Question concerning STP

A new switch gets turned up with a lower priority on those vlans would cause the root to change. Else are you sure it wasn't just a link going down which caused spanning tree to re-run on your core or someplace else....this could cause a network wide outage while spanning tree re-calculates blocking/forwarding on all the l2 links. Other than that, only reason a root changes is if the priority is changed on an existing switch that makes it preferred, a new switch is added with a lower priority, or the switch that is root had a problem.

A "root bridge election" will happen any time that a new device is added to your network that runs spanning tree (another layer-2 device). (This is not

100% correct, but to outline when a new STP device wouldn't cause a spanning-tree election would be long-winded and possibly confusing.) The fact that the root bridge changed is immaterial to the fact that you had an outage, you would have had one even if the root bridge had remained the same because during the election process (which takes about 40 seconds) no traffic is passed between switches. To minimize spanning-tree changes, turn on spanning-tree port-fast on each port that is not used to connect to another layer 2 device. When enabled, port-fast will disable the port if it receives an STP packet preventing the device from connecting to the network and causing STP elections. Port-fast only works on access ports, any port that is trunking ignores port-fast settings.

A new switch gets turned up with a lower priority on those vlans would cause the root to change. Else are you sure it wasn't just a link going down which caused spanning tree to re-run on your core or someplace else....this could cause a network wide outage while spanning tree re-calculates blocking/forwarding on all the l2 links. Other than that, only reason a root changes is if the priority is changed on an existing switch that makes it preferred, a new switch is added with a lower priority, or the switch that is root had a problem.

Thanks for your infos. Actually the switches dont report any link failures nor were new devices added to the network which running STP.

Thats why its kind of strange to me that this reelection took place.

Best...AJ

There are only two ways that a re-election can occur.

1. 'New' higher (lower number) priority bridge appears.
2. Existing root disappears.

These events though can obviously be the result of a number of different situations.

A new higher priority bridge could be the result of someone connecting a new device to the network or by re-configuring an existing device.

The existing root could disappear for various reasons. Communications between it and other bridges could be disrupted for a number of reasons. Busy link Busy CPU on any bridge in network High error rate on link (very high) Failed link disconnecting root from network.

Root re-configured with a lower priority Root crashes/reboots

You don't mention whether the old root was restored. If it was not then it has either gone away or someone has re-configured the network equipment or added a new switch.

You should design the STP configuration and arrange the root bridge and a standby root in the core of the network. This location is required since some traffic may have to flow through the root and if it is near the edge then that is where the traffic must go. Usually, having traffic go to the edge and back is not what you want.

Remembering that the full priority field includes the bridge MAC address, so a bridge with a numerically lower MAC address will preempt a higher MAC with the same priority setting.

Sam