Configuring the Link Services Interface with a Configuration Editor

This section contains the following topics:

Configuring MLPPP Bundles and LFI on Serial Links

Figure 31 shows a network topology that is used as an example in this section. In this example, your company's branch office is connected to its main branch using J Series devices R0 and R1. You transmit data and voice traffic on two low-speed 1-Mbps serial links. To increase bandwidth, you configure MLPPP and join the two serial links se-1/0/0 and se-1/0/1 into a multilink bundle lsq-0/0/0.0. Then you configure LFI and CoS on R0 and R1 to enable them to transmit voice packets ahead of data packets.

Figure 31: Configuring MLPPP and LFI on Serial Links

Image g017255.gif

Configuring a multilink bundle on the two serial links increases the bandwidth by 70 percent from approximately 1 Mbps to 1.7 Mbps and prepends each packet with a multilink header as specified in the FRF.12 standard. To increase the bandwidth further, you can add up to 8 serial links to the bundle. In addition to a higher bandwidth, configuring the multilink bundle provides load balancing and redundancy. If one of the serial links fails, traffic continues to be transmitted on the other links without any interruption. In contrast, independent links require routing policies for load balancing and redundancy. Independent links also require IP addresses for each link as opposed to one IP address for the bundle. In the routing table, the multilink bundle is represented as a single interface.

This example uses MLPPP for providing multilink services. For information about configuring MLFR, see Configuring MLFR FRF.15 Bundles and Configuring MLFR FRF.16 Bundles.

You can use the LFI and CoS configurations provided in this example with MLFR FRF.15 and MLFR FRF.16 bundles, too. You can also use the same LFI and CoS configurations for other interfaces, such as on T1 or E1.

To configure MLPPP bundles and LFI, perform the following tasks:

Configuring an MLPPP Bundle

In this example, you create an MLPPP bundle (lsq-0/0/0.0) at the logical unit level of the link services interface (lsq-0/0/0) on J Series devices R0 and R1. Then you add the two serial interfaces se-1/0/0 and se-1/0/1 as constituent links to the multilink bundle. Adding multiple links does not require you to configure and manage more addresses.

To configure an MLPPP bundle on a J Series device:

  1. Navigate to the top of the interfaces configuration hierarchy.
  2. Perform the configuration tasks described in Table 79 on Device R0 and Device R1.
  3. Go on to Enabling Link Fragmentation and Interleaving.

Table 79: Configuring an MLPPP Bundle

Task

CLI Configuration Editor

Navigate to the Interfaces level in the configuration hierarchy. Specify the link services interface to be configured.

From the [edit] hierarchy level, enter

edit interfaces lsq-0/0/0

Configure a logical unit on the lsq-0/0/0 interface and define the family type—for example, Inet.

Configure an IP address for the multilink bundle at the unit level of the link services interface.

Set the appropriate source address for the interface:

  • On R0, enter

    set unit 0 family inet address 10.0.0.10/24

  • On R1, enter

    set unit 0 family inet address 10.0.0.9/24

From the Interfaces level in the configuration hierarchy, specify the names of the constituent links to be added to the multilink bundle—for example, se-1/0/0 and se-1/0/1.

From the [edit] hierarchy level, add the constituent links to the multilink bundle.

  • To add se-1/0/0 to the multilink bundle, enter

    edit interfaces se-1/0/0

  • To add se-1/0/1 to the multilink bundle, enter

    edit interfaces se-1/0/1

Create the multilink bundle by specifying a logical unit on each constituent link and defining it as an MLPPP bundle—for example, lsq-0/0/0.0.

Enter

set unit 0 family mlppp bundle lsq-0/0/0.0

Set the serial options to the same values for both interfaces on R0—se-1/0/0 and se-1/0/1.

Note: In this example, R0 is set as a data circuit-terminating equipment (DCE) device. The serial options are not set for interfaces on R1. You can set the serial options according to your network setup.

  1. On R0, from the [edit] hierarchy level, set serial options for the interface.
    • To set options on se-1/0/0, enter

      edit interfaces se-1/0/0

    • To set options on se-1/0/1, enter

      edit interfaces se-1/0/1

  2. Enter

    set serial-options clocking-mode dce clock-rate 2.0mhz

Enabling Link Fragmentation and Interleaving

To configure link fragmentation and interleaving (LFI), you define the MLPPP encapsulation type and enable fragmentation and interleaving of packets by specifying the following properties—the fragmentation threshold and fragmentation—maps with a no-fragmentation knob mapped to the forwarding class of choice. In this example, a fragmentation threshold of 128 bytes is set on the MLPPP bundle that applies to all traffic on both constituent links, so that any packet larger than 128 bytes transmitted on these links is fragmented.

For more information about LFI, see Link Fragmentation and Interleaving Overview.

To enable LFI:

  1. Navigate to the top of the interfaces configuration hierarchy.
  2. Perform the configuration tasks described in Table 80 on Device R0 and Device R1.
  3. Go on to Defining Classifiers and Forwarding Classes.

Table 80: Enabling LFI

Task

CLI Configuration Editor

Navigate to the Interfaces level in the configuration hierarchy.

Specify the link services intelligent queuing interface for fragmentation.

From the [edit] hierarchy level, enter

edit interfaces lsq-0/0/0

Specify the multilink encapsulation type, enable LFI, and set the fragmentation threshold for the multilink interface.

Fragment Threshold—Set the maximum size, in bytes, for multilink packet fragments—for example, 128. Any nonzero value must be a multiple of 64 bytes. The value can be between 128 and 16320. The default is 0 bytes (no fragmentation).

Enter

set unit 0 encapsulation multilink-ppp fragment-threshold 128

Defining Classifiers and Forwarding Classes

By defining classifiers you associate incoming packets with a forwarding class and loss priority. Based on the associated forwarding class, you assign packets to output queues. To configure classifiers, you specify the bit pattern for the different types of traffic. The classifier takes this bit pattern and attempts to match it to the type of packet arriving on the interface. If the information in the packet’s header matches the specified pattern, the packet is sent to the appropriate queue, defined by the forwarding class associated with the classifier.

In this example, an IP precedence classifier, classify_input, is assigned to all incoming traffic. The precedence bit value in the type of service (ToS) field is assumed to be 000 for all incoming data traffic and 010 for all incoming voice traffic. This classifier assigns all data traffic to Q0 and all voice traffic to Q2, and fragmentation-map maps Q2 to no-fragmentation. On a J Series device, when LFI is enabled, all traffic assigned to Q2 is treated as LFI (voice) traffic. You do not need to assign network control traffic to a queue explicitly, because it is assigned to Q3 by default.

For more information about configuring CoS components, see Junos OS Class of Service Configuration Guide for Security Devices.

To define classifiers and forwarding classes:

  1. Navigate to the top of the interfaces configuration hierarchy.
  2. Perform the configuration tasks described in Table 81 on Device R0 and Device R1.
  3. Go on to Defining and Applying Scheduler Maps.

Table 81: Defining Classifiers and Forwarding Classes

Task

CLI Configuration Editor

Navigate to the Class of service level in the configuration hierarchy.

From the [edit] hierarchy level, enter

edit class-of-service

Configure a behavior aggregate (BA) classifier for classifying packets.

In this example, you specify the default IP precedence classifier, which maps IP precedence bits to forwarding classes and loss priorities.

Enter

edit classifiers inet-precedence classify_input

For the classifier to assign an output queue to each packet, it must associate the packet with a forwarding class.

Assign packets with IP precedence bits 000 to the DATA forwarding class, and specify a low loss priority.

Enter

set forwarding-class DATA loss-priority low code-points 000

Assign packets with IP precedence bits 010 to the VOICE forwarding class, and specify a low loss priority.

Enter

set forwarding-class VOICE loss-priority low code-points 010

Assign each forwarding class one-to-one with the output queues.

  • DATA—Assign to Queue 0.
  • VOICE—Assign to Queue 2.
  • NC (Network Control)—Assign to Queue 3. NC is assigned to Queue 3 by default.

From the [edit class-of-service] hierarchy level, enter

set forwarding-classes queue 0 DATA

set forwarding-classes queue 2 VOICE

set forwarding-classes queue 3 NC

Apply the behavior aggregate classifier to the incoming interface.

  1. From the [edit class-of-service] hierarchy level, enter

    edit interfaces ge-0/0/1

  2. Enter

    set unit 0 classifiers inet-precedence classify_input

Configure fragmentation-map

Enter

set class-of-service fragmentation-maps FM forwarding-class VOICE no-fragmentation

Attach fragmentation-map to lsq

Enter

set class-of-service interfaces lsq-0/0/0 unit unit fragmentation-map

Defining and Applying Scheduler Maps

By defining schedulers you configure the properties of output queues that determine the transmission service level for each queue. These properties include the amount of interface bandwidth assigned to the queue, the size of the memory buffer allocated for storing packets, and the priority of the queue. After defining schedulers you associate them with forwarding classes by means of scheduler maps. You then associate each scheduler map with an interface, thereby configuring the hardware queues and packet schedulers that operate according to this mapping.

In this example, you define and apply scheduler maps as follows:

To define and apply scheduler maps:

  1. Navigate to the top of the interfaces configuration hierarchy.
  2. Perform the configuration tasks described in Table 82 on Device R0 and Device R1.
  3. Go on to Applying Shaping Rates to Interfaces.

Table 82: Defining and Applying Scheduler Maps

Task

CLI Configuration Editor

Navigate to the Interface level in the configuration hierarchy.

From the [edit] hierarchy level, enter

edit interfaces

To configure CoS components for each multilink bundle, enable per-unit scheduling on the interface.

Enter

set lsq-0/0/0 per-unit-scheduler

set se-1/0/0 per-unit-scheduler

set se-1/0/1 per-unit-scheduler

Navigate to the Interfaces level in the Class of Service configuration hierarchy and specify the link services intelligent queuing interface to be configured.

From the [edit class-of-service] hierarchy level, enter

edit interfaces lsq-0/0/0

Define a scheduler map—for example, s_map.

Enter

set unit 0 scheduler-map s_map

Apply the scheduler map to the constituent links of the multilink bundle—for example, se-1/0/0 and se-1/0/1.

  1. From the [edit] hierarchy level, specify the interface to be configured.
    • To apply the scheduler map to se-1/0/0, enter

      edit interfaces se-1/0/0

    • To apply the scheduler map to se-1/0/1, enter

      edit interfaces se-1/0/1

  2. Apply the scheduler map to the logical interface.

    set unit 0 scheduler-map s_map

Associate a scheduler with each forwarding class.

  • DATA—A scheduler associated with the DATA forwarding class.
  • VOICE—A scheduler associated with the VOICE forwarding class.
  • NC—A scheduler associated with the NC forwarding class.

A scheduler receives the forwarding class and loss priority settings, and queues the outgoing packet based on those settings.

From the [edit class-of-service] hierarchy level, enter

set scheduler-maps s_map forwarding-class DATA scheduler DATA

set scheduler-maps s_map forwarding-class VOICE scheduler VOICE

set scheduler-maps s_map forwarding-class NC scheduler NC

Define the properties of output queues for the DATA scheduler:

  • Transmit rate—Specify a percentage of transmission capacity—49.
  • Buffer size—Specify a percentage of total buffer—49.
  • Priority—Do not specify the transmission priority for the DATA scheduler to apply the default setting—low.

For more information about transmit rate and buffer size, see Junos OS Class of Service Configuration Guide for Security Devices.

Enter

set schedulers DATA transmit-rate percent 49

set schedulers DATA buffer-size percent 49

Define the properties of output queues for the VOICE scheduler:

  • Transmit rate—Specify a percentage of transmission capacity—50.
  • Buffer size—Specify a percentage of total buffer—5.
  • Priority—Specify a transmission priority—high.

Enter

set schedulers VOICE transmit-rate percent 50

set schedulers VOICE buffer-size percent  5

set schedulers VOICE priority high

Define the properties of output queues for the NC scheduler:

  • Transmit rate—Specify a percentage of transmission capacity—1.
  • Buffer size—Specify a percentage of total buffer—1.
  • Priority—Specify a transmission priority—high.

Enter

set schedulers NC transmit-rate percent  1

set schedulers NC buffer-size percent  1

set schedulers NC priority high

Applying Shaping Rates to Interfaces

To control the voice traffic latency within acceptable limits, you configure the shaping rate on constituent links of the MLPPP bundle. Shaping rate at the interface level is required only when you enable LFI. To apply shaping rates to interfaces, you have to first enable per-unit scheduling. For information about shaping rates and LFI, see Configuring CoS Components with LFI.

You must configure the shaping rate to be equal to the combined physical interface bandwidth for the constituent links. In this example, the combined bandwidth capacity of the two constituent links—se-1/0/0 and se-1/0/1—is 2 Mbps. Hence, configure a shaping rate of 2 Mbps on each constituent link.

To apply a shaping rate to the constituent links of the multilink bundle:

  1. Navigate to the top of the interfaces configuration hierarchy.
  2. Perform the configuration tasks described in Table 83 on Device R0 and Device R1.
  3. Go on to Verifying the Link Services intelligent queuing Interface Configuration, to verify your configuration.

Table 83: Applying Shaping Rate to Interfaces

Task

CLI Configuration Editor

Navigate to the Class of service level in the configuration hierarchy.

From the [edit] hierarchy level, enter

edit class-of-service

Apply the shaping rate to the constituent links of the multilink bundle—for example, se-1/0/0 and se-1/0/1. The shaping rate specifies the amount of bandwidth to be allocated for this multilink bundle.

  1. Set the shaping rate on both the constituent links:
    • To set the shaping rate for se-1/0/0, enter

      edit interfaces se-1/0/0

    • To set the shaping rate for se-1/0/1, enter

      edit interfaces se-1/0/1

  2. Set the shaping rate:

    set unit 0 shaping-rate 2000000

Configuring MLFR FRF.15 Bundles

J Series devices support Multilink Frame Relay end-to-end (MLFR FRF.15) on the link services intelligent queuing interface lsq-0/0/0.

With MLFR FRF.15, multilink bundles are configured as logical units on the link services intelligent queuing interface, such as lsq-0/0/0.0. MLFR FRF.15 bundles combine multiple permanent virtual circuits (PVCs) into one aggregated virtual circuit (AVC). This process provides fragmentation over multiple PVCs on one end and reassembly of the AVC on the other end. For more information about multilink bundles, see Multilink Bundles Overview.

You can configure LFI and CoS with MLFR in the same way that you configure them with MLPPP. For information about configuring LFI and CoS, see Configuring MLPPP Bundles and LFI on Serial Links.

In this example, you aggregate two T1 links to create an MLFR FRF.15 bundle on two J Series devices—Device R0 and Device R1.

To configure an MLFR FRF.15 bundle:

  1. Navigate to the top of the configuration hierarchy on Device R0 and Device R1.
  2. Perform the configuration tasks described in Table 84.
  3. If you are finished configuring the device, commit the configuration.
  4. Go on to Verifying the Link Services intelligent queuing Interface Configuration, to verify your configuration.

Table 84: Configuring MLFR FRF.15 Bundles

Task

CLI Configuration Editor

Navigate to the Interfaces level in the configuration hierarchy. Specify the link services intelligent queuing interface as an interface to be configured.

From the [edit] hierarchy level, enter

edit interfaces lsq-0/0/0

Configure a logical unit on the lsq-0/0/0 interface, and define the family type—for example, Inet.

Configure an IP address for the multilink bundle on the unit level of the link services intelligent queuing interface.

Set the appropriate source address for the interface:

  • On R0, enter

    set unit 0 family inet address 10.0.0.4/24

  • On R1, enter

    set unit 0 family inet address 10.0.0.5/24

Define the multilink bundle as an MLFR FRF.15 bundle by specifying the Multilink Frame Relay end-to-end encapsulation type.

From the [edit interfaces lsq-0/0/0] hierarchy level, enter

set unit 0 encapsulation multilink-frame-relay-end-to-end

Specify the names of the constituent links to be added to the multilink bundle—for example, t1-2/0/0 and t1-2/0/1.

Define the Frame Relay encapsulation type.

  1. From the [edit] hierarchy level, enter
    • For configuring t1-2/0/0

      edit interfaces t1-2/0/0

    • For configuring t1-2/0/1

      edit interfaces t1-2/0/1

  2. Enter

    set encapsulation frame-relay

Define R0 to be a data circuit-terminating equipment (DCE) device. R1 performs as a data terminal equipment (DTE) device, which is the default with Frame Relay encapsulation.

For more information about DCE and DTE, see Serial Interface Overview.

On R0 only, enter

set dce

On the logical unit level of the interface, specify the data-link connection identifier (DLCI). The DLCI field identifies which logical circuit the data travels over. DLCI is a value from 16 through 1022—for example, 100. (Numbers 1 through 15 are reserved for future use.)

Specify the multilink bundle to which the interface is to be added as a constituent link—lsq-0/0/0.0.

Enter

set unit 0 dlci 100 family mlfr-end-to-end bundle lsq-0/0/0.0

Configuring MLFR FRF.16 Bundles

J Series devices support Multilink Frame Relay (MLFR) user-to-network interface (UNI) network-to-network interface (NNI) (MLFR FRF.16) on the link services intelligent queuing interface lsq-0/0/0.

MLFR FRF.16 configures multilink bundles as channels on the link services intelligent queuing interface, such as lsq-0/0/0:0. A multilink bundle carries Frame Relay permanent virtual circuits (PVCs), identified by their data-link connection identifiers (DLCIs). Each DLCI is configured at the logical unit level of the link services intelligent queuing interface and is also referred as a logical interface. Packet fragmentation and reassembly occur on each virtual circuit. For more information about multilink bundles, see Multilink Bundles Overview.

You can configure LFI and CoS with MLFR in the same way that you configure them with MLPPP. For information about configuring LFI and CoS, see Configuring MLPPP Bundles and LFI on Serial Links.

In this example, you aggregate two T1 interfaces to create an MLFR FRF.16 bundle on two J Series devices—Device R0 and Device R1.

To configure an MLFR FRF.16 bundle:

  1. Navigate to the top of the configuration hierarchy on Device R0 and Device R1.
  2. Perform the configuration tasks described in Table 85.
  3. If you are finished configuring the device, commit the configuration.
  4. Go on to Verifying the Link Services intelligent queuing Interface Configuration, to verify your configuration.

Table 85: Configuring MLFR FRF.16 Bundles

Task

CLI Configuration Editor

Navigate to the Chassis level in the configuration hierarchy.

From the [edit] hierarchy level, enter

edit chassis

Specify the number of multilink frame relay UNI NNI (FRF.16) bundles to be created on the interface. You can specify a number from 1 through 255.

Enter

set fpc 0 pic 0 mlfr-uni-nni-bundles 1

Specify the channel to be configured as a multilink bundle.

From the [edit] hierarchy level, enter

edit interfaces lsq-0/0/0:0

Define the multilink bundle as an MLFR FRF.16 bundle by specifying the Multilink Frame Relay UNI NNI encapsulation type.

Enter

set encapsulation multilink-frame-relay-uni-nni

Define R0 to be a data circuit-terminating equipment (DCE) device. R1 performs as a data terminal equipment (DTE) device, which is the default with Frame Relay encapsulation.

For more information about DCE and DTE, see Serial Interface Overview

On R0 only, enter

set dce

Configure a logical unit on the multilink bundle lsq-0/0/0:0, and define the family type—for example, Inet.

Assign a data link connection identifier (DLCI) to the multilink bundle. The DLCI field identifies which logical circuit the data travels over. DLCI is a value from 16 through 1022—for example, 400. (Numbers 1 through 15 are reserved for future use.)

Assign an IP address to the multilink bundle.

Set the appropriate address for the interface:

  • On R0, enter

    set unit 0 dlci 400 family inet address 10.0.0.10/24

  • On R1, enter

    set unit 0 dlci 400 family inet address 10.0.0.9/24

Create the T1 interfaces that are to be added as constituent links to the multilink bundle—t1-2/0/0 and t1-2/0/1.

Define the Frame Relay encapsulation type.

  1. From the [edit] hierarchy level, enter
    • For configuring t1-2/0/0

      edit interfaces t1-2/0/0

    • For configuring t1-2/0/1

      edit interfaces t1-2/0/1

  2. Enter

    set encapsulation multilink-frame-relay-uni-nni

Specify the multilink bundle to which the interface is to be added as a constituent link—lsq-0/0/0:0.

Enter

set unit 0 family mlfr-uni-nni bundle lsq-0/0/0:0

Configuring CRTP

Compressed Real-Time Transport Protocol (CRTP) is typically used for compressing voice and video packets. You can configure CRTP with LFI on the link services interface of a J Series device.

On the J Series device, CRTP can be configured as a compression device on a T1 or E1 interface with PPP encapsulation, using the link services interface.

For more information about configuring CRTP on a single link, see the Junos Network Interfaces Configuration Guide and the Junos Services Interfaces Configuration Guide.

To configure CRTP on the device:

  1. Navigate to the top of the interfaces configuration hierarchy.
  2. Perform the configuration tasks described in Table 86.
  3. If you are finished configuring the device, commit the configuration.

Table 86: Adding CRTP to a T1 or E1 Interface

Task

CLI Configuration Editor

Navigate to the Interfaces level in the configuration hierarchy.

From the [edit] hierarchy level, enter

edit interfaces interface-name

Select an E1 or T1 interface—for example, t1–1/0/0.

Set PPP as the type of encapsulation for the physical interface.

  1. Enter

    set encapsulation ppp

  2. Enter

    edit unit 0

Add the link services intelligent queuing interface, lsq-0/0/0.0, to the physical interface.

Enter

set compression-device lsq-0/0/0.0

Add the link services intelligent queuing interface, lsq-0/0/0, to the device.

From the [edit interfaces] hierarchy level, enter

edit interfaces lsq-0/0/0 unit 0

Configure the link services intelligent queuing interface, lsq-0/0/0, properties.

F-max period—Maximum number of compressed packets allowed between transmission of full headers. It has a range from 1 to 65535.

Maximum and Minimum—UDP port values from 1 to 65536 reserve these ports for RTP compression. CRTP is applied to network traffic on ports within this range. This feature is applicable only to voice services interfaces.

Enter

set compression rtp f-max-period 2500 port minimum 2000 maximum 64009

Copyright© 1999-2010 Juniper Networks, Inc. All rights reserved.