The next topic of discussion
is rapid spanning tree protocol or rapid STP.
The IEEE designation for rapid spanning tree protocol is it? 2.1 w
now spanning tree takes a total of 15
a total of 50 seconds to converge from going from blocking to 14.
If you remember from blocking to listening, it takes 20 seconds. Then, between listening and learning, it takes 15 seconds, the time being called forward DeLay and then between learning and forging it takes another 15 seconds and that time also being called for DeLay.
50 seconds is too long off a time for user's to wait before they can transmit data.
Rapid spanning tree,
however, goes through a quick proposal agreement process before
Now the rapid spanning three ports, rules and states have changed slightly.
Porch upon power up a port in the discarding state when spanning tree has its ports in the blocking
spanning tree transitions to the learning state. Rapid spanning tree is still in the discarding state.
Now these states are named after what the switch is doing to use their data.
So, of course, in the discarding state,
rapid spanning tree will simply discard user frames.
Then you go into the learning state, where you're learning your Mac addresses and populating your Mac address table exactly like spanning tree, and then you go into the fording state.
However, this succession of ST changes happens quicker in rapid spanning tree than it does in spanning, tree
spanning tree. So to say, is timer base
rapid spanning tree goes through a quick proposal agreement process between two switches before the porch can go into 14.
So since rapid spanning tree is based on a quick proposal agreement and it's not timer based, let's go over that quick proposal agreement. Statement
our proposal Agreement process.
Now on this diagram here, I have switched one connected to switch to over port at zero slash one on switch one and zero slash one on switch to
now we're gonna assume that's which one's Mac address is. A.
Now, folks, I know that A is actually 16 bits, not 48 bits, and a Mac address is supposed to be 48 bits. But I just don't have the space here to write 48 bits worth of A's so
e checks Character is four bits of which makes a A 16 bits, but we're gonna pretend that this is just 48 bits.
Switch two's Mac address is B B B B.
the bridge I D Field in rapid spanning tree or purvey land rapid spanning tree remains the same
four bits for priority and 12. It's for villain i D,
which would make the default priority for this switch. 32 768 Just like just like with spanning tree. And let's presume we're just running being on one, which would make the priority field
a total of 32 769 32 768 plus one.
Same. Which switch to the priority would be 32 769
So the tiebreaker in this case would have to be the Mac address. And since which one has the Lower Mac address, then switched to a intact being lower than B B B B in hacks switch one we know is the superior bridge.
Now each one of these switches
sends itself upon power up
a proposal B p d. You
so switch to sends a proposal. BP due to switch one and switch one sends a proposal v. P D. you to switch to
this proposal. Bpd. You is actually in the proposal. BP do the switch is actually claiming that it will have the designated port on this segment between switch one and switch to.
And if we remember, a segment is a connection or a link between two switches
right here between switch Juan and switch to
so in. Each switch in its proposal is claiming that it will have the designated port on the segment.
And if you remember how things worked out in spanning tree, the superior switch always ends up having the designated port on the segment.
So switch to sense. Which one?
A proposal saying I am going to have the designated port auras in my eggs, you know, slash One is going to be the designated port on this segment.
Switch one rejects switch to his proposal, baby to you. Because which one is clearly superior?
Which one also sends
a proposal VP to you to switch to
and switch to seize that? Yes. Which one in fact is superior and Swiss to accept switch ones. Proposal baby to you
at this point if switch to has any downstream ports.
If Swiss who has any down see imports. It puts all these ports in the discarding state,
and this process is called synchronization. So switch to synchronize its itself,
basically isolating itself from the rest of the topology if there are others, which is,
and this is done in order to prevent temporary layer to bridging loops.
So you isolate each segment away from each other segment
to lessen the chance of having in there to nuke.
At this point, switch to sends back
an agreement bpd and makes its zero slash one into a route port.
So this agreement, BP do, reaches which one
and switch one makes its zero slash juan into a designated port.
Now both these ports are put in forging,
and the proposal agreement process between these two switches is over and and then moves one segment downstream. If switch to is connected to, let's say, switch three, same process will happen between switch to and so it's three
now. As you see, this took a total of maybe four seconds as opposed to 50 seconds with regular spanning tree protocol.
So if you remember bpd, use ascent in spanning tree, whether it be regular spanning tree or rapid spanning tree every two seconds. So the 1st 2 seconds were for the proposal. BP do the next 2nd 2 seconds were for the agreement
and within force to six seconds.
These switches have put their ports and fording as opposed to waiting 50 seconds.
This concludes the rapid spanning tree lecture.