portfast on a trunk with bpduguard

Answered Question
Jul 16th, 2007
User Badges:

I am configuring a trunk port to a VMWare ESX server. I have had problems with virtualization on AIX servers bleed STP BPDU's across the hypervisor through a redundant link thereby causing a bridging loop. I want to protect myself from this same thing occurring with this ESX server. Any thoughts on whether this is going to do it or possibly better ways to do it?



switchport

switchport trunk encapsulation dot1q

switchport trunk allowed vlan 101,102

switchport mode trunk

no ip address

speed 1000

duplex full

no cdp enable

spanning-tree portfast trunk

spanning-tree bpduguard



Correct Answer by Amit Singh about 9 years 11 months ago

I really dont know the topology you have but based on the description above the "spanning-tree portfast trunk" and "bpduguard" should help you to achieve what you are looking for. You have done the correct configuration on the ports. Please see the link below for more info:


http://www.cisco.com/univercd/cc/td/doc/product/lan/cat3750/12235se/cr1/cli3.htm#wp1946989


HTH,Please rate if it does.


-amit singh

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 3 (2 ratings)
Loading.
Correct Answer
Amit Singh Mon, 07/16/2007 - 08:34
User Badges:
  • Cisco Employee,

I really dont know the topology you have but based on the description above the "spanning-tree portfast trunk" and "bpduguard" should help you to achieve what you are looking for. You have done the correct configuration on the ports. Please see the link below for more info:


http://www.cisco.com/univercd/cc/td/doc/product/lan/cat3750/12235se/cr1/cli3.htm#wp1946989


HTH,Please rate if it does.


-amit singh

Francois Tallet Mon, 07/16/2007 - 08:36
User Badges:
  • Gold, 750 points or more

Your configuration is going to shut down the link as soon as a bpdu is received. That is probably fine if you never expect a bpdu (which is usually the case on a link to a server).

Now, I would just point that the bpdu is not the source of the loop. It's just an indicator that you have some unexpected bridging on the port.

If you had some bleeding of bpdus between your uplink, you probably also had a bleeding of user traffic -> and that's the loop. Looping back the bpdus will actually result in the spanning tree starting to block some ports (if you are running it on the path covered by the loop).

Regards,

Francois

Actions

This Discussion