The port is configured to allow only that VID to
pass in or out. When frames belonging to different customers pass over the same trunk
between provider bridges, those frames may have both a C-tag and an S-tag. This is the
???Q-in-Q??? format shown in Figure 13.4c. Whether the C-tag is present is the choice of the
customer; it makes no difference to the provider. The S-tag is necessary for the provider
to keep each customer??™s traffic separate from the other customers.
Figure 13.5 points out one limitation of the Q-in-Q model. In Figure 13.5, there are
three customer VLAN bridges, X, Y, and Z, connected via four provider bridges, P, Q,
R, and S. For clarity, no other customers??™ ports or equipment is shown. The customer
is using two C-VLANs, the solid and the dotted lines. The provider is carrying the
customer??™s traffic in a single S-VLAN, shown as a wide gray band in the illustration.
Each of the customer??™s five stations, of which two are routers and three are workstations,
are labeled with their MAC addresses, A through D.
Every Ethernet station is built with a globally unique MAC address. A network can,
however, be operated by overriding that globally unique MAC address with a locally
administered MAC address.
Pages:
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867