STP Etherchannel Question

[visitor68]

Does STP consider the cumulative bandwidth of an etherchannel when calculating cost to the root?

Thanks

[Laurent Aubert]

Hi,

The answer is yes.

IOS will update dynamically the cost based on the number of active links in the bundle. If you don't want STP to converge in case you loose one of link of the bundle, you need to statically configure the cost for that bundle.

An internal formula is used to determine the default cost for an etherchannel which leads to the following default values:

2-ports GE bundle -> 3

3-ports GE to 7-ports GE bundle -> 2

8-ports GE  bundle -> 1

2-ports 10GE  bundle -> 1

If you want more granularity and still have the dynamic cost update, you can change the coding from 16bits to 32bits. In this case you will have a unique value based on the number of links and the speed of the link. You need to configure spanning-tree pathcost method long on the swith.

HTH

Laurent.

[visitor68]

laurent:

"IOS will update dynamically the cost based on the number of active links  in the bundle"

Isnt that true just for LACP and not for PAgP?

Also, can you be a little more specific regaring the last part of your post? The 32-bit to 64-bit thing...

Thanks!

[Jon Marshall]

ex-engineer wrote:
laurent:
"IOS will update dynamically the cost based on the number of active links  in the bundle"
Isnt that true just for LACP and not for PAgP?
Also, can you be a little more specific regaring the last part of your post? The 32-bit to 64-bit thing...
Thanks!

It's not LACP or PaGP that is dynamically updating, it is STP.

Currently STP uses 16bit numbers for cost. Because nowadays we can have multi-gigabit and 10Gbps links the granularity isn't really there with 16bit numbers so you can use 32bit numbering instead. Below is how the costs would change using 32bit numbering -

Bandwidth	Old STP value	New Long STP value
10 Mbps	100	2,000,000
100 Mbps	19	200,000
1 Gbps	4	20,000
N X 1 Gbps	3	10,000
10 Gbps	2	2,000
100 Gbps	N/A	200
1 Tbps	N/A	20
10 Tbps	N/A	2

Jon