Each of the two copies of the broadcast frame is relayed
back and forth between the two trunks forever, or until a transmission error causes the
frame to be lost. In the meantime, each station on both trunks receives thousands of copies
of the broadcast, so frames destined for station A are often erroneously forwarded to
the wrong trunk 2, where A was mistakenly (though briefly) learned, and the network??™s
bandwidth is quickly saturated. This condition is called a broadcast storm.
The solution to this problem is that bridges do not automatically start forwarding
frames from one port to another the instant they are turned on. Instead, they run a
version of the Spanning Tree Protocol (STP). The two bridges, X and Y in Figure 13.2,
detect each others??™ presence, and one of them (let??™s suppose bridge Y) disables the forwarding
of frames through one of its ports (port Y2 in Figure 13.2, which is marked
with a circle). With no data frames passing through this one port, there are no longer
any loops in the network and therefore no broadcast storm. If bridge X later fails, bridge
Y detects the loss of STP packets and re-enables its port Y2, restoring connectivity
between the LANs.
Pages:
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853