is to notify all bridges connected in the tree that there has been a topology change and
restructuring of the tree is in progress (due to a communications link failure somewhere in
the network or a new switch added in the network). The root bridge is determined by the
bridge priority assigned to it and the MAC address.
By default, it is the bridge with the lowest MAC address that gets assigned the role as “root”,
but a specific bridge can be forced to be the root bridge by changing its bridge priority setting
(a lower number with respect to other bridges means higher priority, set on the Spanning
Tree Settings page).
Every communication path between each bridge (managed switch) on the network has an
associated cost. This “path cost” may be determined by the speed of each segment, because it
costs more time to move data at a slower speed, or the path cost can be manually configured
to encourage or discourage the use of a particular network. For example, you may not want
to use a particular high-speed link except when absolutely necessary because you pay a fee to a
service providor for data using that path, while another path is free (no monetary cost).
The path cost is the cumulative cost of all the hops from the root bridge to a particular port
on the network. A Spanning Tree network always uses the lower cost path available between
a port and the root bridge. When the available network connections change, it reconfigures
itself as necessary.
See the RSTP examples topic in this section for an example of how the path cost can be
utilized to establish the primary and backup connections.
During the start-up of a Spanning Tree Network, all bridges (managed switches) are
transmitting configuration messages (BPDUs) claiming to be the root. If a switch receives
a BPDU that is “better” than the one it is sending, it will immediately stop claiming itself
as the root and send the “better” root information instead. Assuming the working network
segments actually connect all of the switches, after a certain period of time there will be only
one switch that is sending its own root information and this switch is the root. All other
switches transmit the root bridge’s information at the rate of the root bridge’s “hello time” or
when the root bridge’s BPDU is received on one of their ports.
The factor for determining which switch is the root (has the “best” root information) is
the bridge priority and its tie-breaker, the switch MAC address. If a switch has more than
one path to get messages from the root, other information in the configuration message
determines which path is the best.
Once the root bridge is determined, all other switches see the root bridge’s information and
information about path (or paths) to the root. If more than one port provides a path to the
root the non-root switches must decide which port to use. They check all of their ports to
select the port that is receiving messages indicating the best path to the root.
The selected port for each bridge is called the root port. It provides the best path to
communicate with the root. The best path is determined first by the lowest total path
cost to the root (root path cost). Each port is assigned a cost (usually based on the speed)
for messages received on that port. The root path cost for a given path is just sum of the
individual port costs for that path. The lowest path cost indicates the shortest, fastest path to
the root. If more than one path has the same cost the port priority assigned to each port, and
its tie-breaker the port number pick the best path.
4-16
Chapter 4 - Managed Switch Software Setup
Stride Industrial Ethernet Switches User Manual 2nd Ed. Rev. A
To Next Page
Loading...