8.2 Configuring Frame Relay
8.2.2 Frame Relay encapsulation and mapping
There are two possible Frame Relay encapsulations. The Cisco encapsulation has a four-byte header, with two bytes for the DLCI and two bytes to identify the packet type. The IETF standard is in accordance with RFCs 1294 and 1490 and it uses a two-byte header, as shown in the Figure. To configure Frame Relay on a Cisco router to interoperate with another vendor's equipment, IETF encapsulation and Frame Relay map entries (per protocol) are used for more flexibility; if backward compatibility and interoperability are required, then this is the preferred method.

Regarding DLCI-to-Layer 3 mapping, some administrators let inverse ARP discover the DLCIs, and then manually map the Layer 3 addresses to DLCI numbers with the frame-relay map command. This strategy makes it easier for support staff to gain information on the Frame Relay network.

If the Cisco encapsulation is configured on a serial interface, then that encapsulation applies to all VCs on that serial interface, with the following caveat. If some of the equipment at the destination is non-Cisco equipment and some is Cisco equipment, the solution is to configure the Cisco encapsulation on the interface and selectively configure IETF encapsulation per DLCI (or vice versa). Because the default encapsulation is Cisco, you would not have to reference it explicitly in the encapsulation frame-relay command:

router(config-if)#encapsulation frame-relay

router(config-if)#frame-relay map ip 192.1.1.7 73 ietf

If you use dynamic address mapping, Frame Relay inverse ARP requests next-hop protocol addresses based on a VC's DLCI address. The router then updates its mapping table and uses the information in the table to route outgoing traffic. Dynamic address mapping is enabled by default for all protocols enabled on a physical interface. No additional commands are necessary.

If you use static address mapping, you must use the frame-relay map command to statically map destination network protocol addresses to a designated DLCI. If your Frame Relay environment supports LMI and inverse ARP, dynamic address mapping will take place. Therefore, no static address mapping is required.

Again, the encapsulation method can be configured either on an interface or per-destination basis. The following example represents a Frame Relay interface using IETF frames to communicate to the two destinations shown:

Router(config-if)#encapsulation frame-relay IETF
Router(config-if)#frame-relay map ip 131.108.123.2 48 broadcast
Router(config-if)#frame-relay map ip 131.108.123.3 49 broadcast

On the other hand, it is possible to configure encapsulation strictly per destination. In the following example, IETF encapsulation is configured on a per-DLCI basis. This configuration has the same result as the configuration in the first example.

Router(config-if)#encapsulation frame-relay
Router(config-if)#frame-relay map ip 131.108.123.2 48 broadcast ietf
Router(config-if)#frame-relay map ip 131.108.123.3 49 broadcast ietf

The following example shows a case in which most destinations use Cisco encapsulation, but one site requires IETF encapsulation:

Router(config-if)#encapsulation frame-relay
Router(config-if)#frame-relay map ip 131.108.123.2 48 broadcast
Router(config-if)#frame-relay map ip 131.108.123.3 49 broadcast ietf
Router(config-if)#frame-relay map ip 131.108.123.4 50 broadcast