Go Premium for a chance to win a PS4. Enter to Win

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 504
  • Last Modified:

STP - what's wrong with defaults

Experts,

I keep hearing that you should never let the defaults determine the root bridge, and it's better to manually determine that based on your network topology.

My question is, during the Root bridge election process, the link COST are taken into consideration. So shouldn't we trust that whichever was elected to be the root bridge is also the best way to get frames from point A to Point B?
0
trojan81
Asked:
trojan81
4 Solutions
 
Robert Sutton JrSenior Network ManagerCommented:
If this is for a switch then it is usually best for you to determine the root "switch" since you will typically have multiple vlans. Each vlan must have its own root "switch" since each one is its own broadcast domain.  You really didn't specify the equipment being used except for the topic that you chose of Switches & Hubs.


According to Cisco.com:

With STP, the key is for all the switches in the network to elect a root bridge that becomes the focal point in the network. All other decisions in the network, such as which port to block and which port to put in forwarding mode, are made from the perspective of this root bridge. A switched environment, which is different from a bridge environment, most likely deals with multiple VLANs. When you implement a root bridge in a switching network, you usually refer to the root bridge as the root switch. Each VLAN must have its own root bridge because each VLAN is a separate broadcast domain. The roots for the different VLANs can all reside in a single switch or in various switches.

Note: The selection of the root switch for a particular VLAN is very important. You can choose the root switch, or you can let the switches decide, which is risky. If you do not control the root selection process, there can be suboptimal paths in your network.

All the switches exchange information for use in the root switch selection and for subsequent configuration of the network. Bridge protocol data units (BPDUs) carry this information. Each switch compares the parameters in the BPDU that the switch sends to a neighbor with the parameters in the BPDU that the switch receives from the neighbor.

In the STP root selection process, less is better. If Switch A advertises a root ID that is a lower number than the root ID that Switch B advertises, the information from Switch A is better. Switch B stops the advertisement of its root ID, and accepts the root ID of Switch A.

Hope this helps.
0
 
Craig BeckCommented:
Also, adding a switch can cause the STP topology to change, and that can cause the paths to also become suboptimal.

If someone connects a switch with a manually-set STP root value for a particular VLAN that could cause important links to transition into an undesirable state.  That could be very bad.
0
 
Don JohnstonCommented:
My question is, during the Root bridge election process, the link COST are taken into consideration. So shouldn't we trust that whichever was elected to be the root bridge is also the best way to get frames from point A to Point B?
Spanning Tree doesn't determine the best way to get frames from "point A" to "point B".  Unless one of those two points is the root. :-)

Spanning tree will only allow one path from any place on the network to/from the root. Which means that there may be a server one switch away from you, but spanning tree will requires your traffic to go all the way to the other side of the network (because that's where the root bridge ended up.

And only Cisco's implementation of Spanning Tree utilizes Per VLAN Spanning Tree (with one root bridge per VLAN). All the other vendors use a CST (Common Spanning Tree) approach. Although most vendors support 802.1s Multi Spanning Tree which uses a user-defined number of spanning-tree instances.
0
 
AkinsdNetwork AdministratorCommented:
It is best practice to influence the root bridge for your vlans. If 1 switch is centralized, then you can have just that one switch function as the root bridge. Otherwise, you can use different switches function as root bridge for particular vlans.

Also, to buttress craigbeck's point. It is a good idea to implement root guard on your access layer switches to prevent a rogue switch from disrupting your  network

Food for thought
Rather that have a redundant link sitting idle and waiting for something to fail, you could make your distribution a stack of switches and then make the 2 uplinks ether channels to the stack, one link per member of the stack. This way, you benefit from additional bandwidth and if one fails, there is no disruption as the other link continues to serve. You can then turn of spanning tree and conserve resources for your switches.

There are more ways to better tune spanning tree for more granular functionality if you desire. Several options are available to you and you have the option of choosing what is best for your network, functionally and manageability
0
 
trojan81Author Commented:
Thank you gentlemen!
0

Featured Post

Free Tool: ZipGrep

ZipGrep is a utility that can list and search zip (.war, .ear, .jar, etc) archives for text patterns, without the need to extract the archive's contents.

One of a set of tools we're offering as a way to say thank you for being a part of the community.

Tackle projects and never again get stuck behind a technical roadblock.
Join Now