IP modeThis mode requires that endpoints exchange traffic running IP at Layer 3. However, a mix of Layer 2 services can be in place at customer sites. Consider the case of exchanging traffic between one site running Frame Relay and another running Ethernet services across an MPLS network, for example.
In this scenario, a site with a Frame Relay router interface generates a packet with an IP payload. When the frame arrives at the service providers edge router, the router strips off the Frame Relay header and adds an outer MPLS label for switching and an innerlabel for demultiplexing. This is the standard encapsulation scheme for Cisco AToM.
The edge router then transmits the IP payload across an IP pseudo-wire; at the other end, the egress provider-edge (PE) router removes the outer label and inspects the inner label to determine the appropriate interface for forwarding. The PE adds an Ethernet header to the packet and forwards it out the proper interface to the customer site.
This mode requires no configuration changes to customer premises equipment (CPE). In addition, each end of an MPLS tunnel plays an active role in the interworking, fairly balancing the processing required to perform the translation.
In the case of using L2TPv3 in IP mode to achieve service interworking over a native IP network, a packet arrives from the customer, and the PE strips off the Frame Relay header. The PE node appends an L2TPv3 header and a delivery IP header, which identifies the appropriate L2TPv3 pseudo-wire for transmission.
Ethernet modeHere, one end of a tunneled connection currently must be running Ethernet. However, traffic running any protocol at Layer 3 can be passed. In this scenario, CPE requires route bridge encapsulation (RBE) or integrated routing and bridging (IRB) to be configured on it.
The WAN access router at the customer site accepts an Ethernet packet from the LAN and appends both a different Ethernet header and a Frame Relay header to it. The extra Ethernet header is called a shim header, or one that is sandwiched between two other headers. The RBE/IRB function makes sure the shim header is included.
The packet is then transmitted to the ingress PE device. That PE removes the Frame Relay header, establishes a pseudo-wire, and forwards the Ethernet payload to the other end of the pseudo-wire. From there, the egress PE forwards the packet across the local Ethernet interface.
In the L2VPN Interworking implementation, as Aditya correctly pointed out, you can have either a routed/IP or bridged/Ethernet IW (Interworking) mode. So if the IW mode is IP, then the ingress LSR will strip the L2 encap (FR, ATM, Ethernet) and send only the underlying IPv4 packet. If the L3 packet is not IPv4 or an ARP packet, this model will not work.
Conversely, if the IW mode is Ethernet, then the Pseudowire is EoMPLS (VC Type 5). At the ingress LSR, the router will remove the L2 encap (FR, ATM) and send only the underlying SNAP bridged frame.
Also, note that you can do FR-AAL5 IW, but like Aditya noted, you will need to use IRB/RBE on the CE side to create the bridged frames.
Introduction: The "external-out enable" command is available for
configuration under the "router ospf process" in case of the IOS-XR
operating system. This command basically enables advertisement of
intra-area routes on the device as external routes in th...
Introduction Basic configuration for netflow Scale parameters for
netflow Netflow support Architecture Packet flow for netflow Inside the
LC CPU Netflow Cache size, maintenance and memory Sample usage Cache
Size Aging Permanent cache Characteristics Which...