Spanning Tree Protocol (STP) is a link management protocol that provides path redundancy while preventing undesirable loops in the network. When it comes to ethernet networks, only one active path can exist between two stations in order for them to function properly. Loops occur in networks for a variety of reasons. The most common reason you find loops in networks is from trying to provide multiple levels of redundancy.
When having multiple root switches in a network used to provide this redundancy, the problem that arises is infinite loops between the switches and the end workstations they’re connected to. If a loop exists in the network topology, the potential exists for duplication of messages. When loops occur, some switches see stations appear on both sides of the switch. This confuses the forwarding algorithm and allows duplicate frames to be forwarded.
Spanning Tree comes into play by only allowing a singular path to destinations when having multiple (redundant) links such as this. In a STP environment, the switches exchange information among themselves using bridge protocol data units (BPDU) and will then listen in on all ports for this BPDU message. Once a bridge is turned on, it automatically assumes that it is the root bridge in the STP tree. The STP software chooses a root bridge and calculates all paths from the lower bridges back to itself. In the event of hardware failure of a root bridge in the redundant environment, a new root is elected and port paths would be recalculated.
With multiple uplinks, STP is a must in our switched environment to provide multiple redundancies in case one goes down, another link is there to takes it place within a matter of seconds. At this stage in our continuously growing network topology, there’s virtually no way you’ll ever lose the connection to your dedicated server. That is of course you don’t reboot the server itself, then that’ll be your expected momentary loss of connectivity.
Updated: January 2019