G-4 User’s Reference Guide
Encapsulation and Fragmentation
RFC 1490 describes an encapsulation method for carrying packets across a Frame Relay network. All protocol packets are encapsulated within a Q.922 Annex A frame (a CCITT specification for data frames). The frames must also contain information necessary to identify the protocol being carried, allowing the receiver to properly process the incoming packets.
RFC 1490 also specifies a packet fragmentation and reassembly procedure for carrying frames over a Frame Relay network which are larger than the network’s maximum frame size.
Note: RFC 1490 compliant devices from different vendors should interoperate with each other.
Network Protocol Addressing and Virtual Interfaces
Routing between LANs across a Frame Relay network is similar to routing across a
Network protocol addresses are associated with each PVC using one of two methods: manual configuration or
InARP, as outlined in RFC 1293, allows dynamic mapping of protocol addresses to a DLCI. It can be used for any network protocol, but is most commonly implemented for IP protocol.
When a router discovers a new, active PVC by communicating with the Frame Relay switch using the LMI protocol, it uses InARP to discover the IP address of the remote router on the other end of the PVC, when the PVC becomes active.
However, not all devices support InARP, so IP/IPX RIP is more common. The source address of the RIP response packet gives the Netopia R3100 the other router’s address. In contrast, manual configuration involves manually configuring the association between the PVC DLCI and network protocol addresses. However,
Frame Relay partial mesh support
The following diagram depicts a partially meshed Frame Relay network:
2.0.0.1
router 1
1.0.0.11.0.0.2
router 2
3.0.0.1
4.0.0.1
router 3
1.0.0.31.0.0.4
router 4
5.0.0.1
