6

We have some Cisco switches running PVST. I need to convert them to use extended system-id so I can use high numbered vlans.

The Cisco stp docs say

When you enable or disable the extended system ID, the bridge IDs of all active STP instances are updated, which might change the spanning tree topology.

If I start at the PVST root and change my primary root, then secondary root, and then all other switches, are there any risks of loops or nasty downtime?

3
  • As long as you know all Bridge_IDs in your network and configure the existing root switch with the lowest value, I believe you will be fine. Nevertheless I never tested it in production.
    – laf
    Commented Jul 25, 2013 at 11:12
  • 3
    Make sure your defined bridge priorities are multiples of 4096.
    – Santino
    Commented Jul 25, 2013 at 12:59
  • Did any answer help you? If so, you should accept the answer so that the question doesn't keep popping up forever, looking for an answer. Alternatively, you can post and accept your own answer.
    – Ron Maupin
    Commented Jan 4, 2021 at 23:33

2 Answers 2

5

If anything it may just need to reconverge. About 45 seconds of interruption IIRC. If you can, while you are changing it on all the switches anyway, use rapid PVST. That will allow for faster convergence.

2

Be aware - when you change your root, you'll cause a topology change, at which time all your STP will pause and recalculate. While 45 seconds doesn't sound like very long - it's an eternity to the systems on your network, some of which think .15 seconds (150ms) is a long time.

I caused a topology change accidentally on my "storage" vlan (NAS). SQL servers lost their connection to NAS attached hard drives - and all crashed, which caused front end systems to crash. It was a poor design, that was like that when I got there, and I stepped on a mine in the mine field - so I didn't get fired. But the server team had to spend the rest of the day bringing servers back up. Then they had to do it again, when we fixed spanning tree and it recalculated again.

Make sure to CYA - advise management, in writing, that this is highly risky to the uptime of the network, and make sure you get a change window, with an approved change control (with an OK from the Server Team management). All of course depending on your network and environment.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.