 0011.2222.3333
8 0.0.0.0 0011.2222.3333
9 0.0.0.0 0011.2222.3333

IP ARP Responder (page 4-1)

Example of IP ARP Responder (page A-21)

Cisco Company Confidential Command Reference 2-43

show Displaying Traffic Stream and Summary Information

show burst Displaying Burst Configurations

show burst [selection-options]
Shows a summary of burst configurations. For the selection options, see show Displaying
Traffic Stream and Summary Information (page 2-38).
For example:
tools75(TGN:OFF,Et2/2:54/54)#sh bur
Summary of burst configurations on Ethernet2/2
burst duration.on duration.off
ts# template state min max min max
1 IP off 1000 1000 1000 1000
2 IP off 1000 1000 1000 1000
3 IP off 1000 1000 1000 1000
4 ARP off 1000 1000 1000 1000
5 ARP off 1000 1000 1000 1000

burst Sending Traffic Stream in Bursts (page 2-16)

show clns Displaying CLNS Header Information

There are so many fields in a CLNS header that it takes two separate display commands to show it
all. One is only for the source and destination addresses, the other for the remaining header fields.
show clns [selection-options]
Displays a summary of CLNS header non-address fields. For the selection options, see show
Displaying Traffic Stream and Summary Information (page 2-38).
For example:
k4700-p(TGN:OFF,Et0:35/35)#sh clns

Summary of CLNS traffic streams on Ethernet0

header
ts# id length version lifetime flags segment.length checksum
8 129 39 1 100 0x1E 39 0000
20 129 39 1 100 0x1E 39 0000
35 129 39 1 100 0x1E 39 0000

show clns address [selection-options]

Displays a summary of CLNS header address fields. Each source and destination address is on
its own line.
For example:
k4700-p(TGN:OFF,Et0:35/51)#sh clns add

Summary of CLNS traffic streams addresses on Ethernet0

ts# len area host mac protocol
8 dest 14 47.0000.0000.0000 0000.0000.0000 00
src 14 47.0000.0000.0000 0000.0000.0000 00
20 dest 14 47.0000.0000.0000 0000.0000.0000 00
src 14 47.0000.0000.0000 0000.0000.0000 00
35 dest 14 47.0000.0000.0000 0000.0000.0000 00
src 14 47.0000.0000.0000 0000.0000.0000 00

CLNS Area Fields (page 3-5)

CLNS Network Header Field Update Commands (page 2-6)
Example of CLNS Traffic Stream (page A-9)
Explanation of CLNS Header Fields (page B-12)
2-44 TGN (Traffic GeNerator) User Manual Cisco Company Confidential
 show clns-hello-generator Displaying CLNS Hello-generators

show clns-hello-generator Displaying CLNS Hello-generators

show clns hello-generator [selection-options]
Displays a summary of CLNS hello-generator configurations. For the selection options, see show
Displaying Traffic Stream and Summary Information (page 2-38).
For example:
tools75(TGN:OFF,Et2/2:54/54)#sh clns hell
Summary of CLNS Hello-Generators on Ethernet2/2
ts# clns hello-address
52 47.0000.0000.0000.0011.2222.3333.00
53 47.0000.0000.0000.0011.2222.3333.00
54 47.0000.0000.0000.0011.2222.3333.00

CLNS Hello-Generator (page 4-2)

Example of CLNS Hello-Generator (page A-21)

show config Displaying Traffic Stream Configuration Commands

show config [all]
Displays the commands used to configure either the currently selected traffic stream or all traffic
streams (if the all keyword is included). These are the same commands that would be written to
an IFS file by the save-config command.
You can use these commands in a script. The resulting configuration commands must be executed
at the router exec prompt.
This command displays the following output:
k4700-p(TGN:OFF,Et0:3/3)#show config

tgn Add ICMP

tgn name ""
tgn on
tgn rate 10
tgn variability 0
tgn send 0
tgn repeat 1 no-update
tgn delayed-start random
tgn burst off
tgn burst duration on 1000 to 1000
tgn burst duration off 1000 to 1000
!
tgn datalink user-defined
tgn length 1000
tgn fragmentation enable mtu auto
tgn fragmentation option-length 0
!
tgn L2-encapsulation arpa
tgn L2-dest-addr 0000.0000.0000
tgn L2-src-addr 0060.3E58.1E1A
tgn L2-protocol 0x0800
!
tgn L3-version 4
tgn L3-header-length auto
tgn L3-tos 0x00
tgn L3-length auto
tgn L3-id 0x0000
tgn L3-fragmentation 0x0000
tgn L3-ttl 60
tgn L3-protocol 1

Cisco Company Confidential Command Reference 2-45

show Displaying Traffic Stream and Summary Information

tgn L3-checksum auto

tgn L3-src-addr 0.0.0.0
tgn L3-dest-addr 0.0.0.0
tgn L3-option-length 0
!
tgn L4-type 0
tgn L4-code 0
tgn L4-checksum auto
tgn L4-option 0x00000000
!
tgn data-length 0
!
tgn fill-pattern 0x00 0x01

show debug Displaying Debugging Information for Program Developers

show debug
Displays internal program data that is of interest only to the program developers.

show decnet Displaying DECnet Header Information

show decnet [selection-options]
Displays a summary of DECnet header field configurations. For the selection options, see show
Displaying Traffic Stream and Summary Information (page 2-38).
For example:
k4700-p(TGN:OFF,Et0:35/38)#sh dec

Summary of DECnet traffic streams on Ethernet0

ts# length flag dest source nl2 visits service protocol
4 21 0x06 0.0 0.0 0 0 0 0
20 21 0x06 0.0 0.0 0 0 0 0
35 21 0x06 0.0 0.0 0 0 0 0

DECnet Network Header Field Update Commands (page 2-6)

Example of DECnet Traffic Stream (page A-10)
Explanation of DECnet Header Fields (page B-11)

show decnet-hello Displaying DECnet Hello-Generators

show decnet hello-generator [selection-options]
Displays a summary of DECnet hello-generator configurations. For the selection options, see
show Displaying Traffic Stream and Summary Information (page 2-38).
For example:
tools75(TGN:OFF,Et2/2:54/54)#sh dec hello
Summary of DECNET Hello-Generators on Ethernet2/2
ts# decnet hello-address designated-router
46 0.0 0.0
47 0.0 0.0
48 0.0 0.0

DECnet Hello-Generator (page 4-3)

Example of DECnet Hello-Generator (page A-21)

2-46 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 show flow Displaying Summary of Packet Flows

show flow Displaying Summary of Packet Flows

show flow [selection-options]
Displays a summary of packet flow traffic streams.
For example:
k4700-p(TGN:OFF,Et0:10/10)#sh flow

Summary of packet flows on Ethernet0

ts# total enabled interval/rate state name
3 5 5 10 on
6 5 3 10 on
10 3 0 10 on

show global Displaying Global Parameters

show global
Displays TGN configuration variables that are not specific to any traffic stream. If URLs have
been established for configuration and/or log files, these are also displayed.
This is an example of output on a single processor router:
c4700-p(TGN:OFF,Et1:2/2)#sh gl
ifs logging tftp://192.1.1.2/tgn/test/log1
ifs config pram://192.1.1.2/tgn-config
decode-templates built-in
display-level terse
mixed-interface off
flow-mode off
output-mode fast
sre off
verbose off
verbose logging-to console
wait-to-release 1

This is an example of output on a router with PRIMARY and SECONDARY processors:

c7513a-(TGN:OFF,Et0/0/0:none)#sh gl
ifs logging tftp://192.1.1.2/tgn/test/log2
ifs config pram://192.1.1.2/tgn-config2
decode-templates built-in
display-level terse
mixed-interface off
output-mode primary fast
output-mode secondary 0 fast
output-mode secondary 1 fast
output-mode secondary 2 fast
output-mode secondary 3 fast
output-mode secondary 4 fast
output-mode secondary 5 fast
secondary 0 on
secondary 1 on
secondary 2 on
secondary 3 on
secondary 4 on
secondary 5 on
sre off
verbose off
verbose logging-to console

Cisco Company Confidential Command Reference 2-47

show Displaying Traffic Stream and Summary Information

open-logfile Opening an IFS Log File (page 2-31)

replace - Selectively replacing IP Address and TCP/UDP Port Number (page 2-36)
max-bit-rate Setting Interface Bandwidth Control (page 2-30)
ordered-traffic Setting Ordered-Traffic Scheduling (page 2-32)
secondary Selecting a SECONDARY Processor for Transmission (page 2-37)
sre Defining Traffic Streams for TGN or SRE (page 2-59)
verbose - Configuring for Activity Messages (page 2-60)
wait-to-release Sierra Wait-to-Release Paktype (page 2-61)

show fragments-sent Displaying Number of Fragments Sent

show fragments-sent [selection-options]
Displays a summary of the number of fragments sent by each traffic stream. It shows a
cumulative number since the last time the counter was cleared. The counter can be cleared with
the clear counts command.
For example:
k4700-p(TGN:OFF,Et0:3/3)#sh fragments-sent

Summary of packet fragments sent for traffic streams on Ethernet0

ts# template state total-fragments-sent
1 TCP on 270
2 UDP on 450
3 ICMP on 89

k4700-p(TGN:OFF,Et0:3/3)#

clear Clearing Configurations or Counters (page 2-16)

fragmentation Configuring IP Fragmentation (page 2-26)
show send Displaying Summary of Send Process (page 2-56)

show icmp Displaying ICMP Header Information

show icmp [selection-options]
Displays a summary of ICMP header field configurations. For the selection options, see show
Displaying Traffic Stream and Summary Information (page 2-38). For example:
tools75(TGN:OFF,Et2/2:54/54)#sh icmp
Summary of ICMP traffic streams on Ethernet2/2
ts# type code checksum option
25 0 0 0x0000 0x00000000
26 0 0 0x0000 0x00000000
27 0 0 0x0000 0x00000000

ICMP Transport Header Field Update Commands (page 2-7)

Example of ICMP Traffic Stream (page A-15)
Explanation of ICMP Header Fields (page B-4)

show igmp Displaying IGMP Header Information

show igmp [selection-options]
Displays a summary of IGMP header field configurations. For the selection options, see show
Displaying Traffic Stream and Summary Information (page 2-38).

2-48 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 show interface Displaying Interface Status

For example:
tools75(TGN:OFF,Et2/2:54/54)#sh igmp
Summary of IGMP traffic streams on Ethernet2/2
ts# version type checksum group.address
28 0 0 0x0000 0.0.0.0
29 0 0 0x0000 0.0.0.0
30 0 0 0x0000 0.0.0.0

IGMP Transport Header Field Update Commands (page 2-9)

Example of IGMP Traffic Stream (page A-16)
Explanation of IGMP Header Fields (page B-5)

show interface Displaying Interface Status

show interface
Displays a summary of TGN interface status and the number of traffic streams configured on
each interface.
For example:
c4700-p(TGN:ON,Et1:3/4)#sh int

# interface admin.state operational.state num.traffic.streams

1 Ethernet0 up up 0
2 Ethernet1 up up 4
3 Serial0 down down 0
4 Serial1 shut down 0
5 Serial2 shut down 0
6 Serial3 shut down 0
7 TokenRing0 up up 0
8 TokenRing1 shut down 0

Admin.state is the software status of the interface.

Operational.state is the hardware status of the interface.
If you have a back-to-back crossover connection between two Ethernet 10BaseT interfaces and the
interface on the router running TGN is not shut, but the other ethernet interface is shut, the other
interface will report the admin.state as up and the operational.state as down.

show interface config Displaying Interface Configurations

show interface config
Displays the traffic scheduling style and maximum bit rate of each interface. For example:
b7513-p(TGN:OFF,Fa4/0:none)#show interface config

ordered- maximum bit

# interface traffic rate (bps)
1 Hssi2/0/0 off off
2 Serial2/1/0 off off
3 FastEthernet3/0/0 off off
4 FastEthernet4/0 off 1000
5 FastEthernet4/1 off off
6 FastEthernet10/0/0 off off

max-bit-rate Setting Interface Bandwidth Control (page 2-30)

ordered-traffic Setting Ordered-Traffic Scheduling (page 2-32)

Cisco Company Confidential Command Reference 2-49

show Displaying Traffic Stream and Summary Information

show interface max-bit-rate Displaying Maximum Bit Rates of Interfaces

show interface max-bit-rate
This command has been replaced with the show interface config command. See show interface
config Displaying Interface Configurations (page 2-49).

show interface tcl-output Displaying Interface Info in TCL-Friendly Format

show interface tcl-output
Displays the same information for the currently selected interface that is available with the show
interface and show rate summary commands but in a TCL-friendly format. It is easy to extract
data in TCL-friendly format from the output text because it follows a unique keyword and is not
row- and column-position dependent, which can change with Pagent releases.
For example:
c7200-p(TGN:OFF,Et1/0:1/1)#sh int tcl

interface Ethernet1/0
admin-state up
operational-state up
traffic-stream-count 1
ordered-traffic off
max-bit-rate off
measured-rate 10856.200
packets-sent 369361

show rate Displaying Traffic Stream Rates (page 2-55)

show ip Displaying IP Header Information

show ip [selection-options]
Displays a summary of IP header field configurations. For the selection options, see show
Displaying Traffic Stream and Summary Information (page 2-38).
For example:
k4700-p(TGN:OFF,Et0:24/46)#sh ip

Summary of IP traffic streams on Ethernet0

prot-
ts# tos len id frag ttl col chksm source destination
1 TCP 00 40 0000 0000 60 6 7ED1 0.0.0.0 0.0.0.0
2 TCP 00 40 0000 0000 60 6 7ED1 0.0.0.0 0.0.0.0
3 TCP 00 40 0000 0000 60 6 7ED1 0.0.0.0 0.0.0.0
5 IGMP 00 28 0000 0000 60 2 7EE1 0.0.0.0 0.0.0.0
6 ICMP 00 28 0000 0000 60 1 7EE2 0.0.0.0 0.0.0.0
9 TCP 00 40 0000 0000 60 6 7ED1 0.0.0.0 0.0.0.0
10 TCP 00 40 0000 0000 60 6 7ED1 0.0.0.0 0.0.0.0
11 TCP 00 40 0000 0000 60 6 7ED1 0.0.0.0 0.0.0.0
12 IGMP 00 28 0000 0000 60 2 7EE1 0.0.0.0 0.0.0.0
14 UDP 00 28 0000 0000 60 17 7ED2 0.0.0.0 0.0.0.0
15 UDP 00 28 0000 0000 60 17 7ED2 0.0.0.0 0.0.0.0
16 UDP 00 28 0000 0000 60 17 7ED2 0.0.0.0 0.0.0.0
17 UDP 00 28 0000 0000 60 17 7ED2 0.0.0.0 0.0.0.0
18 ICMP 00 28 0000 0000 60 1 7EE2 0.0.0.0 0.0.0.0
19 UDP 00 28 0000 0000 60 17 7ED2 0.0.0.0 0.0.0.0
20 UDP 00 28 0000 0000 60 17 7ED2 0.0.0.0 0.0.0.0
21 UDP 00 28 0000 0000 60 17 7ED2 0.0.0.0 0.0.0.0

2-50 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 show ipx Displaying IPX Header Information

22 UDP 00 28 0000 0000 60 17 7ED2 0.0.0.0 0.0.0.0

23 UDP 00 28 0000 0000 60 17 7ED2 0.0.0.0 0.0.0.0
24 IGMP 00 28 0000 0000 60 2 7EE1 0.0.0.0 0.0.0.0

IP Network Header Field Update Commands (page 2-4)

Example of IP Traffic Stream (page A-6)
Explanation of IP Header Fields (page B-1)

show ipx Displaying IPX Header Information

show ipx [selection-options]
Displays a summary of IPX header field configurations. For the selection options, see show
Displaying Traffic Stream and Summary Information (page 2-38). For example:
tools75(TGN:OFF,Et2/2:54/54)#sh ipx
Summary of IPX traffic streams on Ethernet2/2
tran pkt <------ destination---------> <----------source----------->
ts# len ctr typ network host skt network host skt
31 30 00 00 00000000 0000.0000.0000 0 00000000 0000.0000.0000 0
32 30 00 00 00000000 0000.0000.0000 0 00000000 0000.0000.0000 0
33 30 00 00 00000000 0000.0000.0000 0 00000000 0000.0000.0000 0

IPX Network Header Field Update Commands (page 2-5)

Example of IPX Traffic Stream (page A-7)
Explanation of IPX Header Fields (page B-6)

show mac Displaying MAC Addresses

show mac [selection-options]
Displays a summary of datalink MAC addresses from LAN datalink header field configurations.
For the selection options, see show Displaying Traffic Stream and Summary Information
(page 2-38).
For example:
tools75(TGN:OFF,Et2/2:54/54)#sh mac
Summary of MAC addresses on Ethernet2/2
ts# template encap mac.destination mac.source dsap ssap protocol
1 IP arpa 0000.0000.0000 0000.0000.0000 0x0800
2 IP arpa 0000.0000.0000 0000.0000.0000 0x0800
3 IP arpa 0000.0000.0000 0000.0000.0000 0x0800
4 ARP arpa 0000.0000.0000 0000.0000.0000 0x0806
5 ARP arpa 0000.0000.0000 0000.0000.0000 0x0806
6 ARP arpa 0000.0000.0000 0000.0000.0000 0x0806

show name Displaying Traffic Stream Names and Delayed-Start Information

show name [selection-options]
Displays a summary of names assigned to traffic streams. This command also shows a summary
of delayed-start definitions. For the selection options, see show Displaying Traffic Stream and
Summary Information (page 2-38).
For example:
c4700-p(TGN:OFF,Et1:3/3)#sh na

Summary of traffic stream names on Ethernet1

Cisco Company Confidential Command Reference 2-51

show Displaying Traffic Stream and Summary Information

ts# template state delayed-start name

1 IP on 0 test traffic 1
2 TCP on random TCP small packets
3 TCP on 1000ms TCP big packets

name Assigning a Name to a Traffic Stream (page 2-30)

datalink Specifying the Datalink Header Encapsulation (page 2-18)
select Selecting a Traffic Stream by Name (page 2-37)

show output-mode Displaying Output Mode Information

show output-mode
Displays what the output mode is on all processors.
For example:
c7513a-(TGN:OFF,Et0/0/0:none)#sh out

current optimal-send only concurrent

processor output-mode available optimal-send optimal-send
primary fast yes no no
secondary 0 process yes no no
secondary 1 fast yes no no
secondary 2 dedicated yes no no
secondary 3 optimal yes no no
secondary 4 optimal yes no no
secondary 5 fast yes no no

Processor identifies the processor. Current output-mode lists the current mode on the processor
(see ordered-traffic Setting Ordered-Traffic Scheduling (page 2-32)). Optimal output-mode
states whether the optimal mode is available on this processor. If optimal output mode is available,
only-optimal send states if it is the only mode available. Some SECONDARY processors cannot
run a full IOS, but might have an optimal output mode implemented in microcode. Concurrent
optimal-send identifies whether this optimal mode implementation allows other IOS processes to
run concurrently (yes) or does it grab all processor cycles (no).

show packet Displaying Packet Sent by Traffic Stream

show packet [n | traffic-stream ts-name-or-number] [hex]
Displays a formatted version of a packet resulting from a traffic stream definition. By default, the
currently selected traffic stream is displayed, or you can select a specific traffic stream by name
or number. Use the hex option to display the packets in hex.
For example:
tools75(TGN:OFF,Et1/1/0,1/1)#sh pac

Ethernet Packet: 54 bytes

Dest Addr: 0000.0000.0000, Source Addr: 0000.0000.0000
Protocol: 0x0800

IP Version: 0x4, HdrLen: 0x5, TOS: 0x00

Length: 40, ID: 0x0000, Flags-Offset: 0x0000
TTL: 60, Protocol: 6 (TCP), Checksum: 0x7ED1 (OK)
Source: 0.0.0.0, Dest: 0.0.0.0

TCP Src Port: 0 (Reserved), Dest Port: 0 (Reserved)

Seq #: 0x00000000, Ack #: 0x00000000, Hdr_Len: 5
Flags: 0x00, Window: 0, Checksum: 0x0000

2-52 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 show packet fragments Displaying IP Packet Fragments

Urgent Pointer: 0

This example displays the packet in hex format.

tools75(TGN:OFF,Et1/1/0,1/1)#sh pac hex

0 : 0000 0000 0000 0000 0000 0000 0800 4500 0028 0000 ..............E..(..
20 : 0000 3C06 7ED1 0000 0000 0000 0000 0000 0000 0000 ....................
40 : 0000 0000 0000 5000 0000 0000 0000 0000 0000 0000 ......P.............

show packet fragments Displaying IP Packet Fragments

show packet fragments
Displays a formatted version of packet fragments resulting from a traffic stream definition if
fragmentation is enabled (see fragmentation Configuring IP Fragmentation (page 2-26)).
For example:
r4500a-(TGN:OFF,Et0:1/1)#show packet fragments

Original Packet:

Ethernet Packet: 200 bytes

Dest Addr: 0000.0000.0000, Source Addr: 0010.7B2C.79F2
Protocol: 0x0800

IP Version: 0x4, HdrLen: 0x5, TOS: 0x00

Length: 186, ID: 0x0000, Flags-Offset: 0x0000
TTL: 60, Protocol: 6 (TCP), Checksum: 0x7E3F (OK)
Source: 0.0.0.0, Dest: 0.0.0.0

TCP Src Port: 0 (Reserved), Dest Port: 0 (Reserved)

Seq #: 0x00000000, Ack #: 0x00000000, Hdr_Len: 5
Flags: 0x00, Window: 0, Checksum: 0x126E (OK)
Urgent Pointer: 0

Data:
0 : 0001 0203 0405 0607 0809 0A0B 0C0D 0E0F 1011 1213 ....................
<snip>
140 : 8C8D 8E8F 9091 ......

Fragment 1:

Ethernet Packet: 114 bytes

Dest Addr: 0000.0000.0000, Source Addr: 0010.7B2C.79F2
Protocol: 0x0800

IP Version: 0x4, HdrLen: 0x5, TOS: 0x00

Length: 100, ID: 0x0000, Flags-Offset: 0x2000 (more fragments)
TTL: 60, Protocol: 6 (TCP), Checksum: 0x5E95 (OK)
Source: 0.0.0.0, Dest: 0.0.0.0

TCP Src Port: 0 (Reserved), Dest Port: 0 (Reserved)

Seq #: 0x00000000, Ack #: 0x00000000, Hdr_Len: 5
Flags: 0x00, Window: 0, Checksum: 0x126E ERROR: 33B4
Urgent Pointer: 0

Data:
0 : 0001 0203 0405 0607 0809 0A0B 0C0D 0E0F 1011 1213 ....................
20 : 1415 1617 1819 1A1B 1C1D 1E1F 2021 2223 2425 2627 ............ !"#$%.
40 : 2829 2A2B 2C2D 2E2F 3031 3233 3435 3637 3839 3A3B ()*+,-./0123456789:;

Cisco Company Confidential Command Reference 2-53

show Displaying Traffic Stream and Summary Information

Fragment 2:

Ethernet Packet: 114 bytes

Dest Addr: 0000.0000.0000, Source Addr: 0010.7B2C.79F2
Protocol: 0x0800

IP Version: 0x4, HdrLen: 0x5, TOS: 0x00

Length: 100, ID: 0x0000, Flags-Offset: 0x200A (more fragments)
TTL: 60, Protocol: 6 (TCP), Checksum: 0x5E8B (OK)
Source: 0.0.0.0, Dest: 0.0.0.0

Data:
0 : 3C3D 3E3F 4041 4243 4445 4647 4849 4A4B 4C4D 4E4F .=.?@ABCDEFGHIJKLMNO
20 : 5051 5253 5455 5657 5859 5A5B 5C5D 5E5F 6061 6263 PQRSTUVWXYZ[\
40 : 6465 6667 6869 6A6B 6C6D 6E6F 7071 7273 7475 7677 defghijklmnop
60 : 7879 7A7B 7C7D 7E7F 8081 8283 8485 8687 8889 8A8B xyz{|}..............

Fragment 3:

Ethernet Packet: 40 bytes

Dest Addr: 0000.0000.0000, Source Addr: 0010.7B2C.79F2
Protocol: 0x0800

IP Version: 0x4, HdrLen: 0x5, TOS: 0x00

Length: 26, ID: 0x0000, Flags-Offset: 0x0014
TTL: 60, Protocol: 6 (TCP), Checksum: 0x7ECB (OK)
Source: 0.0.0.0, Dest: 0.0.0.0

Data:
0 : 8C8D 8E8F 9091 ......

r4500a-(TGN:OFF,Et0:1/1)#

show pagent-format Displaying a Packet in Pagent Format

show pagent-format [n]
Displays a traffic stream packet in a format that can be used to input the definition into the classic
Pagent program.
By default, this command displays the currently selected traffic stream, unless n is used to select
another traffic stream.
For example:
c7513a-(TGN:OFF,Et0/0/0:2/2)#sh pag

interface Ethernet0/0/0
add $$ byte 0
00E0F759 E50700E0 34C5E800 08004500 00640001 0000FF01 79758009 0116C001
01020800 6B630798 1A440000 00000002 F108ABCD ABCDABCD ABCDABCD ABCDABCD
ABCDABCD ABCDABCD ABCDABCD ABCDABCD ABCDABCD ABCDABCD ABCDABCD ABCDABCD
ABCDABCD ABCDABCD ABCDABCD ABCDABCD ABCD
quit

show program-status Displaying Current Program Status

show program-status

2-54 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 show rate Displaying Traffic Stream Rates

Displays the same information available in the TGN option prompt. This command is useful
when running a script from the router exec prompt, and the program option prompt is not
available.
For example:
k4700-p(TGN:OFF,Et0:22/22)#sh prog

Program: TGN
State: OFF
Broadcast: OFF
Flow_mode: OFF
Selected Interface: Ethernet0
Selected Traffic Stream: 22 of 22

show rate Displaying Traffic Stream Rates

show rate [summary] [selection-options]
Displays a summary of the traffic stream variables that set the rate, what the measured rate was,
and the number of packets sent by the traffic stream.For the selection options, see show
Displaying Traffic Stream and Summary Information (page 2-38).
If the keyword summary is included, only the interface totals are displayed.
The display can post one of the following messages to explain what the rates are based on:
The rates are since traffic generation was started.

The rates are since the last rate change during traffic generation.
Traffic generation is currently off.

These rates are from the last time traffic generation was active.

The following example shows output during traffic generation. Note that the rate or interval is
measured by taking the time at which traffic generation is started and when it is stopped, and the
number of packets sent during that period. This means that the measurement can be badly off when
the time between packets is long, and the transmit period is short (for example the interval
measurement here).
c4700-p(TGN:OFF,Et1:5/5)#sh rate

Traffic generation is currently off.

These rates are from the last time traffic generation was active.

Summary of traffic stream rates on Ethernet1

measured
ts# template state repeat interval/rate interval/rate packets_sent
1 IP on 1 10 pps 9.923 186
2 IP on 1 100 pps 99.934 1864
3 TCP on 1 1234 pps 1233.869 23003
4 UDP on 1 500 pps 499.941 9321
5 ICMP on 1 2000 millisec 2071 msec 10
6 IP on 1 1200 bps 1199.372 613
Totals for Ethernet1 1864.420 34997

c4700-p(TGN:OFF,Et1:5/5)#sh rate sum

Traffic generation is currently off.

These rates are from the last time traffic generation was active.

Summary of traffic stream rates on interfaces

measured

Cisco Company Confidential Command Reference 2-55

show Displaying Traffic Stream and Summary Information

interface interval/rate packets_sent

Ethernet1 1864.420 34997

rate Setting the Packet Send Rate (page 2-35)

bit-rate - Setting the transmission Rate in bits per second (page 2-15)
interval Setting the Interval Between Sending Packets (page 2-27)
clear Clearing Configurations or Counters (page 2-16)

show secondary Displaying Activity Status of SECONDARY Processors

show secondary
Displays which SECONDARY processors will transmit packets and which SECONDARY
processors will have the PRIMARY processor transmit packets for them.This command is only
available if there are SECONDARY processors.
For example:
c7513a-(TGN:OFF,Et0/0/0:2/2)#sh secondary

on = SECONDARY to transmit
off = PRIMARY to transmit
SECONDARY-Transmit-Active
SECONDARY slot 0 on
SECONDARY slot 1 on
SECONDARY slot 2 on
SECONDARY slot 3 on
SECONDARY slot 5 on

The SECONDARY processors that are set to ON will transmit packets out their active interfaces.
The SECONDARY processes that are set to OFF will let the PRIMARY transmit the packets through
the SECONDARY for the SECONDARYs active interfaces.
secondary Selecting a SECONDARY Processor for Transmission (page 2-37)

show send Displaying Summary of Send Process

show send [selection-options]
Displays a summary of how many packets have been sent by the start send process. For the
selection options, see show Displaying Traffic Stream and Summary Information (page 2-38).
If this command is used while sending packets in the process or fast mode, the left-to-send
column shows how many packets are left to send on a traffic stream before the send process is
complete.
For example:
c4700-p(TGN:OFF,Et1:3/4)#sh send

Summary of sending traffic streams on Ethernet1

ts# template state interval/rate send-amount/left-to-send total-sent
1 IP on 5000 pps 10000 0 0
2 TCP on 1000 pps 2000 0 0
3 UDP on 12000 pps 24000 0 0
4 ARP on 2000 millisec 2 0 0

c4700-p(TGN:OFF,Et1:3/4)#start send
c4700-p(TGN:SEND,Et1:3/4)#

You have started packet send in dedicated output-mode.

2-56 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 show sequence Displaying Summary of Packet Sequences

TGN will go into a send loop that locks out all other process.
Enter shift-control-6 to stop traffic generation or wait for completion.

Send process complete.

c4700-p(TGN:OFF,Et1:3/4)#sh send

Summary of sending traffic streams on Ethernet1

ts# template state interval/rate send-amount/left-to-send total-sent
1 IP on 5000 pps 10000 0 10000
2 TCP on 1000 pps 2000 0 2000
3 UDP on 12000 pps 24000 0 24000
4 ARP on 2000 millisec 2 0 2

c4700-p(TGN:OFF,Et1:3/4)#sh rate

Summary of traffic stream rates on Ethernet1

measured
ts# template state repeat interval/rate interval/rate packets_sent
1 IP on 1 5000 pps 5002.814 10000
2 TCP on 1 1000 pps 1000.375 2000
3 UDP on 1 12000 pps 9452.075 24000
4 ARP on 1 2000 millisec 2000 msec 2

send Sending Packets (page 2-38)

start/stop Starting and Stopping Traffic Generation (page 2-59)

show sequence Displaying Summary of Packet Sequences

show sequence [selection-options]
Displays a summary of special packet sequence traffic streams.
For example:
c4700-p(TGN:OFF,Et1:1/4)#sh seq

Summary of packet sequences on Ethernet1

ts# total enabled configured name
1 3 3 3
c4700-p(TGN:OFF,Et1:1/4)#sh
Traffic stream 1 of 4, Sequence, Ethernet1 (up)
name ""
on
sequence interval 100
sequence add "frag 1"
sequence 1 enable
sequence add "frag 2"
sequence 2 enable
sequence add "frag 3"
sequence 3 enable
! 3 out of 3 packets enabled

c4700-p(TGN:OFF,Et1:1/4)#sh name

Summary of traffic stream names on Ethernet1

ts# template state delayed-start name
1 Sequence
2 IP on random frag 1
3 IP on random frag 2
4 IP on random frag 3

Cisco Company Confidential Command Reference 2-57

show Displaying Traffic Stream and Summary Information

add Adding a Traffic Stream (page 2-10)

sequence Adding and Updating Packet Sequences (page 2-38)

show tcp Displaying TCP Header Information

Displays a summary of TCP header field configurations. This command can also displays header
field configurations for flow members. For the selection options, see show Displaying Traffic
Stream and Summary Information (page 2-38).
For example:
k4700-p(TGN:OFF,Et0:34/50)#sh tcp

Summary of TCP traffic streams on Ethernet0

src dest header option
ts# port port seq ack length flgs win chksm urgent length
1 0 0 00000000 00000000 80 0x00 0 0xAFE5 0 0
2 0 0 00000000 00000000 80 0x00 0 0xAFE5 0 0
3 0 0 00000000 00000000 80 0x00 0 0xAFE5 0 0
10 0 0 00000000 00000000 80 0x00 0 0xAFE5 0 0
11 0 0 00000000 00000000 80 0x00 0 0xAFE5 0 0
12 0 0 00000000 00000000 80 0x00 0 0xAFE5 0 0

TCP Transport Header Field Update Commands (page 2-7)

Example of TCP Traffic Stream (page A-13)
Explanation of TCP Header Fields (page B-3)

show traffic-stream Displaying a Traffic Stream by Name or Number

show traffic-stream ts-name-or-number
Displays the configuration of a traffic stream, allowing it to be selected by name or number.
This is similar to the command show ts#, except this allows the traffic stream to be selected by
name.
The following example displays the output of an IP traffic stream on Ethernet:
k4700-p(TGN:OFF,Et0:4/4)#show traffic-stream 4

Traffic stream 4 of 4, IP, Ethernet0 (up)

name ""
on
rate 10
variability 0
send 0
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
!
datalink user-defined
length 1000
fragmentation enable mtu auto
fragmentation option-length 0
!
L2-encapsulation arpa
L2-dest-addr 0000.0000.0000
L2-src-addr 0060.3E58.1E1A
L2-protocol 0x0800
!

2-58 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 show udp Displaying UDP Header Information

L3-version 4
L3-header-length auto
L3-tos 0x00
L3-length auto
L3-id 0x0000
L3-fragmentation 0x0000
L3-ttl 60
L3-protocol 0
L3-checksum auto
L3-src-addr 0.0.0.0
L3-dest-addr 0.0.0.0
L3-option-length 0
!
data-length 0
!
fill-pattern 0x00 0x01

show udp Displaying UDP Header Information

show udp [selection-options]
Displays a summary of UDP header field configurations. For the selection options, see show
Displaying Traffic Stream and Summary Information (page 2-38).
For example:
tools75(TGN:OFF,Et2/2:1/54)#sh udp
Summary of UDP traffic streams on Ethernet2/2
ts# src.port dest.port length checksum
22 0 0 8 0x0000
23 0 0 8 0x0000
24 0 0 8 0x0000

UDP Transport Header Field Update Commands (page 2-7)

Example of UDP Traffic Stream (page A-14)
Explanation of UDP Header Fields (page B-4)

sre Defining Traffic Streams for TGN or SRE

sre {on | off}
Defines packets for SRE use. off turns off SRE mode and goes to TGN mode. Traffic streams are
then defined for TGN program use. on sets SRE mode, and packets are defined for SRE program
use.

start/stop Starting and Stopping Traffic Generation

start [send]
stop
s
These commands start and stop traffic generation. When traffic generation is started or activated,
the command prompt will indicate ON and all active traffic streams on all interfaces will transmit
packets. The s command is a quick way to toggle the ON/OFF state of traffic generation.
These commands do not affect arp-responder or hello-generator traffic streams. As long as an
arp-responder is defined and active, it will respond to ARP requests. As long as a hello-generator
is defined and active, it will send out hello packets.

Cisco Company Confidential Command Reference 2-59

variability Defining the Variability in Packet Intervals

The start command has two modes.

start
Starts traffic generation. Traffic does not stop until the stop command (or cntrl-shift-6 for
dedicated mode) is entered.
start send
Causes only traffic streams that have a send count entered to send packets until the requested
number of packets are sent for the traffic stream. When all traffic streams have sent their packets,
the program posts the message Send Complete and stops traffic generation.
This send mode can also be aborted with the stop command.

variability Defining the Variability in Packet Intervals

variability percent
Defines the variability, from 0 to 100 percent, in the time interval between packets.
By default, percent is 0, that is, there is a constant time interval between each packet sent by a
traffic stream.
If percent is not zero, it defines a range of interval values, plus and minus from the interval
defined by the rate or interval or bit-rate command. A new and different interval is calculated
within this range after each packet.
For example, assume a 10 pps traffic stream:
If variability is 0 percent, the interval between packets is 100 milliseconds.
If variability is 50 percent, intervals are random, from 50 to 150 milliseconds.
At maximum variability of 100 percent, the intervals are random from 0 to 200 milliseconds.
In all cases, the average rate over a period of time is equal to the rate set by the rate or interval or
bit-rate command, even though individual intervals would vary.
This does not affect the interval between packets sent in a repeat, which are still sent as fast as
possible. Instead, it affects the time interval between the repeat sends (see repeat Resending
Packets Repeatedly (page 2-35)).
rate Setting the Packet Send Rate (page 2-35)
bit-rate - Setting the transmission Rate in bits per second (page 2-15)

verbose - Configuring for Activity Messages

verbose {off | on}
verbose logging-to {ios-logging | console}
These commands are global across all processes.
verbose {off | on}
Determines whether the start and stop commands generate activity messages. The default is off.
verbose logging-to {ios-logging | console}
Determines where the activity messages will be posted. The default is to display the messages on
the console.

2-60 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 wait-to-release Sierra Wait-to-Release Paktype

If this is set for ios-logging, the activity messages are passed to the IOS logging facility as level
6 (informational) messages. Depending on how IOS logging is configured, these messages can
have timestamps added and be posted to the console and/or log file and/or system log. For more
information on how to configure IOS logging, see the logging commands in the Cisco IOS
Configuration Fundamentals Command Reference -> Cisco IOS System Management
Commands -> Troubleshooting Commands.

wait-to-release Sierra Wait-to-Release Paktype

wait-to-release n
This command is only available on the sierra routers (4500 and 4700). It helps get around a
problem of releasing paktypes at fast output rates.
By default, this value is set to 0, that is, in fast and dedicated output modes, paktypes are released
immediately. In some situations, at fast output rates, paktypes might not get released and might
result in a slow memory leak.
If the memory leak is a problem, set a wait-to-release time of n seconds. This causes a wait period
of n seconds after a stop command before the paktypes are forced to be released by setting their
refcount. If you set wait time too short, IOS complains with incorrect refcount error messages,
so you will have to increase the wait time.
In most cases, a wait time of 1 second is enough. In some situations, 10 seconds might not be
enough.
The wait period is reflected in the command prompt options. The program prevents restarting
traffic generation until the wait period is over.

write Writing Information to an IFS Log File

write ....
Commands that start with the keyword show are used to display information on the console. For
every show command, there is an equivalent write command that displays the same information
on the console but also writes it to an opened IFS log file.
IOS File System (page 1-10)
show Displaying Traffic Stream and Summary Information (page 2-38)
open-logfile Opening an IFS Log File (page 2-31)
close-logfile Closing an Open IFS Log File (page 2-17)

Cisco Company Confidential Command Reference 2-61

write Writing Information to an IFS Log File

2-62 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 CHAP TER 3

Defining Header Field Values

Almost all header fields have decimal, hex, mac address, or IP address field data. These can all be
defined as having a constant value, incrementing or being random over a range.
A few fields are defined by entering a string of hex numbers. These fields cannot be incrementing
or random, although it is possible to lay a configurable field on top of these fields to add variability.
These fields include the Token Ring Routing Information Field (RIF), Connectionless Network
Service (CLNS) source and destination areas, IP option field, and TCP option field.
Checksum and header length fields also have the additional ability of being defined automatically to
accommodate changing packet lengths and data.

Decimal and Hex Fields

Most header fields are decimal or hexadecimal numbers, 1 to 4 bytes in length. There are three
command formats to define these fields.
In the following, Ln.... represents an L2-... command to define a datalink header field, an
L3-...command to define a network header field, or an L4-... command to define a transport header
field.
All of these values can be entered as decimal or hex. Hex numbers must start with 0x.
Ln-... value
Sets the field to a constant value.
The generic prompt for these numbers always prompts for a number that is valid for a 4 byte field.
If the number entered is too large, an error message indicates the correct maximum value.
For example:
tools75(TGN:OFF,Et0/0/0,1/3)#l3-ttl 1000
*** The value exceeds field size of 0xFF (255).

Ln-... increment min-value to max-value [by inc-by]

Causes the field value to increment from min-value to max-value with each packet sent. By
default, it increments by 1 byte to max-value and restarts at min-value. The inc-by option allows
the field value to increase by the inc-by] byte amount instead of by 1.
Ln-... random min-value to max-value [by inc-by]
Causes the field value to be random from min-value to max-value. By default, the random value
can be from min-value to max-value. The inc-by option causes the field value to be min-value
plus multiples of inc-by, instead of multiples of 1.

Cisco Company Confidential Defining Header Field Values 3-1

MAC Address Fields

The following are examples of setting hex and decimal fields:

L3-protocol 10
L3-ttl increment 20 to 50
L3-tos increment 20 to 100 by 5
L4-seq random 0x10000 to 0x20000
L4-ack random 0x10000 to 0x20000 by 0x100
Nested Increments (page 3-3)

MAC Address Fields

Ln-... value
Ln-... increment min-value to max-value [by inc-by]
Ln-... random min-value to max-value [by inc-by]
Fields with MAC addresses are entered just like the decimal and hex values defined in Decimal and
Hex Fields (page 3-1), except that all values are entered as full MAC addresses, except for the inc-by,
which is entered as max 4 byte decimal or hex value.
The increment and random options only affect the last 4 bytes of these 6 byte fields. If you need to
have the first 2 bytes changing, lay a configurable field on top of these bytes.
The following are examples of setting MAC address fields:
L2-dest-addr 0000.0c98.a44c
L2-src-addr increment 00ab.0004.0000 to 00ab.0004.1000
L2-src-addr increment 0000.0000.abcd to 0000.0100.abcd by 0x10000
L3-dest-host random 0011.2222.0000 to 0011.2222.3000
L3-src-host random 0000.1111.0000 to 0000.1111.5500 by 0x100
Nested Increments (page 3-3)

IP Address Fields
Ln-... value
Ln-... increment min-value to max-value [by inc-by] [subnet subnet-mask]
Ln-... random min-value to max-value [by inc-by] [subnet subnet-mask]
Fields with IP addresses are entered just like the decimal and hex values defined in Decimal and Hex
Fields (page 3-1), except that all values are entered as full IP addresses, including the inc-by value.
It can include a subnet mask, so subnet broadcast addresses are not generated.
The following are examples of setting IP address fields:
L3-src-addr 100.1.1.1
L3-dest-addr increment 1.1.1.1 to 1.1.1.255
L3-dest-addr increment 1.1.1.1 to 1.1.255.1 by 0.0.1.0
L3-dest-addr random 192.1.1.1 to 192.1.255.255
L3-src-addr random 192.1.1.1 to 192.1.1.255 by 0.0.0.5

3-2 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 IPv6 Address Fields

If you need to generate a wide range of IP addresses that span several subnets, you probably do not
want to send subnet broadcast packets. If you configure a subnet mask for incrementing and random
IP addresses, the TGN program will not generate addresses where the subnet address bits are all
zeros or all ones.
The following is a repeat of the above examples with subnet masks added:
L3-dest-addr increment 1.1.1.1 to 1.1.1.255 subnet 255.255.255.0
L3-dest-addr increment 1.1.1.1 to 1.1.255.1 by 0.0.1.0 subnet 255.255.128.0
L3-dest-addr random 192.1.1.1 to 192.1.255.255 subnet 255.255.255.240
L3-src-addr random 192.1.1.1 to 192.1.1.255 by 0.0.0.5 subnet 255.255.255.192
Nested Increments (page 3-3)

IPv6 Address Fields

Ln-... value
IPv6 address do not directly support incrementing or random values.
For reasons of performance, incrementing and random fields are limited to a 32-bit CPU word, 4
bytes. Four bytes is big enough to hold an entire IPv4 address or the last 4 bytes of a MAC address,
but an IPv6 address is 16 bytes long. This problem is solved by using the configurable field option.
With this you can lay an incrementing or random field up to 4 bytes long on the appropriate bytes of
an IPv6 address.
Example 1:
To increment the last 4 bytes of the IPv6 source address in an IPv6 header, from 1 to 1000:
field add incr_src_ipv6
field start-at network-start offset 20
field type 4 decimal
field data increment 1 to 1000

Note that the incrementing or random field is identified by the number of bytes from the start of
the header. In the IPv6 header, the source address starts 8 from the start of the header.
The last four bytes of the 16-byte IPv6 source address (bytes 13 to 16) start 8 + 12 bytes from the
start of the network header.
Example 2:
To make the seventh and eighth bytes of the destination IPv6 address random from 0 to 0xFFFF:
field add rand_dest_ipv6
field start-at network-start offset 30
field type 2 hex
field data random 0 to 0xFFFF

In the IPv6 header, the destination address starts 24 bytes from the start of the header, so the
seventh byte in the address starts 24 + 6 = 30 bytes from the start of the header.

Nested Increments
Any header field that can be set to increment over a range also has the ability to be linked to another
incrementing field. This is the nest-over option and looks like this:
... increment min to max [nest-over field-name]

Cisco Company Confidential Defining Header Field Values 3-3

Automatic Setting of Length and Checksum Fields

With this option, the incrementing field increments only when the incrementing field it points to
wraps from its max to min value. This allows the traffic stream to step through all possible
combinations of the two fields.
The field-name can be any L2-, L3-, L4-, or configurable field name, but it must adhere to the
following rules:
Entered correctly in full
Valid within the template
Set to increment before traffic generation or expand
Nested incrementing fields can nest several deep, and several incrementing fields can nest to a single
incrementing field.
An incrementing packet length can be set to nest-over an incrementing header field, and
incrementing header fields can be set to nest-over length. This is an exception, because packet
length is not a header field.
In this example, every time the IP destination address increments through its range of addresses, the
IP source address increments by 1.
L3-src-addr increment 22.1.1.1 to 22.1.1.10 nest-over L3-dest-addr
L3-dest-addr increment 100.1.1.1 to 100.1.1.100

Automatic Setting of Length and Checksum Fields

Most header length and several header checksum fields have the additional ability to be set
automatically to a correct value.
When these fields are initially defined, these fields default to auto. This allows packets to be defined
and updated with incrementing and random values or with incrementing or random lengths. The
length and checksum fields are updated automatically to the correct value.
The same is true for checksums in the IP, TCP, UDP, ICMP, and IGMP headers.
The calculated length values and checksum are what is required for a router to accept and forward a
packet.
These length and checksum fields can be set to user-defined constant, incrementing, or random
values, but the only reason to do so is to introduce invalid data into a packet for negative testing.

Token Ring RIF

The Token Ring RIF field is unlike other header fields. It may or may not exist in the datalink header,
and if it exists, it is of variable length up to 32 bytes long.
The default RIF field length is 0. This field is entered as a series of hex bytes. A valid RIF field has
an even number of bytes, but TGN allows a field length of an odd number of bytes for testing
purposes.
In the second byte of the RIF field is a field that indicates the RIF field length. The program forces
the correct RIF length into this field. An incorrect length trashes the rest of the packet.
The RIF field cannot be set to be incrementing or random. If you need variability in this field, lay an
configurable field on top of the RIF field.
The following is an example of a RIF field entry:
L2-rif C006 0021 0030

3-4 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 CLNS Area Fields

CLNS Area Fields

In the CLNS header, the source and destination area fields are variable length fields up to 13 bytes
long. These fields are entered as a series of hex bytes. A valid area is an odd number of bytes long.
The CLNS area fields cannot be set to be incrementing or random. If you need variability in these
fields, lay a configurable field on top.
The following are examples of CLNS area fields:
L3-dest-area 47.0010.aa01.00cc
L3-src-area 47.0102.0304.0506.0708.0910.1112

Cisco Company Confidential Defining Header Field Values 3-5

CLNS Area Fields

3-6 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 CHAP TER 4

ARP-Responder and Hello-Generator

Commands
The ARP-responder and hello-generator commands configure the ARP responders and
hello-generators.
The purpose of these responders and generators is to make the router think there is a live end station
waiting to receive packets to the specific network address. The hello-generators tell the router which
interface the end station is on, and all of these tell the router the MAC address of the end station.

IP ARP Responder
The IP ARP Responder responds to ARP requests for one IP address or a range of them.
The ARP Responder is typically used to act as a fictitious end station. If you are using a traffic
generator (like TGN) to send packets to a non-existent end station or IP address, you need the ARP
responder to act as the end station so that the router forwards the test packets on the destination
network.
If the ARP responder is not used and the end-station IP address has not been statically configured in
the ARP cache, the router sends out an ARP request for each test packet. After timeout, it sends an
ICMP host unreachable message back to the source IP address.
With the ARP responder, the first test packet results in an ARP request. The ARP responder sends a
reply giving a MAC address to which the router can forward packets. With the destination address
now in the routers ARP cache, all subsequent packets to that IP address are forwarded immediately.
To create the ARP Responder, first use the add arp responder command (see add Adding a Traffic
Stream (page 2-10)). Similar to regular traffic streams, multiple ARP responders can be created on
each interface.
Use one of the following commands to define which IP address or range of IP addresses the ARP
responder will respond to:
ip-address a.b.c.d
ip-address a.b.c.d to a.b.c.d
The following is the fictitious MAC address in the ARP response. It can be any MAC address. It is
not related to the bia address of the interface (though in most cases, you configure it to be the MAC
address of the interface). If the ARP responder is to respond to a range of IP addresses, you can
configure it to respond with a constant MAC address for all the IP addresses in the range or with an
equal range of MAC addresses starting with this configured MAC address:
mac-address hhhh.hhhh.hhhh [constant]
The following activates or deactivates the ARP responder:
on | off

Cisco Company Confidential ARP-Responder and Hello-Generator Commands 4-1

AppleTalk ARP Responder

Example of IP ARP Responder (page A-21)

show arp-responder Displaying ARP Responders (page 2-43)

AppleTalk ARP Responder

To create the Appletalk ARP (AARP) Responder, first use the add aarp responder command (see
add Adding a Traffic Stream (page 2-10)).
Use one of the following commands to define which AppleTalk address or range of AppleTalk
addresses the AARP responder will respond to:
appletalk-address net.node
appletalk-address net.node to net.node
The following is the fictitious MAC address in the AARP response. If the AARP responder is to
respond to a range of AppleTalk addresses, it responds with an equal range of MAC addresses
starting with this configured MAC address:
mac-address hhhh.hhhh.hhhh
The following activates or deactivates the AARP responder:
on | off
Example of AARP (AppleTalk ARP) Responder (page A-21)
show aarp-responder Displaying AARP Responders (page 2-41)

VINES Hello-Generator
To create the VINES hello-generator, first use the add vines hello-generator command (see add
Adding a Traffic Stream (page 2-10)). The following commands configure the VINES
hello-generator.
hello-address network:subnet
VINES address the hello-generator advertises as an active end station.
mac-address hhhh.hhhh.hhhh
Fictitious MAC address the router is to send the packets to.
on | off
Activates or deactivates the VINES hello-generator.
Example of VINES Hello-Generator (page A-21)
sre Defining Traffic Streams for TGN or SRE (page 2-59)

CLNS Hello-Generator
To create the CLNS hello-generator, first use the add clns hello-generator command (see add
Adding a Traffic Stream (page 2-10)). The following commands configure the CLNS
hello-generator.
hello-address 47.0000.0000.0000.0011.2222.3333.00
CLNS address the hello-generator advertises as an active end station. It shows the command with
an example CLNS address. The MAC address is extracted from the CLNS address. In this
example, the MAC address is 0011.2222.3333.
on | off

4-2 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 DECnet Hello-Generator

Activates or deactivates the CLNS hello-generator.

Example of CLNS Hello-Generator (page A-21)
show clns-hello-generator Displaying CLNS Hello-generators (page 2-45)

DECnet Hello-Generator
To create the CLNS hello-generator, first use the add decnet hello-generator command (see add
Adding a Traffic Stream (page 2-10)). The following commands configure the DECnet
hello-generator.
hello-address area.node
DECnet address the hello-generator advertises as an active end station. In the DECnet protocol,
the area node address defines the MAC address.
designated-router area.node
The DECnet end station hello packet includes a field that identifies the designated router the end
station sends its packets to.
on | off
Activates or deactivates the DECnet hello-generator.
Example of DECnet Hello-Generator (page A-21)
show decnet-hello Displaying DECnet Hello-Generators (page 2-46)

Cisco Company Confidential ARP-Responder and Hello-Generator Commands 4-3

DECnet Hello-Generator

4-4 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 CHAP TER 5

Using TGN and PKTS Timestamps to

Measure Latency
The timestamp configurable field is included in TGN and PKTS to help measure latency through a
network.
When the timestamp field is configured on a TGN traffic stream, TGN puts a timestamp in the packet
just before sending it. This occurs before any transport checksums are calculated, so that the
timestamp can be added into a valid TCP or UDP packet. Turn off transport checksumming if the
checksum is not important to the test.
When the timestamp field is configured in a PKTS filter, PKTS can extract the timestamp from the
packet and compare it to the time it received the packet. The show timestamp command can be used
to display the time difference.
It is important that the same router is used to send and receive, so that both the send and receive
timestamps come from the same timebase.

Creating a TGN Traffic Stream with a Timestamp

The easiest way to add a timestamp configurable field to a TGN traffic stream is to add the
timestamp keyword when the traffic stream is created. This puts the 8 byte timestamp in the first 8
bytes of the data array.
For example:
add ip timestamp

Creating a PKTS Selective Filter with a Timestamp

The easiest way to add a timestamp configurable field to a PKTS filter is to add the timestamp
keyword when the filter is created. This defines the timestamp to be in the first 8 bytes of the data
array, and it sets this to be a display filter of incoming packets.
In PKTS FILTER mode, do the following:
add ip timestamp

Displaying Timestamp Information

To display timestamp information, use the show timestamp command:
show timestamp [command-line-filter-options] [stats-only]
This command only display packets that match a display filter with a configurable timestamp
field. All the usual command line filter options are available to select a subset of the packets.

Cisco Company Confidential Using TGN and PKTS Timestamps to Measure Latency 5-1
Example of Using Timestamps

Statistics are displayed at the end of the summary. If you want to display only the statistics and
not all the packets, include the stats-only keyword.
For example:
p4700a-(PKTS:1 of 21739)#sh tim from 1 to 15
# interface summary relative time time-diff
1 Et2 IP 0.000990 0.000354
2 Et2 IP 0.001211 0.000476
3 Et2 IP 0.001320 0.000479
4 Et2 IP 0.001419 0.000535
5 Et2 IP 0.001869 0.000321
6 Et2 IP 0.002587 0.000339
7 Et2 IP 0.002793 0.000374
8 Et2 IP 0.002904 0.000433
9 Et2 IP 0.003006 0.000496
10 Et2 IP 0.003819 0.000355
11 Et2 IP 0.004042 0.000502
12 Et2 IP 0.004152 0.000542
13 Et2 IP 0.004255 0.000586
14 Et2 IP 0.004356 0.000628
15 Et2 IP 0.005198 0.000333
Packets with timestamps = 15
Minimum time difference = 0.000321
Maximum time difference = 0.000628
Average time difference = 0.000450
Median time difference = 0.000476
Standard Deviation = 0.000291

The columns contains the following information.

#The number of the packet in the capture buffer.
interfaceThe interface the packet came in on.
summarySummary description of the packet.
relative timeTime the packet was received relative to when capture was started to a selected
packet.
time-diffDifference between the timestamp in the packet and the receive timestamp.

Example of Using Timestamps

Assume we need to create three separate traffic streams of 60 byte IP packets with three different
Type-Of-Service values. For ease of identifying each traffic stream, we give each traffic stream a
different source IP address. We will send the packets out interface Ethernet1, through a router, and
received on Ethernet2.
In this test, two routers are used.
----------------- -----------------
| | | |
| | | |
| eth1|----->-----|eth1 |
| | | |
| | | |
| p4700-pagent | | r4700-uut |
| | | |
| | | |
| eth2|-----<-----|eth2 |
| | | |
| | | |
----------------- -----------------

5-2 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Example of Using Timestamps

Router p4700-pagent is used to run TGN and PKTS:

1 TGN will send packets out interface ethernet1.

2 PKTS will capture packets on interface ethernet2.

Router r4700-uut is used to run the IOS image under test.
The two routers are connected back to back, ethernet1 to ethernet1, ethernet2 to ethernet2.
We enter the following configurations on the two routers and make sure all four interfaces are not
shut.
On the r4700-uut, this configures the router we are sending packets through:
interface eth1
ip address 1.1.1.10 255.255.255.0
mac-address 0000.1111.1111
no shut
interface eth2
ip address 1.1.2.10 255.255.255.0
no shut
router igrp 2
network 1.0.0.0

On the router running Pagent, the following configuration makes sure the two interfaces we are using
are not shut. It also sets a MAC address on the receiving interface.
interface eth1
not shut
interface eth2
mac-address 0000.2222.2222
not shut

The following commands are entered into the p4700-pagent router, the router running the TGN and
PKTS programs. These command create three IP traffic streams. The add commands include the
timestamp keyword, so that a timestamp field is put into the first 8 bytes of the data array. Each
traffic stream is set to a different type of service and a different source IP address. The source IP
address is used to help identify each stream on the receive side.
tgn
ethernet1
add ip timestamp
l3-src-addr 1.1.1.30
l3-tos 0x10
add ip timestamp
l3-src-addr 1.1.1.31
l3-tos 0x20
add ip timestamp
l3-src-addr 1.1.1.32
l3-tos 0x40

These commands are applied to all three traffic streams.

all length 60
all rate 1000
all l3-dest-addr 1.1.2.20
all l2-dest-addr 0000.1111.1111
all l2-src-addr 0000.3333.3333

Cisco Company Confidential Using TGN and PKTS Timestamps to Measure Latency 5-3
Example of Using Timestamps

This creates an ARP responder on the receiving interface to respond to the routers ARP requests.
ethernet2
add arp responder
ip-address 1.1.2.20
mac-address 0000.2222.2222

The next commands configure PKTS. These commands:

create a 5,000,000 byte capture buffer
activate capture of incoming packets on ethernet2
create a selective filter on ethernet2 to capture only IP packets to 1.1.2.20 (by default, filters are
set for selective capture)
create a display filter to display IP packets with timestamps (by default, timestamp filters are set
for selective display).
pkts
add 5000000
et2 in
et2 filter ip ip-dest 1.1.2.20
et2 filter ip timestamp

At this point, start TGN traffic generation, start PKTS capture, wait until the capture buffer is full,
then stop TGN traffic generation.
The following are examples of using show timestamp to view the differences between the transmit
and receive timestamps.
First, a quick overview of the captured packets:
p4700a-(PKTS:1 of 21739)#sh all
# TD interface summary relative time length
1 I Et2 IP 0.000990 60
2 I Et2 IP 0.001211 60
3 I Et2 IP 0.001320 60
4 I Et2 IP 0.001419 60
5 I Et2 IP 0.001869 60
6 I Et2 IP 0.002587 60
7 I Et2 IP 0.002793 60
8 I Et2 IP 0.002904 60
9 I Et2 IP 0.003006 60
10 I Et2 IP 0.003819 60
11 I Et2 IP 0.004042 60
12 I Et2 IP 0.004152 60
13 I Et2 IP 0.004255 60
14 I Et2 IP 0.004356 60
15 I Et2 IP 0.005198 60
16 I Et2 IP 0.005392 60
17 I Et2 IP 0.005500 60
18 I Et2 IP 0.005602 60
19 I Et2 IP 0.005705 60
20 I Et2 IP 0.006548 60
21 I Et2 IP 0.006747 60
--More--

5-4 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Example of Using Timestamps

Then, a quick overview of the IP addresses of the captured packets:

p4700a-(PKTS:1 of 21739)#sh ip
# TD interface destination source summary
1 I Et2 1.1.2.20 1.1.1.30 IP
2 I Et2 1.1.2.20 1.1.1.31 IP
3 I Et2 1.1.2.20 1.1.1.32 IP
4 I Et2 1.1.2.20 1.1.1.30 IP
5 I Et2 1.1.2.20 1.1.1.31 IP
6 I Et2 1.1.2.20 1.1.1.32 IP
7 I Et2 1.1.2.20 1.1.1.30 IP
8 I Et2 1.1.2.20 1.1.1.31 IP
9 I Et2 1.1.2.20 1.1.1.32 IP
10 I Et2 1.1.2.20 1.1.1.30 IP
11 I Et2 1.1.2.20 1.1.1.31 IP
12 I Et2 1.1.2.20 1.1.1.32 IP
13 I Et2 1.1.2.20 1.1.1.30 IP
14 I Et2 1.1.2.20 1.1.1.31 IP
15 I Et2 1.1.2.20 1.1.1.32 IP
16 I Et2 1.1.2.20 1.1.1.30 IP
17 I Et2 1.1.2.20 1.1.1.31 IP
18 I Et2 1.1.2.20 1.1.1.32 IP
19 I Et2 1.1.2.20 1.1.1.30 IP
20 I Et2 1.1.2.20 1.1.1.31 IP
21 I Et2 1.1.2.20 1.1.1.32 IP
--More--

A quick overview of timestamps:

p4700a-(PKTS:1 of 21739)#sh time
# interface summary relative time time-diff
1 Et2 IP 0.000990 0.000354
2 Et2 IP 0.001211 0.000476
3 Et2 IP 0.001320 0.000479
4 Et2 IP 0.001419 0.000535
5 Et2 IP 0.001869 0.000321
6 Et2 IP 0.002587 0.000339
7 Et2 IP 0.002793 0.000374
8 Et2 IP 0.002904 0.000433
9 Et2 IP 0.003006 0.000496
10 Et2 IP 0.003819 0.000355
11 Et2 IP 0.004042 0.000502
12 Et2 IP 0.004152 0.000542
13 Et2 IP 0.004255 0.000586
14 Et2 IP 0.004356 0.000628
15 Et2 IP 0.005198 0.000333
16 Et2 IP 0.005392 0.000455
17 Et2 IP 0.005500 0.000494
18 Et2 IP 0.005602 0.000532
19 Et2 IP 0.005705 0.000574
20 Et2 IP 0.006548 0.000337
21 Et2 IP 0.006747 0.000467
--More--

Cisco Company Confidential Using TGN and PKTS Timestamps to Measure Latency 5-5
Example of Using Timestamps

An overview of timestamp information for one section of the capture buffer so we also get statistics:
p4700a-(PKTS:1 of 21739)#sh tim from 1 to 15
# interface summary relative time time-diff
1 Et2 IP 0.000990 0.000354
2 Et2 IP 0.001211 0.000476
3 Et2 IP 0.001320 0.000479
4 Et2 IP 0.001419 0.000535
5 Et2 IP 0.001869 0.000321
6 Et2 IP 0.002587 0.000339
7 Et2 IP 0.002793 0.000374
8 Et2 IP 0.002904 0.000433
9 Et2 IP 0.003006 0.000496
10 Et2 IP 0.003819 0.000355
11 Et2 IP 0.004042 0.000502
12 Et2 IP 0.004152 0.000542
13 Et2 IP 0.004255 0.000586
14 Et2 IP 0.004356 0.000628
15 Et2 IP 0.005198 0.000333
Packets with timestamps = 15
Minimum time difference = 0.000321
Maximum time difference = 0.000628
Average time difference = 0.000450
Median time difference = 0.000476
Standard Deviation = 0.000291

Timestamp information for one selected stream from a section of the capture buffer:
p4700a-(PKTS:1 of 21739)#sh tim ip-src 1.1.1.30 from 1 to 40
# interface summary relative time time-diff
1 Et2 IP 0.000990 0.000354
4 Et2 IP 0.001419 0.000535
7 Et2 IP 0.002793 0.000374
10 Et2 IP 0.003819 0.000355
13 Et2 IP 0.004255 0.000586
16 Et2 IP 0.005392 0.000455
19 Et2 IP 0.005705 0.000574
22 Et2 IP 0.006854 0.000505
25 Et2 IP 0.007927 0.000364
28 Et2 IP 0.008348 0.000578
31 Et2 IP 0.009485 0.000441
34 Et2 IP 0.010249 0.000378
37 Et2 IP 0.011321 0.000494
40 Et2 IP 0.012674 0.000331
Packets with timestamps = 14
Minimum time difference = 0.000331
Maximum time difference = 0.000586
Average time difference = 0.000451
Median time difference = 0.000455
Standard Deviation = 0.000288

5-6 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Example of Using Timestamps

Timestamp information for a different stream and a different section of the capture buffer:
p4700a-(PKTS:1 of 21739)#sh tim ip ip-src 1.1.1.31 from 1000 to 1040
# interface summary relative time time-diff
1001 Et2 IP 0.288738 0.000582
1004 Et2 IP 0.289848 0.000488
1007 Et2 IP 0.290867 0.000333
1010 Et2 IP 0.292413 0.000457
1013 Et2 IP 0.292732 0.000583
1016 Et2 IP 0.293849 0.000493
1019 Et2 IP 0.294871 0.000329
1022 Et2 IP 0.296428 0.000466
1025 Et2 IP 0.296746 0.000593
1028 Et2 IP 0.298178 0.000450
1031 Et2 IP 0.298499 0.000573
1034 Et2 IP 0.300411 0.000446
1037 Et2 IP 0.301059 0.000899
1040 Et2 IP 0.302057 0.000450
Packets with timestamps = 14
Minimum time difference = 0.000329
Maximum time difference = 0.000899
Average time difference = 0.000510
Median time difference = 0.000488
Standard Deviation = 0.000293

This uses the stats-only keyword to not display all the individual timestamps:
p4700a-(PKTS:1 of 21739)#sh tim ip ip-src 1.1.1.32 stats-only
Packets with timestamps = 7246
Minimum time difference = 0.000305
Maximum time difference = 0.001022
Average time difference = 0.000456
Median time difference = 0.000481
Standard Deviation = 0.000289

Cisco Company Confidential Using TGN and PKTS Timestamps to Measure Latency 5-7
Example of Using Timestamps

5-8 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 A P P E N D I X A

Examples of Traffic Streams

Datalink Traffic Streams

Example of Unknown Datalink Header with IP and TCP Headers

Traffic stream 1 of 1, TCP, ATM4/0/0 (administratively down)
name ""
on
rate 10
send 0
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-data-length 4
L2-data 0 "00 00 00 00"
!
L3-version 4
L3-header-length auto
L3-tos 0x00
L3-length auto
L3-id 0x0000
L3-fragmentation 0x0000
L3-ttl 60
L3-protocol 6
L3-checksum auto
L3-src-addr 0.0.0.0
L3-dest-addr 0.0.0.0
L3-option-length 0
!
L4-src-port 0
L4-dest-port 0
L4-sequence 0x00000000
L4-acknowledge 0x00000000
L4-header-length auto
L4-flags 0x00
L4-window 0
L4-checksum auto
L4-urgent 0
L4-option-length 0

Cisco Company Confidential Examples of Traffic Streams A-1

Datalink Traffic Streams

!
data-length 0
!
fill-pattern 0x00 0x01

HEX Datalink Header Update Commands (page 2-2)

Example of Ethernet with ARPA Encapsulation Traffic Stream

Traffic stream 1 of 1, Datalink, Ethernet0/0/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-encapsulation arpa
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-protocol 0x0000
!
data-length 0
!
fill-pattern 0x00 0x01

Ethernet ARPA Encap Datalink Header Field Update Commands (page 2-2)
show mac Displaying MAC Addresses (page 2-51)

Example of Ethernet with SNAP Encapsulation Traffic Stream

Traffic stream 1 of 1, Datalink, Ethernet0/0/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-encapsulation snap
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-ether-length auto
L2-dsap 0xAA
L2-ssap 0xAA
L2-control 0x03
L2-snap-oui 0x000000
L2-protocol 0x0000
!
data-length 0
!
fill-pattern 0x00 0x01

Ethernet SNAP Encap Datalink Header Field Update Commands (page 2-2)

A-2 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Example of Ethernet with SAP Encapsulation Traffic Stream

Example of Ethernet with SAP Encapsulation Traffic Stream

Traffic stream 1 of 1, Datalink, Ethernet0/0/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-encapsulation sap
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-ether-length auto
L2-dsap 0xAA
L2-ssap 0xAA
L2-control 0x03
!
data-length 0
!
fill-pattern 0x00 0x01

Ethernet SAP Encap Datalink Header Field Update Commands (page 2-2)

Example of Ethernet with Novell-Ether Encapsulation Traffic Stream

Traffic stream 1 of 1, Datalink, Ethernet0/0/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-encapsulation novell-ether
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-ether-length auto
!
data-length 0
!
fill-pattern 0x00 0x01

Ethernet Novell-Ether Encap Datalink Header Field Update Commands (page 2-2)

Cisco Company Confidential Examples of Traffic Streams A-3

Datalink Traffic Streams

Example of Token Ring with SNAP Encapsulation Traffic Stream

Traffic stream 1 of 1, Datalink, TokenRing3/1/0 (administratively down)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-encapsulation snap
L2-access-control 0x10
L2-frame-control 0x40
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-rif
L2-dsap 0xAA
L2-ssap 0xAA
L2-control 0x03
L2-snap-oui 0x000000
L2-protocol 0x0000
!
data-length 0
!
fill-pattern 0x00 0x01

Token Ring SNAP Encap Datalink Header Field Update Commands (page 2-3)

Example of Token Ring with SAP Encapsulation Traffic Stream

Traffic stream 1 of 1, Datalink, TokenRing3/1/0 (administratively down)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-encapsulation sap
L2-access-control 0x10
L2-frame-control 0x40
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-rif
L2-dsap 0xAA
L2-ssap 0xAA
L2-control 0x03
!
data-length 0
!
fill-pattern 0x00 0x01

Token Ring SAP Encap Datalink Header Field Update Commands (page 2-3)

A-4 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Example of FDDI with SNAP Encapsulation Traffic Stream

Example of FDDI with SNAP Encapsulation Traffic Stream

Traffic stream 1 of 1, Datalink, Fddi0/1/0 (administratively down)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-encapsulation snap
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-dsap 0xAA
L2-ssap 0xAA
L2-control 0x03
L2-snap-oui 0x000000
L2-protocol 0x0000
!
data-length 0
!
fill-pattern 0x00 0x01

FDDI SNAP Encap Datalink Header Field Update Commands (page 2-3)

Example of FDDI with SAP Encapsulation Traffic Stream

Traffic stream 1 of 1, Datalink, Fddi0/1/0 (administratively down)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst duration on 1000 to 1000
burst duration off 1000 to 1000
burst off
length auto
!
L2-encapsulation sap
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-dsap 0xAA
L2-ssap 0xAA
L2-control 0x03
!
data-length 0
!
fill-pattern 0x00 0x01

FDDI SAP Encap Datalink Header Field Update Commands (page 2-3)

Cisco Company Confidential Examples of Traffic Streams A-5

Network Protocol Traffic Stream

Example of Serial HDLC Traffic Streams

Traffic stream 1 of 1, Datalink, Serial8/1/0 (administratively down)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
HDLC
L2-flags 0x0F00
L2-protocol 0x0000
!
data-length 0
!
fill-pattern 0x00 0x01

Serial HDLC Datalink Header Field Update Commands (page 2-3)

Network Protocol Traffic Stream

These examples are all on Ethernet.

Example of IP Traffic Stream

Traffic stream 1 of 1, IP, Ethernet1/1/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-encapsulation arpa
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-protocol 0x0800
!
L3-version 4
L3-header-length auto
L3-tos 0x00
L3-length auto
L3-id 0x0000
L3-fragmentation 0x0000
L3-ttl 60
L3-protocol 0
L3-checksum auto
L3-src-addr 0.0.0.0
L3-dest-addr 0.0.0.0
L3-option-length 0
!
data-length 0
!
fill-pattern 0x00 0x01

A-6 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Example of IPX Traffic Stream

IP Network Header Field Update Commands (page 2-4)

show ip Displaying IP Header Information (page 2-50)
Explanation of IP Header Fields (page B-1)

Example of IPX Traffic Stream

Traffic stream 1 of 1, IPX, Ethernet1/1/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-encapsulation novell-ether
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-ether-length auto
!
L3-checksum 0xFFFF
L3-length auto
L3-transport-control 0x00
L3-packet-type 0x00
L3-dest-net 0
L3-dest-host 0000.0000.0000
L3-dest-socket 0
L3-src-net 0
L3-src-host 0000.0000.0000
L3-src-socket 0
!
data-length 0
!
fill-pattern 0x00 0x01

show ipx Displaying IPX Header Information (page 2-51)

IPX Network Header Field Update Commands (page 2-5)
Explanation of IPX Header Fields (page B-6)

Example of AppleTalk Phase 2 Traffic Stream

Traffic stream 1 of 1, APPLE, Ethernet1/1/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-ether-length auto
L2-dsap 0xAA
L2-ssap 0xAA
L2-control 0x03
L2-snap-oui 0x080007

Cisco Company Confidential Examples of Traffic Streams A-7

Network Protocol Traffic Stream

L2-protocol 0x809B
!
L3-phase 2
L3-hopcount 0
L3-length auto
L3-checksum 0x0000
L3-dest-net 0
L3-src-net 0
L3-dest-node 0
L3-src-node 0
L3-dest-socket 0
L3-src-socket 0
L3-ddp-type 0
!
data-length 0
!
fill-pattern 0x00 0x01

show appletalk Displaying AppleTalk Header Information (page 2-42)

AppleTalk Phase 2 Network Header Field Update Commands (page 2-5)
Explanation of AppleTalk Header Fields (page B-10)

Example of AppleTalk Phase 1 Traffic Stream

Traffic stream 1 of 1, APPLE, Ethernet1/1/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-protocol 0x809B
!
L3-phase 1
L3-llap-dest-node 255
L3-llap-src-node 0
L3-llap-type 2
L3-hopcount 0
L3-length auto
L3-checksum 0x0000
L3-dest-net 0
L3-src-net 0
L3-dest-node 0
L3-src-node 0
L3-dest-socket 0
L3-src-socket 0
L3-ddp-type 0
!
data-length 0
!
fill-pattern 0x00 0x01

show appletalk Displaying AppleTalk Header Information (page 2-42)

AppleTalk Phase 1 Network Header Field Update Commands (page 2-5)
Explanation of AppleTalk Header Fields (page B-10)

A-8 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Example of VINES Traffic Stream

Example of VINES Traffic Stream

Traffic stream 1 of 1, VINES, Ethernet1/1/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-encapsulation arpa
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-protocol 0x809B
!
L3-checksum 0x0000
L3-length auto
L3-transport-control 0x0E
L3-packet-type 0x00
L3-dest-net 0
L3-dest-subnet 0
L3-src-net 0
L3-src-subnet 0
!
data-length 0
!
fill-pattern 0x00 0x01

sre Defining Traffic Streams for TGN or SRE (page 2-59)

CLNS Network Header Field Update Commands (page 2-6)
Explanation of Vines Header Fields (page B-9)

Example of CLNS Traffic Stream

Traffic stream 1 of 1, CLNS, Ethernet1/1/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-ether-length auto
L2-dsap 0xFE
L2-ssap 0xFE
L2-control 0x03
!
L3-id 0x81
L3-header-length auto
L3-version 1
L3-lifetime 100
L3-flags 0x1E
L3-segment-length auto
L3-checksum 0x0000
L3-dest-length auto

Cisco Company Confidential Examples of Traffic Streams A-9

Network Protocol Traffic Stream

L3-dest-area 47.0000.0000.0000
L3-dest-host 0000.0000.0000
L3-dest-protocol 0x00
L3-src-length auto
L3-src-area 47.0000.0000.0000
L3-src-host 0000.0000.0000
L3-src-protocol 0x00
L3-option-length 0
!
data-length 0
!
fill-pattern 0x00 0x01

CLNS Area Fields (page 3-5)

show clns Displaying CLNS Header Information (page 2-44)
CLNS Network Header Field Update Commands (page 2-6)
Explanation of CLNS Header Fields (page B-12)

Example of DECnet Traffic Stream

Traffic stream 1 of 1, DECNET, Ethernet1/1/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-protocol 0x6003
!
L3-length auto
L3-flags 0x1E
L3-dest-area 0
L3-dest-node 0
L3-src-area 0
L3-src-node 0
L3-next-level2 0
L3-visit-count 0
L3-service 0
L3-protocol 0
!
data-length 0
!
fill-pattern 0x00 0x01

show decnet Displaying DECnet Header Information (page 2-46)

DECnet Network Header Field Update Commands (page 2-6)
Explanation of DECnet Header Fields (page B-11)

A-10 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Example of XNS Traffic Stream

Example of XNS Traffic Stream

Traffic stream 1 of 1, XNS, Ethernet1/1/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-protocol 0x0600
!
L3-checksum 0xFFFF
L3-length auto
L3-transport-control 0x00
L3-packet-type 0x00
L3-dest-net 0
L3-dest-host 0000.0000.0000
L3-dest-socket 0
L3-src-net 0
L3-src-host 0000.0000.0000
L3-src-socket 0
!
data-length 0
!
fill-pattern 0x00 0x01

sre Defining Traffic Streams for TGN or SRE (page 2-59)

L4-.... Updating the Transport Header Definition (page 2-7)
Explanation of XNS Header Fields (page B-7)

Example of ARP (IP) Traffic Stream

Traffic stream 1 of 1, ARP, Ethernet1/1/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-encapsulation arpa
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-protocol 0x0806
!
L3-hardware 1
L3-protocol 0x0800
L3-hardware-len 6
L3-protocol-len 4
L3-operation 1
L3-sender-haddr 0000.0000.0000
L3-sender-paddr 0.0.0.0
L3-target-haddr 0000.0000.0000
L3-target-paddr 0.0.0.0

Cisco Company Confidential Examples of Traffic Streams A-11

Network Protocol Traffic Stream

!
data-length 0
!
fill-pattern 0x00 0x01

show arp Displaying ARP Header Information (page 2-43)

ARP Network Header Field Update Commands (page 2-4)
Explanation of IP ARP and Appletalk ARP Header Fields (page B-14)

Example of AARP (AppleTalk ARP) Traffic Stream

Traffic stream 1 of 1, AARP, Ethernet1/1/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-ether-length auto
L2-dsap 0xAA
L2-ssap 0xAA
L2-control 0x03
L2-snap-oui 0x000000
L2-protocol 0x80F3
!
L3-hardware 1
L3-protocol 0x809B
L3-hardware-len 6
L3-protocol-len 04
L3-operation 1
L3-sender-haddr 0000.0000.0000
L3-sender-net 0
L3-sender-node 0
L3-target-haddr 0000.0000.0000
L3-target-net 0
L3-target-node 0
!
data-length 0
!
fill-pattern 0x00 0x01

show aarp Displaying AARP Header Information (page 2-41)

AARP (AppleTalk ARP) Network Header Field Update Commands (page 2-6)
Explanation of IP ARP and Appletalk ARP Header Fields (page B-14)

A-12 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 IP Transport Protocol Traffic Streams

IP Transport Protocol Traffic Streams

The following are examples of IP transport protocol traffic streams.

Example of TCP Traffic Stream

Traffic stream 1 of 1, TCP, Ethernet1/1/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-encapsulation arpa
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-protocol 0x0800
!
L3-version 4
L3-header-length auto
L3-tos 0x00
L3-length auto
L3-id 0x0000
L3-fragmentation 0x0000
L3-ttl 60
L3-protocol 6
L3-checksum auto
L3-src-addr 0.0.0.0
L3-dest-addr 0.0.0.0
L3-option-length 0
!
L4-src-port 0
L4-dest-port 0
L4-sequence 0x00000000
L4-acknowledge 0x00000000
L4-header-length auto
L4-flags 0x00
L4-window 0
L4-checksum 0
L4-urgent 0
L4-option-length 0
!
data-length 0
!
fill-pattern 0x00 0x01

show tcp Displaying TCP Header Information (page 2-58)

TCP Transport Header Field Update Commands (page 2-7)
Explanation of TCP Header Fields (page B-3)

Cisco Company Confidential Examples of Traffic Streams A-13

IP Transport Protocol Traffic Streams

Example of UDP Traffic Stream

Traffic stream 1 of 1, UDP, Ethernet1/1/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-encapsulation arpa
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-protocol 0x0800
!
L3-version 4
L3-header-length auto
L3-tos 0x00
L3-length auto
L3-id 0x0000
L3-fragmentation 0x0000
L3-ttl 60
L3-protocol 17
L3-checksum auto
L3-src-addr 0.0.0.0
L3-dest-addr 0.0.0.0
L3-option-length 0
!
L4-src-port 0
L4-dest-port 0
L4-length auto
L4-checksum 0x0000
!
data-length 0
!
fill-pattern 0x00 0x01

show udp Displaying UDP Header Information (page 2-59)

UDP Transport Header Field Update Commands (page 2-7)
Explanation of UDP Header Fields (page B-4)

A-14 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Example of ICMP Traffic Stream

Example of ICMP Traffic Stream

Traffic stream 1 of 1, ICMP, Ethernet1/1/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-encapsulation arpa
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-protocol 0x0800
!
L3-version 4
L3-header-length auto
L3-tos 0x00
L3-length auto
L3-id 0x0000
L3-fragmentation 0x0000
L3-ttl 60
L3-protocol 1
L3-checksum auto
L3-src-addr 0.0.0.0
L3-dest-addr 0.0.0.0
L3-option-length 0
!
L4-type 0
L4-code 0
L4-checksum 0
L4-option 0
!
data-length 0
!
fill-pattern 0x00 0x01

show icmp Displaying ICMP Header Information (page 2-48)

ICMP Transport Header Field Update Commands (page 2-7)
Explanation of ICMP Header Fields (page B-4)

Cisco Company Confidential Examples of Traffic Streams A-15

IP Transport Protocol Traffic Streams

Example of IGMP Traffic Stream

Traffic stream 1 of 1, IGMP, Ethernet1/1/0 (up)
name ""
on
rate 10
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
length auto
!
L2-encapsulation arpa
L2-dest-addr 0000.0000.0000
L2-src-addr 0000.0000.0000
L2-protocol 0x0800
!
L3-version 4
L3-header-length auto
L3-tos 0x00
L3-length auto
L3-id 0x0000
L3-fragmentation 0x0000
L3-ttl 60
L3-protocol 2
L3-checksum auto
L3-src-addr 0.0.0.0
L3-dest-addr 0.0.0.0
L3-option-length 0
!
L4-version 0
L4-type 0
L4-checksum 0x0000
L4-group-address 0.0.0.0
!
data-length 0
!
fill-pattern 0x00 0x01

show igmp Displaying IGMP Header Information (page 2-48)

IGMP Transport Header Field Update Commands (page 2-9)
Explanation of IGMP Header Fields (page B-5)

A-16 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 IPv6 Traffic Streams

IPv6 Traffic Streams

IPv6 headers and derived protocol headers are supported through the Layer command (see layer
Replacing the Template for a Specific Layer (page 2-9)).

Example of IPv6 Header with Routing Header Extension

Traffic stream 1 of 1, IP, Ethernet0/0 (up)
name ""
on
rate 10
variability 0
send 0
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
!
datalink user-defined
length auto
!
L2-encapsulation arpa
L2-dest-addr 0000.0000.0000
L2-src-addr AABB.CC00.0B00
L2-protocol 0x86DD
!
Layer 3 ipv6
L3-version 6
L3-traffic-class 0
L3-flow-label 0x0
L3-payload-length auto
L3-next-header auto
L3-hop-limit 64
L3-src-addr ::
L3-dest-addr ::
L3-header total 1 modules
L3-header 0 is routing
L3-header 0 next-header auto
L3-header 0 hdr-ext-len auto
L3-header 0 type auto
L3-header 0 segments-left 0
L3-header 0 format is type0
L3-header 0 format reserved 0
L3-header 0 format address length 3 units
L3-header 0 format address 0 ::
L3-header 0 format address 1 ::
L3-header 0 format address 2 ::
!
data-length 0
!
fill-pattern 0x00 0x01

IPv6 Layer Header Field Update Commands (page 2-10)

Cisco Company Confidential Examples of Traffic Streams A-17

IPv6 Traffic Streams

Example of TCP over IPv6 Traffic Stream

Traffic stream 1 of 1, TCP, Ethernet0/0 (up)
name ""
on
rate 10
variability 0
send 0
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
!
datalink user-defined
length auto
!
L2-encapsulation arpa
L2-dest-addr 0000.0000.0000
L2-src-addr AABB.CC00.0A00
L2-protocol 0x86DD
!
Layer 3 ipv6
L3-version 6
L3-traffic-class 0
L3-flow-label 0x0
L3-payload-length auto
L3-next-header auto
L3-hop-limit 64
L3-src-addr ::
L3-dest-addr ::
L3-header total 0 modules
!
L4-src-port 0
L4-dest-port 0
L4-sequence 0x00000000
L4-acknowledge 0x00000000
L4-header-length auto
L4-flags 0x00
L4-window 0
L4-checksum auto
L4-urgent 0
L4-option-length 0
!
data-length 0
!
fill-pattern 0x00 0x01

IPv6 Layer Header Field Update Commands (page 2-10)

A-18 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Example of UDP over IPv6 Traffic Stream

Example of UDP over IPv6 Traffic Stream

Traffic stream 1 of 1, UDP, Ethernet0/0 (up)
name ""
on
rate 10
variability 0
send 0
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
!
datalink user-defined
length auto
!
L2-encapsulation arpa
L2-dest-addr 0000.0000.0000
L2-src-addr AABB.CC00.0A00
L2-protocol 0x86DD
!
Layer 3 ipv6
L3-version 6
L3-traffic-class 0
L3-flow-label 0x0
L3-payload-length auto
L3-next-header auto
L3-hop-limit 64
L3-src-addr ::
L3-dest-addr ::
L3-header total 0 modules
!
L4-src-port 0
L4-dest-port 0
L4-length auto
L4-checksum auto
!
data-length 0
!
fill-pattern 0x00 0x01

IPv6 Layer Header Field Update Commands (page 2-10)

Cisco Company Confidential Examples of Traffic Streams A-19

IPv6 Traffic Streams

Example of ICMPv6 Echo Request Message Traffic Stream

Traffic stream 1 of 1, IP, Ethernet0/0 (up)
name ""
on
rate 10
variability 0
send 0
repeat 1 no-update
delayed-start random
burst off
burst duration on 1000 to 1000
burst duration off 1000 to 1000
!
datalink user-defined
length auto
!
L2-encapsulation arpa
L2-dest-addr 0000.0000.0000
L2-src-addr AABB.CC00.0A00
L2-protocol 0x86DD
!
Layer 3 ipv6
L3-version 6
L3-traffic-class 0
L3-flow-label 0x0
L3-payload-length auto
L3-next-header auto
L3-hop-limit 64
L3-src-addr ::
L3-dest-addr ::
L3-header total 0 modules
!
Layer 4 icmpv6
L4-type auto
L4-code 0
L4-checksum auto
L4-message is echo-request
L4-message identifier 0
L4-message sequence-number 0
L4-message data length 0 bytes
!
data-length 0
!
fill-pattern 0x00 0x01

IPv6 Layer Header Field Update Commands (page 2-10)

ICMPv6 Layer Header Field Update Commands (page 2-10)

A-20 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 ARP Responder and Hello-Generator Traffic Streams

ARP Responder and Hello-Generator Traffic Streams

The following are examples of ARP responder and hello-generator traffic streams.

Example of IP ARP Responder

Traffic stream 1 of 1, ARP Responder, Ethernet1/1/0 (up)
on
ip-address 0.0.0.0
mac-address 0011.2222.3333

IP ARP Responder (page 4-1)

show arp-responder Displaying ARP Responders (page 2-43)

Example of AARP (AppleTalk ARP) Responder

Traffic stream 1 of 1, AARP Responder, Ethernet1/1/0 (up)
on
appletalk-address 0.0
mac-address 0011.2222.3333

AppleTalk ARP Responder (page 4-2)

show aarp-responder Displaying AARP Responders (page 2-41)

Example of VINES Hello-Generator

Traffic stream 1 of 1, Vines Hello-Generator, Ethernet1/1/0 (up)
on
hello-address 0:0000
mac-address 0011.2222.3333

VINES Hello-Generator (page 4-2)

sre Defining Traffic Streams for TGN or SRE (page 2-59)

Example of CLNS Hello-Generator

Traffic stream 1 of 1, CLNS Hello-Generator, Ethernet1/1/0 (up)
on
hello-address 47.0000.0000.0000.0011.2222.3333.00

CLNS Hello-Generator (page 4-2)

show clns-hello-generator Displaying CLNS Hello-generators (page 2-45)

Example of DECnet Hello-Generator

Traffic stream 1 of 1, DECnet Hello-Generator, Ethernet1/1/0 (up)
on
hello-address 0.0
designated-router 0.0

DECnet Hello-Generator (page 4-3)

show decnet-hello Displaying DECnet Hello-Generators (page 2-46)

Cisco Company Confidential Examples of Traffic Streams A-21

ARP Responder and Hello-Generator Traffic Streams

A-22 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 A P P E N D I X B

Explanation of Fields in Headers

The following sections give a cursory explanation of the fields in the headers supported by this
program.

Explanation of IP Header Fields

Version:
4 bit field
IP version Current IP protocol version is 4
Header Length:
4 bit field
Length of IP header in number of 32 bit words. Typical IP header is 20 bytes long, which means
5 in this field (5 of 32 bit words).
Type of Service (TOS):
8 bit field
What bit positions indicate
11100000 - Precedence (importance of datagram)
00010000 - Low delay
00001000 - High throughput
00000100 - High reliability
00000011 - unused
Length:
16 bit field
Total IP length including IP header and data
Identification:
16 bit field
Unique identifier for each packet in case of fragmentation

Cisco Company Confidential Explanation of Fields in Headers B-1

Explanation of IP Header Fields

Fragment:
16 bit field
What bit positions indicate
0x4000 : Dont Fragment indication
0x2000 : More Fragments indication
0x1FFF : Fragment Offset Indicates start of this fragment in original IP header
Time To Live (TTL):
8 bit field
Specifies how long, in seconds, the datagram is allowed to live on the Internet. Each router either
decrements this field by 1 or by the number of seconds it holds the packet. A datagram is
discarded if this field reaches zero.
Protocol:
8 bit field
Identifies the higher level data
A few common protocols:
1: ICMP
2: IGMP
6: TCP
8: EGP
9: IGRP
17: UDP
88: EIGRP
89: OSPF
Header Checksum:
16 bit field
Used to ensure integrity of IP header Ones complement arithmetic of IP header as sequence of
16-bit integers. Does not include IP data.
Source IP Address:
32 bit field
IP address of the datagram sender
Destination IP Address:
32 bit field
IP address of the datagram intended recipient
Option:
Variable length up to 40 bytes max
Used to record routes, timestamps

B-2 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Explanation of TCP Header Fields

Explanation of TCP Header Fields

Source Port:
16 bit field
Identifies application program at source
Destination Port:
16 bit field
Identifies application program at destination
Sequence Number:
32 bit field
Identifies the position in the senders byte stream of the data in this segment
Acknowledge Number:
32 bit field
Identifies the number of the octet that the source expects to receive next
Header Length:
First 4 bits of 8 bit field
Length of TCP header in 32-bit words
Flags:
8 bit field
00100000 - URG: Urgent pointer field is valid
00010000 - ACK: Acknowledge field is valid
00001000 - PSH: This segment requests a push
00000100 - RST: Reset the connection
00000010 - SYN: Synchronize sequence numbers
00000001 - FIN: Sender has reached end of its byte stream
Window:
16 bit field
Sender advertises how much data it is willing to accept in its buffer
Checksum:
16 bit field
1s complement checksum calculated on TCP header, TCP data, and IP pseudo header (Source
IP address, Destination IP address, Protocol and Length)
Urgent:
16 bit field
If URG bit is set, this points to urgent, or out-of-band information in byte stream
Option:
Variable length

Cisco Company Confidential Explanation of Fields in Headers B-3

Explanation of UDP Header Fields

Explanation of UDP Header Fields

Source Port:
16 bit field
Identifies application program at source
Destination Port:
16 bit field
Identifies application program at destination
Length:
16 bit field
Total UDP length including UDP header and data
Checksum:
16 bit field
1s complement checksum calculated on UDP header, UDP data, and IP pseudo header
(Source IP address, Destination IP address, Protocol and Length)

Explanation of ICMP Header Fields

Type:
8 bit field
Identifies ICMP message
0: Echo Reply
3: Destination Unreachable
4: Source Quench
5: Redirect
8: Echo Request
11: Time Exceeded for a Datagram
12: Parameter problem on a Datagram
13: Timestamp Request
14: Timestamp Reply
17: Address Mask Request
18: Address Mask Reply

B-4 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Explanation of IGMP Header Fields

Code:
8 bit field
Further information on ICMP message
0: Network Unreachable
1: Host Unreachable
2: Protocol Unreachable
3: Port Unreachable
4: Fragmentation Needed and DF set
5: Source Route Field
6: Destination network unknown
7: Destination host unknown
8: Source host isolated
9: Communication with destination network administratively prohibited
10: Communication with destination host administratively prohibited
11: Network unreachable for type of service
12: host unreachable for type of service
Checksum:
16 bit field
1s complement checksum calculated on ICMP header and ICMP data

Explanation of IGMP Header Fields

Version:
4 bit field
IGMP protocol version Currently 1
Type:
4 bit field
Identifies message as:
1: Query
2: Response
Unused:
8 bit field
Unused byte in header; should be set to zero
Checksum:
16 bit field
Checksum calculated on IGMP header and IGMP data

Cisco Company Confidential Explanation of Fields in Headers B-5

Explanation of IPX Header Fields

Group Address:
32 bit field
Hosts use this field to report their membership to a particular multicast group.

Explanation of IPX Header Fields

Checksum
16 bit field
Optional Usually set to 0xFFFF
Length
16 bit field
Length of IPX header and data
Transport Control
8 bit field
11110000 - reserved
00001111 - Hop Count Packet sent with hop count of 0. Field is incremented by each router.
Packet is dropped by router receiving hop count of 15.
Packet Type
8 bit field
Defines higher level data
0: Unknown Packet Type
1: RIP
2: Echo Packet
3: Error Packet
4: Packet Exchange Protocol (PEP)
5: Sequenced Packet Protocol (SPP)
17: Netware Core Protocol (NCP)
Destination Network
32 bit field
Indicates which network the destination station is on
Destination Address
48 bit field
MAC address of destination station

B-6 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Explanation of XNS Header Fields

Destination Socket
16 bit field
Specifies process within destination station
0x0001: Routing Information Packet
0x0002: Echo Protocol Packet
0x0003: Error Handler Packet
0x0451: File Service Packet
0x0452: Service Advertising Packet
0x0453: Routing Information Packet
0x0455: NetBIOS Packet
0x0456: Diagnostic Packet
Source Network
32 bit field
Indicates which network the source station is on
Source Address
48 bit field
MAC address of source station
Source Socket
16 bit field
Process sending packet See Destination Socket for values

Explanation of XNS Header Fields

Checksum
16 bit field
Optional Usually set to 0xFFFF
Length
16 bit field
Length of XNS header and data
Transport Control
8 bit field
11110000 - reserved
00001111 - Hop Count Packet sent with hop count of 0. Field is incremented by each router.
Packet is dropped by router receiving hop count of 15

Cisco Company Confidential Explanation of Fields in Headers B-7

Explanation of XNS Header Fields

Packet Type
8 bit field
Defines higher level data
0: Unknown Packet Type
1: RIP
2: Echo Packet
3: Error Packet
4: Packet Exchange Protocol (PEP)
5: Sequenced Packet Protocol (SPP)
Destination Network
32 bit field
Indicates which network the destination station is on
Destination Address
48 bit field
MAC address of destination station
Destination Socket
16 bit field
Specifies process within destination station
0x0001: Routing Information Packet
0x0002: Echo Protocol Packet
0x0003: Error Handler Packet
Source Network
32 bit field
Indicates which network the source station is on
Source Address
48 bit field
MAC address of source station
Source Socket
16 bit field
Process sending packet (see Destination Socket for values)

B-8 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Explanation of Vines Header Fields

Explanation of Vines Header Fields

Checksum
16 bit field
Optional Usually set to 0xFFFF
Length
16 bit field
Length of VINES header and data
Transport Control
8 bit field
11110000 - reserved
00001111 - Hop Count Packet sent with hop count of 0. Field is incremented by each router.
Packet is dropped by router receiving hop count of 15.
Packet Type
8 bit field
Defines higher level data
0: Unknown Packet Type
1: RIP
2: Echo Packet
3: Error Packet
4: Packet Exchange Protocol (PEP)
5: Sequenced Packet Protocol (SPP)
Destination Network
32 bit field
Indicates which network the destination station is on
Destination Sub-Network
16 bit field
Address assigned to destination station
Source Network
32 bit field
Indicates which network the source station is on
Source Sub-Network
16 bit field
Address assigned to source station

Cisco Company Confidential Explanation of Fields in Headers B-9

Explanation of AppleTalk Header Fields

Explanation of AppleTalk Header Fields

The AppleTalk network header is also call Datagram Delivery Protocol (DDP).
Hop Count
4 bit field
Packet sent with hop count of 0. Field is incremented by each route.r Packet is dropped by router
receiving hop count of 15.
Length
10 bit field
Length of DDP header and data
Checksum
16 bit field
Checksum calculated on DDP header and data set to 0x0000 if checksum not performed
Destination Network
16 bit field
Network address of destination station
Source Network
16 bit field
Network address of source station
Destination Node
8 bit field
Node address of destination station
Source Node
8 bit field
Node address of source station
Destination Socket
8 bit field
Process addressed on destination node
Source Socket
8 bit field
Process sending packet
1: RTMP
2: NIS
4: Echoer
6: ZIS

B-10 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Explanation of DECnet Header Fields

Type
8 bit field
Identification of higher level data
1: RTMP Response or Data
2: NBP
3: ATP
4: AEP
5: RTMP Request
6: ZIP
7: ADSP

Explanation of DECnet Header Fields

Length
16 bit field
Length of DECnet header plus data
Flag
8 bit field
1000 0000: Padding
0100 0000: Version
0010 0000: Intra-Ethernet Packet
0001 0000: Return packet
0000 1000: Do not return to sender
0000 0110: Packet Type: Long Data Packet
0000 0001: Control packet
Dest Area
8 bit field
Not used Keep at 0
Dest SubArea
8 bit field
Not used Keep at 0
Dest Address
48 bit field
MAC address of Destination station Station area and node are encoded in the MAC address
Source Area
8 bit field
Not used Keep at 0

Cisco Company Confidential Explanation of Fields in Headers B-11

Explanation of CLNS Header Fields

Source SubArea
8 bit field
Not used Keep at 0
Source Address
48 bit field
MAC address of Source station Station area and node are encoded in the MAC address
NL2
8 bit field
Next Level Router
Visits
8 bit field
The number of routers the packet has passed through
Class
8 bit field
Service Class Not used Keep at 0
Protocol
8 bit field
Protocol Type Not used Keep at 0

Explanation of CLNS Header Fields

ID
8 bit field
Format of this OSI header. The format of the rest of this header is for a data packet; hello packets
have different formats.
129: Data Packet
130: End System Hello
131: Intermediate System Hello
Header Length
8 bit field
Length of this header in bytes
Version
8 bit field
Version of header format; current version in use is 1
Lifetime
8 bit field
Remaining life time of packet in 1/2 seconds. Sets how long a packet can exist on the Internet
before it is dropped.
B-12 TGN (Traffic GeNerator) User Manual Cisco Company Confidential
 Explanation of CLNS Header Fields

Flags
8 bit field
10000000: SP = Segmentation permitted
01000000: MS = More segments
00100000: E/R = Error Report Requested
00011111: Type = Packet Type
00011100: Data Packet
00000001: Error Packet
Segment Length
16 bit field
Length of Header and data
Checksum
16 bit field
Checksum on Header. If checksum not calculated, set to 0x0000.
Destination Length
8 bit field
This the length of the destination address in number of bytes. Normally, the destination address
is considered to be a single number. Here it is split into Area, Address, and Protocol, This length
covers all three together,
Destination Area
Variable length
This is the area of the destination station
Destination Address
48 bit field
This is the MAC address of the destination station
Destination Protocol
8 bit field
This is the process on the destination station that should receive this packet
Source Length
8 bit field
This the length of the destination address in number of bytes, Normally, the destination address
is considered to be a single number. Here it is split into Area, Address, and Protocol. This length
covers all three together.
Source Area
Variable length
This is the area of the source station

Cisco Company Confidential Explanation of Fields in Headers B-13

Explanation of IP ARP and Appletalk ARP Header Fields

Source Address
48 bit field
This is the MAC address of the source station
Source Protocol
8 bit field
This is the process on the source station sending the packet

Explanation of IP ARP and Appletalk ARP Header Fields

Hardware
16 bit field
Identification of sender media
Protocol
16 bit field
Identification of protocol
IP is 0x0800
Appletalk is 0x809B
Hardware Length
8 bit field
Length of hardware field in bytes, (MAC address) Value is 6
Protocol Length
8 bit field
Length of protocol address field in bytes, (IP or AppleTalk address). Value is 4. For AppleTalk,
the first byte is zero, the second and third have the network address, and the forth byte has the
node address.
Operation
16 bit field
Purpose of this ARP packet
1 = Request
2 = Reply
Sender Hardware Address
48 bit field
MAC address of the station sending the packet. Holds the requested information in reply.
Sender Protocol Address
32 bit field
IP or Appletalk address of the station sending the packet

B-14 TGN (Traffic GeNerator) User Manual Cisco Company Confidential

 Explanation of IP ARP and Appletalk ARP Header Fields

Target Hardware Address

48 bit field
MAC address of the destination station. Is zero in request.
Target Protocol Address
32 bit field
IP or AppleTalk address of destination station

Cisco Company Confidential Explanation of Fields in Headers B-15

Explanation of IP ARP and Appletalk ARP Header Fields

B-16 TGN (Traffic GeNerator) User Manual Cisco Company Confidential