STP limitations

[lieven]

Hi,

I once read the limitation of the maximum switches that can be connected in a loop within single STP domain, but I can’t find any document today...

Knowing that we can modify the timers, we are aiming for a fast converging network, so we won’t 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?

(3) A last mind-breaker I have is: if we create, in a single STP domain, 2 independent rings with one common switch. Each ring contains, let say 5 switches, each connected to each neighbor and the 2 boundary switches to that common switch. I assume this setup work if we elect the common switch as Root Switch? What if we elect the Root Switch in the middle of ring A? We likely will exceed the maximum number of switches meaning HELLO packets will not arrive at time to middle switches of ring B … ? I’m I correct?

Thanks in advance!

Regards,

Lieven

[amit-singh]

Hi Lieven,

I think you are getting confused with the STP diamater value in a single STP domain with the number of switches used in a domain. STP converges well and its behaviour can be prediscted with in a diameter of 7 hops which is called as STP diameter. STP will still wotk above this value but it will hard to predict the behaviuor of it and how will it converge. That's why the recommended STP diamater is 7 hops from the root bridge. It actually doesnot matter the number of switches that you are using in your n/w ( dont know the max in an/w though :( ) but just keep in mind that when you are designing you n/w topology the 7 hops vlaue should not exceeded.

RSTP also uses the same vlaue.

I hope all the questions are answered :)

HTH,

-amit singh

[lieven]

Hi Amit Singh,

Thanks! "STP diameter" was the parameter that I was looking for. I knew all the way that it had the value of 7, in contradiction of some colleagues which have to buy me a beer now :D but one thing still kooks my noodles…

Cisco quoted diameter of the STP domain (dia) as following: “It is the maximum number of switches that you cross to link any two switches in the bridged network (including source and destination), if you assume the worst cases (without crossing twice the same switch, of course)”, which seems to be something else, according you described?

Anyway, to cut down a large complex network to match that dia, implementing MST Regions or routing seems to be inevitable!

L.

[amit-singh]

“It is the maximum number of switches that you cross to link any two switches in the bridged network (including source and destination), if you assume the worst cases (without crossing twice the same switch, of course)”, which seems to be something else, according you described ".

I think what was mentioned is absolutely correct and I do agree with this. 7 hops is the diameter inlcuding your source and the destination switches i.e inlcuding your root to the destination switch.

HTH,

-amit singh