Flexible Packet Matching XML Configuration

Last Updated: August 11, 2011

The Flexible Packet Matching XML Configuration feature allows the use of eXtensible Markup Language (XML) to define traffic classes and actions (policies) to assist in blocking network attacks. The XML file used by Flexible Packet Matching (FPM) is called the traffic classification definition file (TCDF).

The TCDF gives you an alternative to the command-line interface (CLI) as a method to define traffic classification behavior. Traffic classification behavior is identical regardless of the method you use.

Finding Feature Information

Your software release may not support all the features documented in this module. For the latest feature information and caveats, see the release notes for your platform and software release. To find information about the features documented in this module, and to see a list of the releases in which each feature is supported, see the Feature Information Table at the end of this document.

Use Cisco Feature Navigator to find information about platform support and Cisco software image support. To access Cisco Feature Navigator, go to www.cisco.com/go/cfn. An account on Cisco.com is not required.

Prerequisites for the Flexible Packet Matching XML Configuration

  • A protocol header definition file (PHDF) relevant to the TCDF must be loaded on the router.
  • Although access to an XML editor is not required, using one might make the creation of the TCDF easier.
  • You must be familiar with XML file syntax.

Restrictions for the Flexible Packet Matching XML Configuration

TCDF Image Restriction

TCDF is part of the FPM subsystem. FPM is not included in the Cisco 871 securityk9 image; therefore, TCDF parsing is not present in the Cisco 871 securityk9 image.

The Flexible Packet Matching XML Configuration has the following restrictions:

  • The FPM TCDF cannot be used to mitigate an attack that requires stateful classification.
  • Because FPM is stateless, it cannot keep track of port numbers being used by protocols that dynamically negotiate ports. Thus, when using the FPM TCDF, port numbers must be explicitly specified.
  • FPM cannot perform IP fragmentation or TCP flow reassembly.

Information About the Flexible Packet Matching XML Configuration

Traffic Classification Definition Files for the Flexible Packet Matching XML Configuration

FPM uses a TCDF to define policies that can block attacks on the network. FPM is a packet classification feature that allows users to define one or more classes of network traffic by pairing a rich set of standard matching operators with user-defined protocol header fields. FPM users can create their own stateless packet classification criteria and define policies with multiple actions (such as drop, log, or send Internet Control Message Protocol [ICMP] unreachable) to immediately block new viruses, worms, and attacks on the network.

Before the release of the Flexible Packet Matching XML Configuration feature, FPM defined traffic classes (class maps), policies (policy maps), and service policies (attach policy maps to a class maps) through the use of CLI commands. With TCDFs, FPM can use XML as an alternative to the CLI to define classes of traffic and specify actions to apply to the traffic classes. Traffic classification behavior is the same whether you create the behavior using a TCDF or configure it using CLI commands. Once a TCDF is created, it can be loaded on any FPM-enabled device in the network.

Protocol Header Definition Files for Traffic Classification Definitions

TCDFs require that a relevant PHDF is already loaded on the device. A PHDF defines each field contained in the header of a particular protocol. Each field is described with a name, optional comment, an offset (the location of the protocol header field in relation to the start of the protocol header), and the length of the field. The total length is specified at the end of each PHDF.

The description of a traffic class in a TCDF file can contain header fields defined in a PHDF. If the PHDF is loaded on the router, the class specification to match begins with a list of the protocol headers in the packet. In the TCDF, the traffic class is associated with a policy that binds the match to an action, such as drop, log, or send ICMP unreachable.

FPM provides ready-made definitions for these standard protocols, which can be loaded onto the router with the load protocol command: ether.phdf, ip.phdf, tcp.phdf, and udp.phdf. You can also write your own custom PHDFs using XML if one is required for the TCDF.


Note


Because PHDFs are defined via XML, they are not shown in a running configuration.

Traffic Classification Description File Format and Use

In the TCDF, you can define one or more classes of traffic and policies that describe specified actions for each class of traffic. The TCDF is an XML file that you create in a text file or with an XML editor. The file that you create must have a filename that has the .tcdf extension.

The TCDF has the following basic format. XML tags are shown in bold text for example purposes only.

<tdcf
>
       <class
 ...> ... </class
>
         ...
       <policy
> ... </policy
>
         ...
</tdcf
>

For a traffic class, you can identify a match for any field or fields against any part of the packet.


Note


FPM is stateless and cannot be used to mitigate an attack that requires stateful classification, that is classify across IP fragments, across packets in a TCP stream, or peer-to-peer protocol elements.

Policies can be anything from access control, quality of service (QoS), or even routing decisions. For FPM, the associated actions (policies) might include permit, drop, log, or send ICMP unreachable.

Once loaded, the TCDF-defined classes and policies can be applied to any interface or subinterface and behave in an identical manner as the CLI-defined classes and policies. You can define policies in the TCDF and apply then to any entry point to the network to block new attacks.

Traffic Class Definitions for a Traffic Classification Definition File

A class can be any traffic stream of interest. You define a traffic stream of interest by matching a particular interface or port, a source address or destination IP address, a protocol or an application. The following sections contain information you should understand before you define the traffic class in the TCDF for FPM configuration:

Class Element Attributes for a Traffic Classification Definition File

The table below lists and describes the attributes that you can associate with the class element in a TCDF for the FPM XML configuration. The class element contains attributes you can use to specify the traffic class name, its description and type, where to look in the packet, what kind of match, and when the actions should apply to the traffic.

Table 1 Attributes for Use with the Class Element in a TCDF for the FPM XMLConfiguration

Attribute Name

Use

Type

name (required)

Specifies the name of the class.

Note    When you use the class element inside policy elements, you need specify the name attribute only.

String

type (required)

Specifies the type of class.

Keywords: stack or access-control

stack start

Specifies where to look in the packet. By default, the match starts at Layer 3.

Keyword: l2-start

match

Specifies the type of match to be performed on the class.

Keywords: all or any

  • all--All class matches must be met to perform the policy actions.
  • any--One or more matches within the class must be met to perform the policy actions.

undo

Directs the device to remove the class-map when set to true.

Keywords: true or false

For example, XML syntax for a stack class describing an IP, User Datagram Protocol (UDP), Simple Management Protocol (SNMP) stack might look like this:

<class
 name
="snmp-stack" 
type
="stack">
    <match
>
       <eq
 
field
="ip.protocol" value="x"></eq
>
       <eq
 
field
="udp.dport" 
value
="161"></eq
>
    </match
>
</class
>

Match Element for a Traffic Classification Definition File

The match element in the TCDF for FPM XML configuration contains operator elements. Operator elements are the following: eq (equal to), neq (not equal to), lt (less than), gt (greater than), range (a value in a specific range, for example, range 1 - 25), and regex (regular expression string with a maximum length of 32 characters).

In following sections, these various operators are collectively called the operator element.

Operator Element Attributes for a Traffic Classification Definition File

The table below lists and describes direct matching attributes that you can associate with the operatorelement in a TCDF for the FPM XML configuration.

Table 2 Direct Matching Attributes to Use with a Match Element in a TCDF for the FPM XML Configuration

Attribute Name

Use

Type

start

Begin the match on a predefined keyword or Protocol.Field, if given.

Keyword: l2-start or l3-start

Otherwise, a field of a protocol as defined in the PHDF, for example, the source field in the IP protocol.

offset

Used with start attribute. Offset from the start point.

Hexadecimal or decimal number, or string constants, Protocol.Field, or combination of a constant and Protocol.Field with +, -, *, /, &, or |.

size

Used together with start and offset attributes. How much to match.

Specifies the size of the match in bytes.

mask

Number specifying bits to be matched in protocol or field attributes.

Used exclusively with field type of bitset to specify the bits of interest in a bit map.

Decimal or hexadecimal number

value

Value on which to match.

String, number, or regular expression

field

Specifies the name of the field to be compared.

Name of field as defined in the PHDF

next

Identifies the next layer of the protocol. This attribute can be used only in stack type classes.

Keyword that is the name of a protocol defined in the PHDF.

undo

Directs the device to remove the particular match operator when set to true.

Keywords: true or false

Policy Definitions for a Traffic Classification Definition File

A policy is any action that you apply to a class. You should understand the following information before defining the policy in a TCDF for the FPM XML configuration:

Policy Element Attributes for a Traffic Classification Definition File

Policies can be anything from access control, QoS, or even routing decisions. For FPM, the associated actions or policies might include drop, log, or send ICMP unreachable. Policies describe the action to take to mitigate attacks on the network.

The table below lists and describes the attributes that you can use with the policy element in the TDCF for FPM XML configuration.

Table 3 Attributes for Use with the Policy Element in a TCDF for the FPM XML Configuration

Attribute Name

Use

Type

name

Name of the policy.

String

type

Specifies the type of policy map.

Keyword: access-control

undo

Directs the device to remove the policy map when set to true.

Keywords: true or false

The policy name in this example is sql-slammer, and the action defined for the policy is to drop the packet. This action is to be applied to the class that has the same name as the policy (class name= "sql-slammer").

<policy
 
name
="sql-slammer">
    <class
 
name
="sql-slammer"></class
>
    <action
>drop</action
>
</policy
>

Action Element for a Traffic Classification Definition File

The action element is used to specify actions to associate with a policy. The policy with the action element is applied to a defined class. The action element can contain any of the following: permit, drop, Log, SendBackIcmp, set, RateLimit, alarm, ResetTcpConnection, and DropFlow. For example:

<action
>
   log
</action
>

Traffic Classification Definition File Syntax Guidelines

The following list describes required and optional syntax for the TCDF:

  • The TCDF filename must end in the .tcdf extension, for example, sql_slammer.tcdf.
  • The TCDF contains descriptions for one or more traffic classes and one or more policy actions.
  • The file is encoded in the XML notation.
  • The TCDF file should begin with the following version encoding:

<?xml version="1.0" encoding="UTF-8"?>

The TCDF is used to define traffic classes and the associated policies with specified actions for the purpose of blocking new viruses, worms, and attacks on the network.

The TCDF is configured in a text or XML editor. The syntax of the TCDF must comply with the XML Version 1.0 syntax and the TCDF schema. For information about Version 1.0 XML syntax, see the document at the following url:

http://www.w3.org/TR/REC-xml/

How to Create and Load Traffic Classification Definition Files

Creating a Definition File for the FPM XML Configuration

SUMMARY STEPS

1.    Open a text file or an XML editor and begin the file with the XML version and encoding declaration.

2.    Identify the file as a TCDF. For example:

3.    Define the traffic class of interest.

4.    Identify matching criteria for the defined classes of traffic. For example:

5.    Define the action to apply to the defined class. For example:

6.    End the traffic classification definition. For example:

7.    Save the TCDF file with a filename that has a .tcdf extension, for example: slammer.tcdf.


DETAILED STEPS
Step 1   Open a text file or an XML editor and begin the file with the XML version and encoding declaration.

Example:

<?xml version="1.0" encoding="UTF-8"?>

Step 2   Identify the file as a TCDF. For example:

Example:

<tcdf >

Step 3   Define the traffic class of interest.

For example, a stack class describing an IP and UDP stack might be described as follows. In this example, the name of the traffic class is "ip-udp," and the class type is "stack."



Example:

<class name ="ip-udp" type ="stack"></class >

In the following example, the name of the traffic class is slammer, the class type is access control, and the match criteria is all:



Example:

<class name=" slammer " type ="access-control" match ="all"></class >

Step 4   Identify matching criteria for the defined classes of traffic. For example:

Example:

<class name ="ip-udp" type ="stack"> <match > <eq field ="ip.protocol" value ="0x11" next ="udp"></eq > </match > </class > <class name=" slammer " type ="access-control" match ="all"> <match > <eq field ="udp.dest-port" value ="0x59A"></eq > <eq field ="ip.length" value ="0x194"></eq > <eq start ="l3-start" offset ="224" size ="4" value ="0x00401010"></eq > </match > </class >

The traffic of interest in this TCDF matches fields defined in the PHDF files, ip.phdf and udp.phdf. The matching criteria for slammer packets is a UDP destination port number 1434 (0x59A), an IP length not to exceed 404 (0x194) bytes, and a Layer 3 position with a pattern 0x00401010 at 224 bytes from start (offset) of the IP header.

Step 5   Define the action to apply to the defined class. For example:

Example:

<policy name ="fpm-udp-policy"> <class name ="slammer"></class > <action >Drop</action > </policy >

The policy name in this example is fpm-udp-policy, and the action defined for the policy is to drop the packet. This action is to be applied to the class that has the name slammer.

Step 6   End the traffic classification definition. For example:

Example:

</tcdf >

Step 7   Save the TCDF file with a filename that has a .tcdf extension, for example: slammer.tcdf.

Loading a Definition File for the FPM XML Configuration

SUMMARY STEPS

1.    enable

2.    show protocol phdf protocol-name

3.    configure terminal

4.    load protocol location:filename

5.    load classification location : filename

6.    end

7.    show class-map [type {stack | access-control}] [class-map-name]


DETAILED STEPS
  Command or Action Purpose
Step 1
enable


Example:

Router> enable

 

Enables privileged EXEC mode.

  • Enter your password if prompted.
 
Step 2
show protocol phdf protocol-name


Example:

Router# show protocol phdf ip

 

Displays protocol information from a specific PHDF.

  • Use this command to verify that a PHDF file relevant to the TCDF is loaded on the device.
 
Step 3
configure terminal


Example:

Router# configure terminal

 

Enters global configuration mode.

 
Step 4
load protocol location:filename


Example:

Router(config)# load protocol localdisk1:ip.phdf

 

(Optional) Loads a PHDF onto a router.

  • The specified location must be local to the router.
Note    If the required PHDF is already loaded on the router (see Step 2), skip this step and proceed to Step 5).
 
Step 5
load classification location : filename


Example:

Router(config)# load classification localdisk1:slammer.tcdf

 

Loads a TCDF onto a router.

  • The specified location must be local to the router.
 
Step 6
end


Example:

Router(config)# end

 

Exits to privileged EXEC mode.

 
Step 7
show class-map [type {stack | access-control}] [class-map-name]


Example:

Router# show class-map sql-slammer

 

(Optional) Displays a class map and its matching criteria.

  • Use this command to verify that a class defined in the TCDF file is available on the device.
  • The class-map-name argument is the name of a class in the TCDF.
 

Examples

The following is sample output from a show class-map command that displays the traffic classes defined in the TCDF after it is loaded on the router:

Router# show class-map
.
.
.
class-map type stack match-all ip-udp
   match field IP protocol eq 0x11 next UDP
class-map type access-control match-all slammer 
   match field UDP dest-port eq 0x59A 
   match field IP length eq 0x194 
   match start l3-start offset 224 size 4 eq 0x4011010 
.
.
.

What to Do Next

After you have defined the TCDF, you must apply that policy to an interface as shown in the following task "Associating a Traffic Classification Definition File."

Associating a Traffic Classification Definition File

Perform this task to associate the defination file with an interface or subinterface.

Before You Begin

The TCDP and FPM must be configured on the device.


SUMMARY STEPS

1.    enable

2.    configure terminal

3.    interface type slot / port

4.    service-policy type access-control ] {input | output} policy-map-name

5.    end

6.    show policy-map interface type access-control ] interface-name slot/port[input | output]


DETAILED STEPS
  Command or Action Purpose
Step 1
enable


Example:

Router> enable

 

Enables privileged EXEC mode.

  • Enter your password if prompted.
 
Step 2
configure terminal


Example:

Router# configure terminal

 

Enters global configuration mode.

 
Step 3
interface type slot / port


Example:

Router(config)# interface gigabitEthernet 0/1

 

Configures an interface type and enters interface configuration mode.

 
Step 4
service-policy type access-control ] {input | output} policy-map-name


Example:

Router(config-if)# service-policy type access-control input sql-slammer

 

Specifies the type and the name of the traffic policy to be attached to the input or output direction of an interface.

  • The policy-map-name argument is the name of a policy in the TCDF.
 
Step 5
end


Example:

Router(config-if)# end

 

Exits to privileged EXEC mode.

 
Step 6
show policy-map interface type access-control ] interface-name slot/port[input | output]


Example:

Router# show policy-map interface gigabitEthernet 0/1

 

(Optional) Displays the packet statistics of all classes that are configured for all service policies either on the specified interface or subinterface.

  • Use this command to verify that policy defined in TCDF is associated with the named interface.
 

Displaying TCDF-Defined Traffic Classes and Policies

SUMMARY STEPS

1.    enable

2.    show class-map [ type { stack | access-control}] [class-map-name]

3.    show class-map type stack [class-map name]

4.    show class-map type access-control [class-map-name]

5.    show policy-map [policy-map]

6.    exit


DETAILED STEPS
Step 1   enable

Use this command to enable privileged EXEC mode. Enter your password if prompted. For example:



Example:

Router> enable Router#

Step 2   show class-map [ type { stack | access-control}] [class-map-name]

Use this command to verify that a class defined in the TCDF file is available on the device. For example:



Example:

Router# show class-map . . . class-map type stack match-all ip-udp match field IP protocol eq 0x11 next UDP class-map type access-control match-all slammer match field UDP dest-port eq 0x59A match field IP length eq 0x194 match start l3-start offset 224 size 4 eq 0x4011010 . . .

Step 3   show class-map type stack [class-map name]

Use this command to display the stack type defined for the class of traffic in the TCDF file. For example:



Example:

Router# show class-map type stack ip-udp class-map type stack match-all ip-udp match field IP protocol eq 0x11 next UDP

Step 4   show class-map type access-control [class-map-name]

Use this command to display the access type defined for the class in the TCDF file. For example:



Example:

Router# show class-map type access-control slammer class-map type access-control match-all slammer match field UDP dest-port eq 0x59A match field IP length eq 0x194 match start l3-start offset 224 size 4 eq 0x4011010

Step 5   show policy-map [policy-map]

Use this command to display the contents of a policy map defined in the TCDF. For example:



Example:

Router# show policy-map fpm-udp-policy policy-map type access-control fpm-udp-policy class slammer drop

Step 6   exit

Use this command to exit to user EXEC mode. For example:



Example:

Router# exit Router>


Configuration Examples for Creating and Loading Traffic Classification Definition Files


Note


The TCDF files are created in a text file or with an XML editor. In the following examples, XML tags are shown in bold text and field names in italic text. The values for the attributes are entered in quotation marks ("value").

Example Traffic Classification Definition File for Slammer Packets

The following example shows how to create and load a TCDF for slammer packets (UDP 1434) for the FPM configuration. The match criteria defined within the class element is for slammer packets with an IP length not to exceed 404 (0x194) bytes, UDP destination port 1434 (0x59A), and pattern 0x00401010 at 224 bytes from start of IP header. This example also shows how to define the policy "sql-slammer" with the action to drop slammer packets.

<?xml version="1.0" encoding="UTF-8"?
>
<tcdf
>
    <class
 
name
="ip-udp" 
type
="stack">
       <match
> 
          <eq
 
field
="ip.protocol" 
value
="0x11" 
next
="udp"></eq
>
       </match
>
    </class
> 
    <class
 name="
slammer
" 
type
="access-control" 
match
="all">
       <match
> 
          <eq
 
field
="udp.dest-port" 
value
="0x59A"></eq
>
          <eq
 
field
="ip.length" 
value
="0x194"></eq
>
          <eq
 
start
="l3-start" 
offset
="224" 
size
="4" 
value
="0x00401010"></eq
>
       </match
>
    </class
>
    <policy
 type="access-control" 
name
="fpm-udp-policy">
       <class
 
name
="slammer"></class
>
       <action
>Drop</action
>
    </policy
>
</tcdf
>

The following example shows how to load the TCDF file onto the device and apply the policy defined in the file to the interface Gigabit Ethernet 0/1:

configure terminal
load classification localdisk1:sql-slammer.tcdf
policy-map type access-control my-policy-1
class ip-udp
service-policy fpm-udp-policy
interface gigabitEthernet 0/1
 service-policy type access-control input my-policy-1
 end

Example Traffic Classification Definition File for MyDoom Packets

The following example shows how to create and load a TCDF for MyDoom packets in a text file or XML editor for the FPM XML configuration. The match criteria for the MyDoom packets are as follows:

  • 90 > IP length > 44
  • pattern 0x47455420 at 40 bytes from start of IP header

or

  • IP length > 44
  • pattern 0x47455420 at 40 bytes from start of IP header
<tcdf
>
    <class
 
name
="md-stack" 
type
="stack">
        <match
>
            <eq
 
field
="ip.protocol" 
value
="6" 
next
="tcp"></eq
>
        </match
> 
    </class
>
    <class
 
type
="access-control" 
name
="mydoom1">
        <match
>
            <gt
 
field
="ip.length" 
value
="44"/>
            <lt
 
field
="ip.length" 
value
="90"/>
            <eq
 
start
="ip.version" 
offset
="tcp.headerlength*4+20" 
size
="4"  
                
value
="0x47455420"/>
        </match
>
    </class
>
    <class
 
type
="access-control" 
name
="mydoom2">
        <match
>
            <gt
 field="ip.length" value="44"/>
            <eq
 start="ip.version" offset="tcp.headerlength*4+58" size="4" 
                value="0x6d3a3830"/>
            <eq
 start="ip.version" offset="tcp.headerlength*4+20" size="4" 
                 value="0x47455420"/>
        </match
>
    </class
>
    <policy
 
name
="fpm-md-stack-policy">
       <class
 
name
="mydoom1"></class
>
       <action
>drop</action
>
    </policy
>
    <policy
 
name
="fpm-md-stack-policy">
       <class
 
name
="mydoom2"></class
>
       <action
>drop</action
>
    </policy
>
</tcdf
>

The following example shows how to load the TCDF file onto the device and apply the policy defined in the file to the interface Ethernet 0/1:

configure terminal
load classification localdisk1:sql-slammer.tcdf
policy-map type access-control my-policy-2
class md-stack
service-policy fpm-md-stack-policy
interface Ethernet 0/1
 service-policy type access-control input my-policy-2
 end

Additional References

Related Documents

Related Topic

Document Title

Cisco IOS commands

Cisco IOS Master Commands List, All Releases

Additional configuration information for class maps and policy maps

"Applying QoS Features Using the MQC" module

Information about and configuration tasks for FPM

"Flexible Packet Matching" module

Standards

Standard

Title

No new or modified standards are supported by this feature, and support for existing standards has not been modified by this feature.

--

MIBs

MIB

MIBs Link

No new or modified MIBs are supported by this feature, and support for existing MIBs has not been modified by this feature.

To locate and download MIBs for selected platforms, Cisco IOS releases, and feature sets, use Cisco MIB Locator found at the following URL:

http://www.cisco.com/go/mibs

RFCs

RFC

Title

No new or modified RFCs are supported by this feature, and support for existing RFCs has not been modified by this feature.

--

Technical Assistance

Description

Link

The Cisco Support and Documentation website provides online resources to download documentation, software, and tools. Use these resources to install and configure the software and to troubleshoot and resolve technical issues with Cisco products and technologies. Access to most tools on the Cisco Support and Documentation website requires a Cisco.com user ID and password.

http://www.cisco.com/cisco/web/support/index.html

Feature Information for Flexible Packet Matching XML Configuration

The following table provides release information about the feature or features described in this module. This table lists only the software release that introduced support for a given feature in a given software release train. Unless noted otherwise, subsequent releases of that software release train also support that feature.

Use Cisco Feature Navigator to find information about platform support and Cisco software image support. To access Cisco Feature Navigator, go to www.cisco.com/go/cfn. An account on Cisco.com is not required.

Table 4 Feature Information for Flexible Packet Matching XML Configuration

Feature Name

Releases

Feature Information

Flexible Packet Matching XML Configuration

12.4(6)T

The Flexible Packet Matching XML Configuration feature provides an Extensible Markup Language (XML)-based configuration file for Flexible Packet Matching (FPM) that can be used to define traffic classes and actions (policies) to assist in the blocking of attacks on a network. The XML file used by FPM is called the traffic classification definition file (TCDF).

The TCDF gives you an alternative to the command-line interface (CLI) as a method to define traffic classification behavior. Traffic classification behavior is identical regardless of the method you use.

This feature was introduced in Cisco IOS Release 12.4(6)T.

The following command was introduced by this feature: load classification.

Cisco and the Cisco Logo are trademarks of Cisco Systems, Inc. and/or its affiliates in the U.S. and other countries. A listing of Cisco's trademarks can be found at www.cisco.com/go/trademarks. Third party trademarks mentioned are the property of their respective owners. The use of the word partner does not imply a partnership relationship between Cisco and any other company. (1005R)

Any Internet Protocol (IP) addresses and phone numbers used in this document are not intended to be actual addresses and phone numbers. Any examples, command display output, network topology diagrams, and other figures included in the document are shown for illustrative purposes only. Any use of actual IP addresses or phone numbers in illustrative content is unintentional and coincidental.