It learns that this particular {customer
source MAC address, EVC} pair is associated with the I-component whose MAC
address is the outer source MAC address of the frame (I-component C). Thus,
I-component E knows how to get back to I-component C when the reply is returned.
10. I-component E does not know the destination MAC address, so it floods the frame
to all ports in the EVC. Only the port to S-cloud 1B belongs to the EVC, so it floods
it only in that direction. I-component E translates the I-SID into the appropriate
S-VLAN value for S-cloud 1. The VLAN ID may be different from the VLAN-ID
used in S-cloud 2; their VLAN ID spaces are totally independent.
11. S-cloud 1B delivers the frame to customer device C-T.
12. When C-T replies to C-S, I-component E gets the frame and uses its learned information
to encapsulate the frame, as shown in Figure 13.9d. It has learned what
unicast outer destination MAC address to use, so the backbone network can deliver
the frame straight to I-component C.
13. While delivering the return frame, I-component C learns from this return frame
what outer MAC address to use to reach customer device C-T.
Pages:
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898