I once read the limitation of the maximum switches that can be connected in a loop within single STP domain, but I cant find any document today...
Knowing that we can modify the timers, we are aiming for a fast converging network, so we wont increase the standard timers/delays.
(1) Can someone provide me the answer (and/or -any- URL)? And does Rapid STP change that amount?
(2) A second thought is, imagine we create a ring with the maximum amount of switches, but each one of them connecting another directly. Do we exceed the limit now? Or in other words, must we count these switches too to define the maximum hops?
From the root bridge, BPDUs can cross a maximum number of bridges before being discarded. With RSTP or between MST region, this limit is max age (the age field has been more or less hijacked and is now in fact rather a hop count). Inside MST regions, you have a max-hops limit. Max age goes up to 40, and in the latest IOS implementation, max-hops is 255. So you should be able to go through 40*255 bridges by combining MST regions.
You don't need to tune any timer with RSTP/MST. It's actually recommended you don't tune any timer. They have no impact on convergence time if you network is adapted to these protocols (ie links are point to point and edge ports correctly identified).
Whether in a ring or any other topology, just make sure that the bpdus sent by the root can reach the edge of the network, even after some link failure (for instance in the ring, the distance from the root to the edge could double in case of a single link failure).
We are pleased to announce availability of Beta software for 16.6.3. 16.6.3 will be the second rebuild on the 16.6 release train targeted towards Catalyst 9500/9400/9300/3850/3650 switching platforms. We are looking for early feedback from custome...