cc/td/doc/product/cable/svc_ctrl/scopsys
hometocprevnextglossaryfeedbacksearchhelp
PDF

Table Of Contents

Configuring the Line Interface

How to Configure the Line Interfaces

Line Interfaces

How to Configure the Gigabit Ethernet Line Interfaces

How to Configure the Fast Ethernet Line Interfaces

How to Configure Tunneling Protocols

About Tunneling Protocols

How to Select the Tunneling Mode

How to Display the Tunneling Configuration

How to Configure VLAN Translation

About VLAN Translation

VLAN Translation Features and Limitations

How to Set the VLAN Translation Constant

How to Disable VLAN Translation

How to Monitor VLAN Translation

How to Configure Traffic Rules and Counters

Information About Traffic Rules and Counters

How to Configure Traffic Counters

How to Configure Traffic Rules

How to Manage Traffic Rules and Counters

How to Configure TOS Marking

About TOS Marking

How to Enable and Disable TOS Marking

How to Modify the TOS Table

How to Count Dropped Packets

About Counting Dropped Packets

How to Disable the Hardware Packet Drop


Configuring the Line Interface


This module describes how to configure the physical line interfaces (ports) as well how to configure those interfaces for tunneling, VLAN translation, TOS marking, and traffic rules.

How to Configure the Line Interfaces 

How to Configure Tunneling Protocols 

How to Configure VLAN Translation 

How to Configure Traffic Rules and Counters 

How to Configure TOS Marking 

How to Count Dropped Packets 

How to Configure the Line Interfaces

Line Interfaces 

How to Configure the Gigabit Ethernet Line Interfaces 

How to Configure the Fast Ethernet Line Interfaces 

Line Interfaces

The Line Interfaces (Subscriber and Network) are used to connect the SCE platform to the network. See the description of network topologies in the Topology section of the Cisco SCE 2000/SCE 1000 Installation and Configuration Guides .

The SCE 1000 2xGBE and the SCE 2000 4xGBE have Gigabit Ethernet line interfaces. You should configure auto-negotiatefor these interfaces.

The SCE 2000 4/8x FE has Fast Ethernet line interfaces. You should configure the speedand duplexfor these interfaces.

How to Configure the Gigabit Ethernet Line Interfaces


Note The maximum packet size supported by the SCE platform is 1600 bytes


SUMMARY STEPS

1. At the SCE# prompt, type configure , and press Enter.

2. At the SCE(config)# prompt, type interface GigabitEthernet 0/ portnumber, and press Enter

3. , where portnumberis the number of the selected port (1-4).

4. At the SCE(config if)# prompt, type auto-negotiate , and press press EnterAt the SCE(config if)# prompt, type exit and press Enter.

5. Repeat steps 2 and 3 for the remaining Gigabit Ethernet interfaces.

DETAILED STEPS


Step 1 At the SCE# prompt, type configure , and press Enter.

Enters Global Configuration mode.

Step 2 At the SCE(config)# prompt, type interface GigabitEthernet 0/ portnumber, and press Enter

Step 3 , where portnumberis the number of the selected port (1-4).

Enters Interface Configuration mode for the selected GBE interface.

Step 4 At the SCE(config if)# prompt, type auto-negotiate , and press press EnterEnables auto-negotiation.

Step 5 At the SCE(config if)# prompt, type exit and press Enter.

Exits to global configuration mode, from which you can access a different Gigabit Ethernet interface.

Step 6 Repeat steps 2 and 3 for the remaining Gigabit Ethernet interfaces.

Auto-negotiation must enabled separately and explicitly for each Gigabit Ethernet interface.


How to Configure the Fast Ethernet Line Interfaces

Note that both sides of the FE link (both the SCE 2000 4/8xFE and the remote device) should have the same configuration. Use either of the following two configuration options:

Autonegotiation = ON

Autonegotiation = ON, speed = 100

SUMMARY STEPS

1. At the SCE# prompt, type configure , and press Enter.

2. At the SCE(config)# prompt, type interface FastEthernet 0/ portnumber, and press Enter, where portnumberis the number of the selected port (1-4).

3. At the SCE(config if)# prompt, type duplex auto|full|half, and press Enter,

4. At the SCE(config if)# prompt, type speed auto|100, and press Enter,

5. At the SCE(config if)# prompt, type exit and press Enter.

6. Repeat steps 2 through 5 for the remaining Fast Ethernet interfaces.

DETAILED STEPS


Step 1 At the SCE# prompt, type configure , and press Enter.

Enters Global Configuration mode.

Step 2 At the SCE(config)# prompt, type interface FastEthernet 0/ portnumber, and press Enter, where portnumberis the number of the selected port (1-4).

Enters Interface Configuration mode for the selected FE interface.

Step 3 At the SCE(config if)# prompt, type duplex auto|full|half, and press press EnterConfigures the duplex mode for the selected FE interface.

Step 4 At the SCE(config if)# prompt, type speed auto|100, and press press EnterConfigures the speed for the selected FE interface.

Step 5 At the SCE(config if)# prompt, type exit and press Enter.

Exits to global configuration mode, from which you can access a different Fast Ethernet interface.

Step 6 Repeat steps 2 through 5 for the remaining Fast Ethernet interfaces.

Speed and duplex mode must enabled separately and explicitly for each Fast Ethernet interface.


How to Configure Tunneling Protocols

About Tunneling Protocols 

How to Select the Tunneling Mode 

How to Display the Tunneling Configuration 

About Tunneling Protocols

Tunneling technology is used across various telecommunications segments to solve a wide variety of networking problems. The SCE platform is designed to recognize and process various tunneling protocols in several ways. The SCE platform is able to either ignore the tunneling protocols ("skip" the header) or treat the tunneling information as subscriber information ("classify"). A special case of classification by tunneling information is MPLS/VPN with private IP support .

The following table shows the support for the various tunneling protocols (the default behavior for each protocol is in bold):

Table 6-1 Tunneling Protocol Summary

Protocol

Supported handling

Mode name

L2TP

Ignore tunnel

IP-tunnel L2TP skip

Don't ignore tunnel - classify by external IP

No IP-Tunnel

VLAN

Ignore tunnel

VLAN symmetric skip

Ignore tunnel - asymmetric

VLAN a-symmetric skip

VLAN tag as subscriber

VLAN symmetric classify

MPLS

Ignore tunnel (inject unlabeled)

MPLS Traffic-engineering skip

Ignore tunnel (inject labeled)

MPLS VPN skip

MPLS L3 VPN as subscriber

MPLS VPN auto-learn


When the tunneling information is ignored, the subscriber identification is the subscriber IP of the IP packet carried inside the tunnel.

L2TP is an IP-based tunneling protocol, therefore the system must be specifically configured to recognize the L2TP flows, given the UDP port used for L2TP. The SCE platform can then skip the external IP, UDP, and L2TP headers, reaching the internal IP, which is the actual subscriber traffic. If L2TP is not configured, the system treats the external IP header as the subscriber traffic, thus all the flows in the tunnel are seen as a single flow.

A single VLAN tag is supported per packet (no QinQ support). MPLS labels are supported up to a maximum of 15 labels per packet.

Subscriber classification by VLAN tag is supported only in symmetric VLAN environments - i.e. where the upstream and downstream tags are identical.


Note All subscribers with tunnel mappings must be cleared in order to change the tunneling mode. If the connection with the SM is down, use the no subscriber all with-tunnel-mappings CLI command (see How to Remove Subscribers with Tunnel Mappings ).


Use these commands to configure tunneling:

ip tunnel

vlan

mpls

L2TP identify-by

How to Select the Tunneling Mode

How to Configure IP Tunnels 

How to Configure the VLAN Environment 

How to Configure the MPLS Environment 

How to Configure the L2TP Environment 

How to Configure IP Tunnels

By default, IP tunnel recognition is disabled. Use this command to configure recognition of L2TP tunnels and skipping into the internal IP packet.

An IP tunnel is mutually exclusive with tunnel-based classification.


Step 1 From the SCE(config if)# prompt, type ip tunnel L2TP skip and press Enter.

Enables IP tunnel mode

To disable IP tunnels, use the following command:

no ip tunnel


How to Configure the VLAN Environment

Use this command to configure the VLAN environment.

Options 

Configuring the VLAN Environment: Example 

Options

There are three options:

symmetric classify

symmetric skip (default)

a-symmetric skip

Symmetric environment refers to an environment in which the same VLAN tags are used for carrying a transaction in the upstream and downstream directions.

Setting the mode to classify means that subscriber and flow classification will use the VLAN tag. Using VLAN classification is mutually exclusive with other tunnel-based classification or IP tunnels.

An a-symmetric environment is an environment in which the VLAN tags might not be the same in the upstream and downstream directions.

The SCE platform is configured by default to work in symmetric environments. A specific command should be used to allow correct operation of the SCE platform in asymmetric environments and instruct it to take into consideration that the upstream and downstream of each flow has potentially different VLAN tags.


Note Using the a-symmetric skip value incurs a performance penalty.



Step 1 From the SCE(config if)# prompt, type vlan {symmetric classify | symmetric skip |a-symmetric skip} and press Enter.

Specify the desired VLAN mode.


Configuring the VLAN Environment: Example

The following example selects symmetric skip VLAN tunnel environment.

SCE(config if)#vlan symmetric skip

How to Configure the MPLS Environment

Use this command to set the MPLS environment.

Options 

Configuring the MPLS Environment: Example 

Options

The following options are available:

traffic-engineering skip (default)—Ignore tunnel (inject unlabeled)

VPN skip —Ignore tunneled (inject labeled)

VPN auto-learn —MPLS L3 VPN as subscriber

When this option is configured, both ip-tunneland VLANmust be set to their default values.

Use the VPN keyword when the labels are mandatory in the traffic, otherwise use traffic-engineering (default).

Note that using the VPN value incurs a performance penalty.

The MPLS VPN auto-learn option is required in an MPLS/VPN environment.


Step 1 From the SCE(config if)# prompt, type mpls {traffic-engineering skip|vpn skip|vpn auto-learn} and press Enter.

Specify the desired MPLS mode.


Configuring the MPLS Environment: Example

The following example selects the MPLS/VPN tunnel environment.

SCE(config if)#mpls vpn auto-learn

How to Configure the L2TP Environment

Use this command to set the port number that the LNS and LAC use for L2TP tunnels.

Options

The following option is available:

portnumber —The port number that the LNS and LAC use for L2TP tunnels.

Default port# = 1701


Step 1 From the SCE(config if)# prompt, type L2TP identify-by port-number portnumberand press Enter.

Enables privileged EXEC mode.


Note If external fragmentation exists in the L2TP environment, it is required to configure a Traffic Rule (see How to Configure Traffic Rules and Counters ) that bypasses all IP traffic targeted to either the LNS or LAC IP address. This will make sure that any packets not having the L2TP port indication (i.e. non-first fragments) will not require handling by the traffic processors.



How to Display the Tunneling Configuration


Step 1 From the SCE# prompt, type show interface linecard 0 MPLS|VLAN|L2TP|IP-tunnel and press Enter.

Displays the current configuration for the specified tunnel option.


How to Configure VLAN Translation

About VLAN Translation 

VLAN Translation Features and Limitations 

How to Set the VLAN Translation Constant 

How to Disable VLAN Translation 

How to Monitor VLAN Translation 

About VLAN Translation

Some topologies require the SCE platform to be able to translate between different VLAN tags.

The following drawing illustrates an example of such a system, in which one router acts as a dispatcher, forwarding traffic and performing load balancing between two SCE 2000 platforms.

Figure 6-1 VLAN Translation

In this example, traffic enters the router via the access ports; it is forwarded to an EtherChannel, which is configured as a trunk, and enters the SCE 2000 platforms.

As can be seen from this drawing, the subscriber side VLAN tags must be different from those on the network side, or the router will simply forward the traffic to the opposite port. This can be supported very simply by having the SCE platform replace the VLAN tags according to a preset configuration.

VLAN Translation Features and Limitations

Features

Configuration of an increment or decrement constant.

Configuration of the constant is global for the line card.

The configured operation (either increment or decrement) is applied to the network side.

The subscriber side automatically performs the opposite operation. That is, if the VLAN is incremented by X on the network side, it is decremented by X on the subscriber side.

VLAN tagged packets are changed (incremented or decremented) before transmission.

Non-tagged packet are not changed.

This feature allows seamless processing with non-VLAN traffic.

Limitations

LIC Bypass not supported - Translation is done in the transmission. Therefore, in LIC bypass, where there is no transmission, there is also no translation.

This means that in general, installations using the VLAN translation feature should rely on cutoff on failure and at upgrade (use redundant SCE platform).

STP hazard - VLAN translation may interfere with Spanning Tree Protocol. This should be taken in consideration when deploying the solution.

The maximum offset that can be configured is 2047. Note that there is no protection for wraparound.

How to Set the VLAN Translation Constant

Use this command to define the VLAN translation constant. Make sure that the same VLAN translation constant is configured for all SCE platforms in the system.

Options 

Setting the VLAN Translation Constant: Example 

Options

The following options are available:

increment | decrement —Keywords indicating whether to increment or decrement the VLAN by the specified value.

value — Integer value by which the VLAN is to incremented or decremented.

The configured translation is applied to the network port side. The reverse operation is performed at the subscriber side.

For example, if "increment 5" is defined, at the network port the VLAN is incremented by 5, and at the subscriber port the VLAN is decremented by 5.

In this case, the network side VLAN tags might be 105, 205, 305, and the subscriber side the VLAN tags would then be 100, 200, 300.

Default = 0

Maximum = 2047 (Note that there is no protection for wraparound of the VLAN value.)


Step 1 From the SCE(config if)# prompt, type vlan translation increment|decrement value valueand press Enter.

Sets the VLAN translation constant.


Setting the VLAN Translation Constant: Example

The following example sets the translation constant to 10, decremented at the network side.

SCE(config if)#vlan translation decrement value 10

How to Disable VLAN Translation


Step 1 From the SCE# prompt, type no vlan translation and press Enter

Disables VLAN translation.


How to Monitor VLAN Translation


Step 1 From the SCE# prompt, type show interface linecard 0 vlan translation and press Enter.

Displays current VLAN translation configuration.


How to Configure Traffic Rules and Counters

Information About Traffic Rules and Counters 

How to Configure Traffic Counters 

How to Configure Traffic Rules 

How to Manage Traffic Rules and Counters 

Information About Traffic Rules and Counters

What are Traffic Rules and Counters? 

Traffic Rules 

Traffic Counters 

What are Traffic Rules and Counters?

Traffic rules and counters may be configured by the user. This functionality enables the user to define specific operations on the traffic flowing through the SCE Platform, such as blocking or ignoring certain flows or counting certain packets. The configuration of traffic rules and counters is independent of the application loaded by the SCE platform, and thus is preserved when the application being run by the SCE platform is changed.

Possible uses for traffic rules and counters include:

Enabling the user to count packets according to various criteria. Since the traffic counters are readable via the SCE SNMP MIB, these might be used to monitor up to 32 types of packets, according to the requirements of the installation.

Ignoring certain types of flows. When a traffic rules specifies an "ignore" action, packets matching the rule criteria will not open a new flow, but will pass through the SCE platform without being processed. This is useful when a particular type of traffic should be ignored by the SCE platform.

Possible examples include ignoring traffic from a certain IP range known to require no service, or traffic from a certain protocol.

Blocking certain types of flows. When a traffic rules specifies a "block" action, packets matching the rule criteria (and not belonging to an existing flow) will be dropped and not passed to the other interface. This is useful when a particular type of traffic should be blocked by the SCE platform.

Possible examples include performing ingress source address filtering (dropping packets originating from a subscriber port whose IP address does not belong to any defined subscriber-side subnet), or blocking specific ports.

It should be noted that using traffic rules and counters does not affect performance. It is possible to define the maximum number of both traffic rules and counters without causing any degradation in the SCE platform performance.

Traffic Rules

A traffic rule specifies that a defined action should be taken on packets processed by the SCE Platform that meet certain criteria. The maximum number of rules is 128, which includes not only traffic rules configured via the SCE platform CLI, but also any additional rules configured by external management systems, such as SCA BB. Each rule is given a name when it is defined, which is then used when referring to the rule.

Packets are selected according to user-defined criteria, which may be any combination of the following:

IP address — A single address or a subnet range can be specified for each of the line ports (Subscriber / Network).

Protocol — TCP/UCP/ICMP/IGRP/EIGRP/IS-IS/OSPF/Other

TCP/UDP Ports — A single port or a port range can be specified for each of the line ports (Subscriber / Network). Valid for the TCP/UDP protocols only.

Direction (Upstream/Downstream) (TCP only).

The possible actions are:

Count the packet by a specific traffic counter

Block the packet (do not pass it to the other side)

Ignore the packet (do not provide service for this packet — No bandwidth metering, transaction reporting etc. is done)

Quick-forward the packet with service — forward delay-sensitive packets through the fast path while maintaining serviceability for these packets

Quick-forward the packet with no service — forward delay-sensitive packets through the fast path with no service provided for these packets

Block and Ignore actions affect only packets that are not part of an existing flow.

Note that Block and Ignore are mutually exclusive. However, blocked or ignored packets can also be counted.

It is possible for a single packet to match more that one rule (The simplest way to cause this is to configure two identical rules with different names). When this happens, the system operates as follows:

Any counter counts a specific packet only once. This means that:

If two rules specify that the packet should be counted by the same counter, it is counted only once.

If two rules specify that the packet should be counted by different counters, it is counted twice, once by each counter.

Block takes precedence over Ignore — If one rule specifies Block , and another rule specifies Ignore , the packet is blocked.

Traffic Counters

Traffic counters count the traffic as specified by the traffic rules. The maximum number of counters is 32. Each counter is given a name when it is defined, which is then used when referring to the counter.

A traffic counter can be configured in one of two ways:

Count packets — the counter is incremented by 1 for each packet it counts.

Count bytes — the counter is incremented by the number of bytes in the packet for each packet it counts.

How to Configure Traffic Counters

A traffic counter must be created before it can be referenced in a traffic rule. Use the following commands to create and delete traffic counters.

How to Create a Traffic Counter 

How to Delete a Traffic Counter 

How to Delete all Existing Traffic Counters 

How to Create a Traffic Counter

Options

The following options are available:

name —The name of the counter

Count packets — the counter is incremented by 1 for each packet it counts.

Count bytes — the counter is incremented by the number of bytes in the packet for each packet it counts.


Step 1 SFrom the SCE(config if)# prompt, type traffic-counter name namecount-bytes|count-packets and press Enter.

Adds a traffic counter with the specified name and counting mode..


How to Delete a Traffic Counter


Step 1 From the SCE(config if)# prompt, type no traffic-counter name nameand press Enter.

Note that a traffic counter cannot be deleted if it is used by any existing traffic rule.


How to Delete all Existing Traffic Counters


Step 1 From the SCE(config if)# prompt, type no traffic-counter all and press Enter.

Removes all traffic counters.

Note that a traffic counter cannot be deleted if it is used by any existing traffic rule.


How to Configure Traffic Rules

Use the following commands to create and delete traffic rules.

How to Create a Traffic Rule 

How to Delete a Traffic Rule 

How to Delete all Traffic Rules 

How to Create a Traffic Rule

Options 

Configuring Traffic Rules: Examples 

Options

The following options are available:

IP specification:

all|([all-but] (ip-address|ip-range))

ip-addressis a single IP address in dotted-decimal notation, such as 10.1.2.3

ip-rangeis an IP subnet range, in the dotted-decimal notation followed by the number of significant bits, such as 10.1.2.0/24.

Use the all-butkeyword to exclude the specified IP address or range of IP addresses

protocol:

Any one of the following protocols:

TCP/UCP/ICMP/IGRP/EIGRP/IS-IS/OSPF/Other

tunnel id specification:

all|([all-but] tunnel id)

tunnel id is an 8-bit Hex value range, in the format '(HEX) Tunnel-id' or '(HEX) MinTunnelId:(HEX) MaxTunnelId', which reflects the lower eight bits of the VLAN tag.

Tunnel-ID-based rules can only be used in " VLAN symmetric classify " mode (see How to Configure the VLAN Environment, and only when tunnel id mode is enabled.

Use the traffic-rule tunnel-id-modecommand.

Note that the VLAN tag itself is a 12-bit value, and therefore aliasing of the lower 8 bits can occur, depending on the VLAN tags used.

direction:

Any of the following:

upstream/downstream/both

traffic-counter:

Either of the following:

name <name of an existing traffic counter>— Packets meeting the criteria of the rule are to be counted in the specified counter. If a counter name is defined, the "count" action is also defined implicitly. The keyword namemust appear as well as the actual name of the counter.

none — If noneis specified, then an action must be explicitly defined via the action option.

action: (not required if the action is count only)

One of the following:

block — Block the specified traffic

ignore — Bypass the specified traffic; traffic receives no service

quick-forwarding — Forward delay-sensitive packets through the fast path while maintaining serviceability for these packets

quick-forwarding-ignore — Forward delay-sensitive packets through the fast path with no service provided for these packets


Step 1 From the SCE(config if)# prompt, type traffic-rule name nameIP-addresses (all|(subscriber-side <IP specification> network-side <IP specification>)) protocol protocol[tunnel-id tunnel-id specification] direction directiontraffic-counter <traffic-counter> [action action]

Enables privileged EXEC mode.


Configuring Traffic Rules: Examples

Example 1 

Example 2 

Example 3 

Example 1

This example creates the following traffic rule:

Name = rule1

IP addresses: subscriber side = all IP addresses, network side = 10.10.10.10 only

Protocol = other

Direction = both

Traffic counter = counter1

The only action performed will be counting

SCE(config if)# traffic-rule rule1 IP-addresses subscriber-side all network-side 10.10.10.10 protocol other direction both traffic-counter name counter1

Example 2

This example creates the following traffic rule:

Name = rule2

IP addresses: subscriber side = all IP addresses, network side = all IP addresses EXCEPT the subnet 10.10.10.0/24

Protocol = TCP

Tunnel id = all

Direction = downstream

Traffic counter = counter2

Action = Block

The actions performed will be counting and blocking

The first command enables tunnel id mode.

SCE(config if)#traffic-rule tunnel-id-mode SCE(config if)# traffic-rule rule2 IP-addresses subscriber-side all network-side all-but 10.10.10.0/24 protocol tcp tunnel-id all direction downstream traffic-counter name counter2 action block

Example 3

This example creates the following traffic rule:

Name = rule3

IP addresses: all

Protocol = IS-IS

Direction = upstream

Traffic counter = none

Action = ignore (required since traffic-counter = none)

The only action performed will be Ignore .

SCE(config if)# traffic-rule rule3 IP-addresses all protocol IS-IS direction upstream traffic-counter none action ignore

How to Delete a Traffic Rule


Step 1 From the SCE(config if)# prompt, type no traffic-rule name nameand press Enter.

Removes the specified traffic rule.


How to Delete all Traffic Rules


Step 1 From the SCE(config if)# prompt, type no traffic-rule all and press Enter.

Removes all existing traffic rules.


How to Manage Traffic Rules and Counters

Use these commands to display existing traffic rule configuration, as well as traffic counter configuration (packets/bytes and the name of the rule using the counter) and traffic counter value.

You can also reset a specific counter or all counters.

How to View a Specified Traffic Rule 

How to View all Traffic Rules 

How to View a Specified Traffic Counter 

How to View all Traffic Counters 

How to Reset a Specified Traffic Counter 

How to Reset all Traffic Counters 

How to View a Specified Traffic Rule


Step 1 From the SCE# prompt, type show interface linecard 0 traffic-rule name rule-nameand press Enter.

Displays the configuration of the specified traffic rule.


How to View all Traffic Rules


Step 1 From the SCE# prompt, type show interface linecard 0 traffic-rule all and press Enter.

Displays the configuration of all existing traffic rules.


How to View a Specified Traffic Counter

Viewing a Traffic Counter: Example 


Step 1 From the SCE# prompt, type show interface linecard 0 traffic-counter name counter-nameand press Enter.

Displays the value of the specified counter and lists the traffic rules that use it.


Viewing a Traffic Counter: Example

The following example displays information for the traffic counter "cnt".

SCE# show interface linecard 0 traffic-counter name cnt Counter 'cnt' value: 0 packets. Rules using it: None.

How to View all Traffic Counters


Step 1 From the SCE# prompt, type show interface linecard 0 traffic-counter all and press Enter.

Displays the value of the each counter and lists the traffic rules that use it.


Viewing the Traffic Counters: Example

The following example displays information for all existing traffic counters.

SCE# show interface linecard 0 traffic-counter all Counter 'cnt' value: 0 packets. Rules using it: None. Counter 'cnt2' value: 0 packets. Rules using it: Rule2. 2 counters listed out of 32 available.

How to Reset a Specified Traffic Counter


Step 1 From the SCE# prompt, type clear interface linecard 0 traffic-counter name counter-nameand press Enter.

Displays the value of the specified counter and lists the traffic rules that use it.


How to Reset all Traffic Counters


Step 1 From the SCE# prompt, type clear interface linecard 0 traffic-counter all and press Enter.

Displays the value of the each counter and lists the traffic rules that use it.


How to Configure TOS Marking

About TOS Marking 

How to Enable and Disable TOS Marking 

How to Modify the TOS Table 

About TOS Marking

The SCE platform TOS marking feature enables marking the TOS field in the IP header of each packet according to two applicative attributes of the packet: its Class (class of service) and its Color (reflects the packet's level of compliance to its relevant bandwidth limitations, where applicable). The actual TOS value set in the IP header is determined according to the configurable TOS table, based on the Class and Color. The default values in the TOS table are based on the Diffserv standard.


Note The first few TCP packets (connection establishment) are associated and marked with a default AF4 class that is mapped to the AF-4 queue and are marked accordingly. This occurs because the SCE platform transmits the first few packets before classifying the flow and identifying the application or service.


How to Enable and Disable TOS Marking

How to Enable TOS Marking 

How to Disable TOS Marking 

How to Enable TOS Marking


Step 1 From the SCE(config if)# prompt, type ToS-marking mode diffserv and press Enter.

Enables TOS marking.


How to Disable TOS Marking


Step 1 From the SCE(config if)# prompt, type no ToS-marking mode diffserv and press Enter.

Disables TOS marking.


How to Modify the TOS Table

Options 

Modify the TOS Table: Example 

Options

The following options are available:

class — the applicative class of the packet (BE, AF1, AF2, AF3, AF4, EF),

color — the applicative color (green, red or any)

value —value to be assigned to the packet (value set to the IP TOS field)

The value parameter must be in hexadecimal format in the range 0x0 to 0x3f .


Step 1 From the SCE(config if)# prompt, type ToS-marking set-table-entry class classcolor colorvalue value and press Enter.


Modify the TOS Table: Example

The following example sets a TOS marking table entry.

SCE(config if)#ToS-marking set-table-entry class AF3 color green value 0x24

How to Count Dropped Packets

About Counting Dropped Packets 

How to Disable the Hardware Packet Drop 

About Counting Dropped Packets

By default, the SCE platform hardware drops red packets (packets that are marked to be dropped due to BW control criteria). However, this presents a problem for the user who needs to know the number of dropped packets per service. To be able to count dropped packets per service, the traffic processor must see all dropped packets for all flows. However, if the hardware is dropping red packets, the traffic processor will not be able to count all dropped packets and the user will not get proper values on the relevant MIB counters.

The user can disable the drop-red-packets-by-hardware mode. This allows the application to access existing per-flow counters. The application can then retrieve the number of dropped packets for every flow and provide the user with better visibility into the exact number of dropped packets and their distribution.

Note that counting all dropped packets has a considerable effect on system performance, and therefore, by default, the drop-red-packets-by-hardware mode is enabled.

How to Disable the Hardware Packet Drop

Use this command to disable the drop-red-packets-by-hardware mode, enabling the software to count all dropped packets.

By default hardware packet drop is enabled.


Note Disabling this feature may have both delay and performance implications.



Step 1 From the SCE(config if)# prompt, type no accelerate-packet-drops and press Enter.

Disables hardware packet drop.

To enable hardware packet drop, use the following command:

accelerate-packet-drops



hometocprevnextglossaryfeedbacksearchhelp

Posted: Wed May 30 08:43:51 PDT 2007
All contents are Copyright © 1992--2007 Cisco Systems, Inc. All rights reserved.
Important Notices and Privacy Statement.