Stp Load Balancing - Cisco Catalyst 2000 Configuration Handbook

Catalyst series lan switching
Table of Contents

Advertisement

122 Cisco LAN Switching Configuration Handbook
switch A become root ports. Switch C1 makes its Gigabit Ethernet link to C2 a designat-
ed port because it has the lower sending BID. And sadly, switch C2 must move its Gigabit
Ethernet link to C1 into the blocking state because it is neither a root nor a designated
port. You can see this in the lower half of the figure.
Clearly, an inefficient topology has surfaced because all the traffic passing across the net-
work core must now pass across lower-speed links through switch A. Switch A, being an
access layer switch, is also likely to have less horsepower than the core layer switches.
To remedy this situation, place the STP root bridge somewhere in the core or highest
hierarchical layer of the network. You can do this with the following command for VLAN
10 on switch C1, for example:
(global) spanning-tree vlan 10 root primary
Alternatively, you can explicitly set the bridge priorities with these commands (available
on all Catalyst models):
(global) spanning-tree vlan 10 priority 8192

STP Load Balancing

Figure 7-3 shows a network diagram consisting of three switches that are connected in a
triangle fashion. Each of the links between switches is a trunk, carrying two VLANs. The
switches will be configured so that the two VLANs are load balanced across the available
trunks. The lower half of the figure shows the resulting spanning-tree topologies for
VLAN 100 and VLAN 101.
Distribution switch Catalyst D1 will be chosen as the root bridge. Some users connected
to access switch Catalyst A1 are on VLAN 100, whereas other users are on VLAN 101.
The idea is to have VLAN 100 traffic forwarded to distribution switch Catalyst D1, while
VLAN 101 traffic goes to Catalyst D2.
Note Switch D1 has been selected as the root bridge for both VLANs for simplicity and
to demonstrate the use of port cost adjustments in load balancing. You can also configure
D1 as the root for VLAN 100 and D2 as the root for VLAN 101. The resulting STP topolo-
gies would be the same, but there would be no need to adjust the port costs in switch A1.
An additional benefit is that the two trunk links will failover to each other. Should one
trunk link fail, the other moves from blocking into forwarding mode, forwarding both
VLANs 100 and 101 across the same trunk. If the STP UplinkFast feature is also used on
both switches, the link failover is almost instantaneous.
Switch Catalyst D1 will be configured as the primary root bridge for both VLANs,
whereas Catalyst D2 will become the secondary root bridge. If D1 fails, D2 becomes the
new root.

Hide quick links:

Advertisement

Table of Contents
loading

Table of Contents