cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
20075
Views
30
Helpful
6
Replies

STP Tie Breakers

kipper41gsk
Level 1
Level 1

The following information is from the CCNP Switch (642-813) Official Certification Guide

(Copyright© 2010 Pearson Education, Inc.)

Chapter 7: Traditional Spanning Tree Protocol - Page 135

Under section: Electing Designated Ports

Can anyone tell me what these 4 steps mean or provide a better explaination and are these four steps only to break a tie during electing a Designated port?

All tie-breaking STP decisions are based on the following sequence of four conditions:

1.Lowest root bridge ID

2.Lowest root path cost to root bridge

3.Lowest sender bridge ID

4.Lowest sender port ID

1 Accepted Solution

Accepted Solutions

Hello,

Without seeing the topology, it's difficult to give you a clear answer. However, I suspect that the answer to your question is examining the speeds and path cost. Follow this link for more information.

A FastEthernet port running at 100Mb/s will have a path cost of 19. A GigabitEthernet port with a speed of 1Gb/s will have a path cost of 4. Using our second rule, the path of lower cost will take precedence and your Fa0/17 port will be changed to Blocking.

Hope this helps!

View solution in original post

6 Replies 6

Giuseppe Larosa
Hall of Fame
Hall of Fame

Hello Kipper41gsk,

The 4 steps are used for electing a designated port, but also to select a root port, an alternate port and for each other decision STP has to take.

In a network segment, let's suppose there are two LAN switches connected to each other on a point to point link.

If the network has already converged and the root bridge has been elected, both switches will send out an STP BPDU indicating the same root bridge ID  ( that is 16 bit bridge priority followed by the 48 bit MAC address).

If a tie on root bridge ID occurs, the next step is to compare the root path cost to the root bridge: each switch puts in this field the total cost to the root bridge from the point of view of their own root port. In STP the root port cost is added to the cost seen on the upstream switch in rx direction. All STP BPDUs sent out non root ports indicate the same root path cost that is the cost from the point of view of the root port.

Most of the times this comparison leads to the choice of the DP in the segment.

If there is a tie in total cost to root bridge, the next step is to consider the sender Bridge ID field, this field is formed by the local switch bridge priority followed by the MAC address. The lowest sender bridge ID wins.

To be noted if there is a tie at the sender bridge ID level this means that or there are two links connected between two switches or there is a single link connecting two ports of the same switch.

The last criteria is the lowest port id, again the port ID is formed by a port priority followed by a port index that usually is the same as SNMP ifindex.

With default settings all ports have the same default port priority of 128 and the choice is for the lowest port index that is associated to the ports with a lower identification number so port gi1/0/15 is preferred over gi1/0/16 to make an example.

All the fields are contained in the STP BPDU.

Hope to help

Giuseppe

I'm aware of the various tie-breakers, right down to the port ID, but when looking at a topology diagram in a practice exam today, I had a situation where the tie-breaker was between Fa0/17 and Gi0/17. Of these two, which is seen as the lower value?

Hello,

Without seeing the topology, it's difficult to give you a clear answer. However, I suspect that the answer to your question is examining the speeds and path cost. Follow this link for more information.

A FastEthernet port running at 100Mb/s will have a path cost of 19. A GigabitEthernet port with a speed of 1Gb/s will have a path cost of 4. Using our second rule, the path of lower cost will take precedence and your Fa0/17 port will be changed to Blocking.

Hope this helps!

The Tie Between Designated ports is broke by following

* The Bridge ID of the switch(Bridge Priority+VLAN#+MAC Address)

* The cost of the link( FE-100mbps-19 , GE-1000mbps-4, E-10mbps-100)

* The port priority of the sending port( check it with show spanning-tree interface priority

* The interface number of the sending port(physical interface number 1/0,1/10,2/1 etc)

 

Note: In STP lowest is always the best 

I know this is an old thread but hopefully, this can still help folks in the future.

When a tiebreaker comes down to port number, it's not actually the port number itself (Fa0/17 or Gi0/17). Each port has a number you can see with the command show spanning-tree. It's the number after the dot in the Prio.Nbr column. This is what is used during a tiebreaker.

 

davidfoss_8-1666384779270.png

 

Here is the output of show spanning-tree on Switch1 in the above configuration:

davidfoss_4-1666384112314.png

Gi0/1 is elected as the root port because its cost is lower.

However, if we manually set the spanning-tree cost of Gi0/1 and Gi0/2 to 19, so all ports have the same cost...we see Fa0/1 wins. 

davidfoss_5-1666384255287.png

However...Fa0/1 doesn't win because its port number is the lowest. It wins because its neighbor's port number is the lowest. In the previous image, Fa0/1 was plugged into the neighbor switch's Fa0/1 port, Gi0/1 into Gi0/1 as well. Now if we switch the connections so Fa0/1 is plugged into Fa0/3 and Gi0/1 is plugged into Fa0/2:

 

davidfoss_6-1666384433470.png

davidfoss_7-1666384520571.png

We can see Gi0/1 becomes the root port again because its neighbor's port number is lower.

You are not supposed to have a tie breaker at port ID level as one link is Fast Ethernet (path cost=19) and GigaEthernet (path cost=4). So the latter would be preferred for Root Port of Designated Port before coming to the port ID tie breaker.

Review Cisco Networking for a $25 gift card