The many
customers of that provider network, assuming for a moment they are using bridges
(as shown in Figure 13.5), are also presumably running STP. This is especially important
if a customer has two or more connections from one of his or her networks to the
provider??™s network.
Forwarding loops and the consequent broadcast storms could be avoided if all of
these interconnected bridges, both customer and provider bridges, ran a single globally
connected instance of the Spanning Tree Protocol. This is not practical, however. One
reason is that the STP protocols simply do not scale to encompass all of the provider
and all of the customer bridges in the world. Another is that these protocols depend on
certain configuration parameters that must be coordinated in the different bridges in a
network. The cooperation required among system administrators would be prohibitive,
and any error by any administrator could affect the whole world.
A very simple way to keep the providers??™ spanning trees separate from the customers??™
spanning trees was adopted by IEEE Std 802.1ad-2005. As mentioned in ???In-Band
Signaling,??? a standard customer bridge (C-bridge) transmits and receives frames carrying
the Spanning Tree Control protocols (bridge protocol data units, or BPDUs) using
the destination MAC address 01-80-C2-00-00-00 and prevents all frames whose
destination MAC address is in the 01-80-C2-00-00-00 through ??¦-0F range from being
relayed from port to port.
Pages:
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873