All Cisco-Network Study Notes: Frame Relay Fragmentation

Frame Relay Fragmentation

Frame Relay Forum IA 12, or FRF.12, defines a accepted adjustment of assuming LFI over a Frame

Relay PVC. Cisco IOS supports two methods for configuring FRF.12. The bequest FRF.12

configuration requires FRTS to be configured, and requires a queuing apparatus to be activated to the

shaped packets. (Example 14-7 in Affiliate 14 shows an FRTS map-class shape-with-LLQ

command that shapes and applies LLQ.)

Figure 15-5 shows the all-embracing argumentation of how FRF.12 interleaves packets application LFI, back configured

using bequest FRF.12 configuration. IOS creates a 2-queue software queuing arrangement on the

physical interface. Any packets abrogation the FRTS LLQ go into the “high” Dual FIFO queue, with

the packets and bits from added queuing activity into the Dual FIFO “normal” queue. On the

interface, IOS treats the Dual FIFO chain as a antecedence queue, which causes interleaving.

Frame Relay Agreement 523

Figure 15-5 Interface Dual FIFO Queues with FRTS Plus FRF.12

To configure bequest FRF.12, the frame-relay fragment admeasurement command is added to the FRTS map class

on both ends of the VC. For example, in Archetype 14-7, the frame-relay fragment 120 command

could be added to the shape-with-LLQ map class, with the aforementioned agreement on the router on the

other end of the VC, to accredit FRF.12. Note that because breach of any affectionate implies that an

additional breach attack is used, breach charge be added on both ends of the articulation or VC.

The additional adjustment of configuring FRF.12 is alleged Frame Relay Breach at the Interface,

and was added to Cisco IOS Software Release 12.2(13)T. This adjustment does not crave FRTS; the

frame-relay fragment command artlessly sits anon on the concrete interface. If no queuing tool

is configured on the interface, the router creates Dual FIFO queuing on the interface, interleaving

all non-fragmented packets amid bits of added packets. Optionally, agreement of a

queuing apparatus that has a PQ affection (for example, LLQ) can be acclimated instead, causing packets in the

PQ to be anon interleaved. Archetype 15-3 shows a sample agreement application the same

router, R1, from the aboriginal two examples in this chapter. In this case, FRF.12 has been enabled on

s0/0/0, with a fragment admeasurement of 120.

NOTE All packets can be fragmented, but Cisco appropriately suggests allotment a fragment

size so that the packets about placed into the LLQ PQ will not be fragmented. Alone packets

from the abstraction LLQ are placed into the Dual FIFO interface aerial queue, and alone those

packets are interleaved.

Example 15-2 FRF.12 on the Interface—Configuration

! No FRTS agreement exists—simply the frame-relay fragment 120 end-to-end

! command. Note that LLQ is not enabled in this case, so nonfragmented packets

! will be interleaved application Dual FIFO.

R1# appearance run int s 0/0/0

interface Serial0/0/0

encapsulation frame-relay

frame-relay fragment 120 end-to-end

! Next, breach stats are listed.

continues

Class 1 – LLQ

Shaping Queues Created

by LLQ Configuration

on a Single VC

FRF.12 Dual FIFO Queues

Interleaving

Attempt to Not

Fragment

Shape

Class 2 – Non-

LLQ

Class 3 – Non-

LLQ

Normal

Queue, with

Fragments

High Queue

TX Ring

524 Affiliate 15: Frame Relay

Table 15-5 summarizes the key credibility apropos both styles of FRF.12 configuration.

In accession to FRF.12, Cisco IOS supports two added methods of LFI over Frame Relay, including

FRF.11-c. This breach adjustment works alone on Articulation over Frame Relay (VoFR) VCs. With

this tool, articulation frames are never fragmented, and articulation frames are consistently interleaved, without

requiring any accurate queuing tool. Once a VoFR VC has been configured, the LFI configuration

is identical to the bequest appearance of FRF.12 configuration.

The aftermost blazon of FR LFI uses MLP over Frame Relay; it additionally happens to be the alone advantage for Frame

Relay-to-ATM Service Interworking. MLP over FR uses PPP headers instead of the Cisco or

RFC 2427 attack apparent in Figure 15-2, thereby enabling abounding PPP appearance accurate by the

PPP headers. MLP and LFI agreement would artlessly charge to be added to that agreement to

achieve LFI.

R1# appearance frame-relay fragment 104

interface dlci frag-type admeasurement in-frag out-frag dropped-frag

Se0/0/0.14 104 end-to-end 120 2759 2762 0

! The appearance queueing command (yes, IOS misspells it) lists statistics for the Dual

! FIFO queuing arrangement added to the interface back FRF.12 is configured.

R1# appearance queueing int s0/0/0

Interface Serial0/0/0 queueing strategy: priority

Output chain appliance (queue/count)

high/354 medium/0 normal/1422 low/0

Table 15-5 Comparing Bequest and Interface FRF.12

Feature Bequest FRF.12 FRF.12 on the Interface

Requires FRTS? Yes No

Interleaves by agriculture Dual FIFO interface aerial queue

from a abstraction PQ?

Yes No

Interleaves by application either Dual FIFO or a configured

LLQ policy-map on the concrete interface.

No Yes

Config approach for the frame-relay fragment command. map-class Concrete interface

Example 15-2 FRF.12 on the Interface—Configuration (Continued)