Pseudowire Forwarding As just discussed, the draft-martini architecture relies on
MPLS LSPs to transport Layer 2 (or Layer 1) information across IP/MPLS networks.
In order to send a Layer 2 frame across the network, an ingress PE prepends a label
stack to the frame and sends the resulting packet into an MPLS LSP.
The bottom label in the stack is known as the PW label. This label is used to identify
the pseudowire and the corresponding egress port at the egress PE and is assigned by
the egress PE. If the payload of the MPLS packet is, for example, an Ethernet frame,
the PW label will correspond to an Ethernet port or port and VLAN at the egress PE.
So when the egress PE receives the packet, it will remove the PW label and switch the
payload to the corresponding port or port and VLAN. This process is unidirectional and
will be repeated independently for bidirectional operation.
Any labels above the PW label in the stack are used simply to transport the packet
from the ingress PE to the egress PE along the selected MPLS LSP. In general, only
one such label is required, and it is known as the tunnel label. Note that the tunnel
label, if present, must be immediately above the PW label (except for the case where
an intermediate label is used for router alert VCCV, as described in the pseudowire
OAM section below).
Pages:
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985