Rapid Spanning-Tree (RSTP)

22.06.2016 Rapid SpanningTree (RSTP) | NetworkLessons.
com
Search...
Table of Contents
CCIE Routing & Switching
Unit 1: Preparation
Unit 2: Switching
Static MAC Address Table Entry
Cisco Switch Virtualization
Introduction to VLANs (Virtual LAN)
How to conᴀ밄gure VLANs
802.1Q Encapsulation
How to conᴀ밄gure a trunk between switches
Cisco DTP (Dynamic Trunking Protocol) Negotiation
802.1Q Tunneling (Q-in-Q)
Etherchannel over 802.1Q Tunneling
How to change the Native VLAN
VTP (VLAN Trunking Protocol)
VTP Version 3
Protected Port
Private VLANs (PVLAN)
Introduction to Spanning-Tree
Spanning-Tree Cost Calculation
PVST (Per VLAN Spanning Tree)
Spanning-Tree Port States
Spanning-Tree TCN (Topology Change Notiᴀ밄cation)
Spanning-Tree Portfast
Spanning-Tree UplinkFast
https://networklessons.com/cisco/ccieroutingswitching/rapidspanningtreerstp/ 1/27
22.06.2016 Rapid SpanningTree (RSTP) | NetworkLessons.com
Spanning-Tree Backbone Fast
Rapid Spanning-Tree
Rapid Spanning-Tree Conᴀ밄guration
MST (Multiple Spanning-Tree)
Spanning-Tree BPDUGuard
Spanning-Tree BPDUFilter
Spanning-Tree RootGuard
Spanning-Tree LoopGuard and UDLD
FlexLinks
Introduction to Etherchannel
Layer 3 Etherchannel
Cisco IOS SPAN and RSPAN
Unit 3: IP Routing
Unit 4: RIP
Unit 5: EIGRP
Unit 6: OSPF
Unit 7: BGP
Unit 8: Multicast
Unit 9: IPv6
Unit 10: Quality of Service
Unit 11: Security
Unit 12: System Management
Unit 13: Network Services
Unit 14: MPLS
You are here: Home » Cisco » CCIE Routing & Switching
Rapid Spanning-Tree (RSTP)

12 votes
Nowadays we see more and more routing in our networks. Routing protocols like OSPF and
EIGRP are much faster than spanning-tree when they have to deal with changes in the network.
To keep up with the speed of these routing protocols another ᴀ밄avor of spanning-tree was
created…rapid spanning-tree.
Rapid spanning-tree is not a revolution of the original spanning-tree but an evolution. Behind
the scenes some things have been changed to speed up the process, conᴀ밄guration-wise it’s the
same as what you have seen so far. I will refer to the original spanning-tree as “classic spanning-
tree”.
Let’s dive into rapid spanning-tree and we’ll see what the diᴀ洅erences are with the classic
spanning-tree. Take a look at the picture below:
Remember the port states of spanning-tree? We have a blocking, listening, learning and
forwarding port state. This is the ᴀ밄rst diᴀ洅erence between spanning-tree and rapid spanning-
tree. Rapid spanning-tree only has three port states:
Discarding
Learning
Forwarding
You already know about learning and forwarding but discarding is a new port state. Basically it
combines the blocking and listening port state. Here’s a nice overview:
Classic Spanning- Rapid Spanning- Port active in Learns MAC

Tree Tree topology? addresses?
Blocking Discarding No No
Listening Discarding Yes No
Learning Learning Yes Yes
Forwarding Forwarding Yes Yes
Do you remember all the other port roles that spanning-tree has? Let’s do a little review and I’ll
show you what is diᴀ洅erent for rapid spanning-tree:
The switch with the best bridge ID (priority + MAC address) becomes the root bridge. The other
switches (non-root) have to ᴀ밄nd the shortest cost path to the root bridge. This is the root port.
There’s nothing new here, this works exactly the same for rapid spanning-tree. The next step is
to select the designated ports:
On each segment there can be only one designated port or we’ll end up with a loop. The port
will become the designated port if it can send the best BPDU. SwitchA as a root bridge will
always have the best ports so all of interfaces will be designated. The fa0/16 interface on
SwitchB will be the designated port in my example because it has a better bridge ID than
SwitchC. There’s still nothing new here compared to the classic spanning-tree. The interfaces
that are left will be blocked:
SwitchC receives better BPDUs on its fa0/16 interface from SwitchB and thus it will be blocked.
This is the alternate port and it’s still the same thing for rapid spanning-tree. Let me show you a
new example with a port state that is new for rapid spanning-tree:
Here is a new port for you, take a look at the fa0/17 interface of SwitchB. It’s called a backup
port and it’s new for rapid spanning-tree. You are very unlikely to see this port on a production
network though. Between SwitchB and SwitchC I’ve added a hub. Normally (without the hub in
between) both fa0/16 and fa0/17 would be designated ports.
Because of the hub the fa0/16 and fa0/17 interface on SwitchB are now in the same collision
domain. Fa0/16 will be elected as the designated port and fa0/17 will become the backup port
for the fa0/16 interface. The reason that SwitchB sees the fa0/17 interface as a backup port is
because it receives its own BPDUs on the fa0/16 and fa0/17 interfaces and understands that it
has two connections to the same segment. If you remove the hub the fa0/16 and fa0/17 will
both be designated ports just like the classic spanning-tree.
Something else that is diᴀ洅erent is the BPDU, take a look:
The BPDU is diᴀ洅erent for rapid spanning-tree. In the classic spanning-tree the ᴀ밄ags ᴀ밄eld only
had two bits in use:
Topology change.
Topology change acknowledgment.
All bits of the ᴀ밄ag ᴀ밄eld are now used. The role of the port that originates the BPDU will be
added by using the port role ᴀ밄eld, it has the following options:
Unknown
Alternate / Backup port.
Root port.
Designated port.
This new BPDU is called a version 2 BPDU. Switches running the old version of spanning-tree
will drop this new BPDU version. In case you are wondering…rapid spanning-tree and the old
spanning are compatible! Rapid spanning-tree has a way of dealing with switches running the
older spanning-tree version.
Let’s walk through the other stuᴀ洅 that has been changed:
BPDUs are now sent every hello time. Only the root bridge generated BPDUs in the classic
spanning-tree and those were relayed by the non-root switches if they received it on their root
port. Rapid spanning-tree works diᴀ洅erently…all switches generate BPDUs every two seconds
(hello time). This is the default hello time but you can change it.
The classic spanning-tree uses a max age timer (20 seconds) for BPDUs before they are
discarded. Rapid spanning-tree works diᴀ洅erently! BPDUs are now used as a keepalive
mechanism similar to what routing protocols like OSPF or EIGRP use. If a switch misses three
BPDUs from a neighbor switch it will assume connectivity to this switch has been lost and it will
remove all MAC addresses immediately.
Rapid spanning tree will accept inferior BPDUs. The classic spanning tree ignores them. Does
this ring a bell? This is pretty much the backbone fast feature of classic spanning-tree.
Transition speed (convergence time) is the most important feature of rapid spanning tree. The
classic spanning tree had to walk through the listening and learning state before it would move
an interface to the forwarding state, this took 30 seconds with the default timers. The classic
spanning tree was based on timers.
Rapid spanning doesn’t use timers to decide whether an interface can move to the forwarding
state or not. It will use a negotiation mechanism for this. I’ll show you how this works in a bit.
Do you remember portfast? If we enable portfast while running the classic spanning tree it will
skip the listening and learning state and put the interface in forwarding state right away.
Besides moving the interface to the forwarding state it will also not generate topology changes
when the interface goes up or down. We still use portfast for rapid spanning tree but it’s now
referred to as an edge port.
Rapid spanning tree can only put interfaces in the forwarding state really fast on edge ports
(portfast) or point-to-point interfaces. It will take a look at the link type and there are only two
link types:
Point-to-point (full duplex)

Shared (half duplex)
Normally we are using switches and all our interfaces are conᴀ밄gured as full duplex, rapid
spanning tree sees these interfaces as point-to-point. If we introduce a hub to our network we’ll
have half duplex which is seen as a shared interface to rapid spanning-tree.
Let’s take a close look at the negotiation mechanism that I described earlier:
Let me describe the rapid spanning tree synchronization mechanism by using the picture
above. SwitchA on top is the root bridge. SwitchB, SwitchC and SwitchD are non-root bridges.
As soon as the link between SwitchA and SwitchB comes up their interfaces will be in blocking
mode. SwitchB will receive a BPDU from SwitchA and now a negotiation will take place called
sync:
After SwitchB received the BPDU from the root bridge it immediately blocks all its non-edge
designated ports. Non-edge ports are the interfaces that connect to other switches while edge
ports are the interfaces that have portfast conᴀ밄gured. As soon as SwitchB blocks its non-edge
ports the link between SwitchA and SwitchB will go into forwarding state. SwitchB will now do
the following:
SwitchB will also perform a sync operation with both SwitchC and SwitchD so they can quickly
move to the forwarding state.
Are you following me so far? The lesson to learn here is that rapid spanning tree uses this sync
mechanism instead of the “timer-based” mechanism that the classic spanning tree uses
(listening > learning > forwarding). I’m going to show you what this looks like on real switches in
a bit. Let’s take a closer look at the sync mechanism, let’s look at what happens exactly between
SwitchA and SwitchB:
At ᴀ밄rst the interfaces will be blocked until they receive a BPDU from each other. At this moment
SwitchB will ᴀ밄gure out that SwitchA is the root bridge because it has the best BPDU information.
The sync mechanism will start because SwitchA will set the proposal bit in the ᴀ밄ag ᴀ밄eld of the
BPDU. When SwitchB receives the proposal it has to do something with it:
SwitchB will block all its non-edge interfaces and will start the synchronization towards SwitchC
and SwitchD, once this is done SwitchB will let SwitchA know about this:
Once SwitchB has its interfaces in sync mode it will let SwitchA know about this by sending an
agreement. This agreement is a copy of the proposal BPDU where the proposal bit has been
switched oᴀ洅 and the agreement bit is switched on. The fa0/14 interface on SwitchB will now go
into forwarding mode. When SwitchA receives the agreement here’s what happens:
Once SwitchA receives the agreement from SwitchB it will put its fa0/14 interface in forwarding
mode immediately.
What about the fa0/16 and fa0/19 interface on SwitchB?
The exact same sync mechanism will take place now on these interfaces. SwitchB will send a
proposal on its fa0/16 and fa0/19 interfaces towards SwitchC and SwitchD. SwitchC and SwitchD
will send an agreement:
SwitchC and SwitchD don’t have any other interfaces so they will send an agreement back to
SwitchB:
SwitchB will place its fa0/16 and fa0/19 interface in forwarding and we are done. This sync 
mechanism is just a couple of messages ᴀ밄ying back and forth and very fast, it’s much faster
than the timer-based mechanism of the classic spanning tree!
What else is new with rapid spanning tree? There are three more things I want to show you:
UplinkFast
Topology change mechanism.
Compatibility with classic spanning tree.
When you conᴀ밄gure the classic spanning tree you have to enable UplinkFast yourself. Rapid
spanning tree uses UpLinkFast by default, you don’t have to conᴀ밄gure it yourself. When a switch
loses its root port it will put its alternate port in forwarding immediately.
The diᴀ洅erence is that the classic spanning tree needed multicast frames to update the MAC
address tables of all switches.
We don’t need this anymore because the topology change mechanism for rapid spanning tree
is diᴀ洅erent.So what’s diᴀ洅erent about the topology change mechanism?
With the classic spanning tree a link failure would trigger a topology change. Using rapid
spanning tree a link failure is not considered as a topology change. Only non-edge interfaces
(leading to other switches) that move to the forwarding state are considered as a topology
change. Once a switch detects a topology change this will happen:
It will start a topology change while timer with a value that is twice the hello time. This will
be done for all non-edge designated and root ports.
It will ᴀ밄ush the MAC addresses that are learned on these ports.
As long as the topology change while timer is active it will set the topology change bit on
BPDUs that are sent out these ports. BPDUs will also be sent out of its root port.
When a neighbor switch receives this BPDU with the topology change bit set this will happen:
It will clear all its MAC addresses on all interfaces except the one where it received the BPDU
with the topology change on.
It will start a topology change while timer itself and send BPDUs on all designated ports and
the root port, setting the topology change bit.
Instead of sending a topology change all the way up to the root bridge like the classic spanning
tree does, the topology change is now quickly ᴀ밄ooded throughout the network. If you want to
know exactly how this topology change mechanism works then take a look at this lesson.
Last but not least, let’s talk about compatibility. The short answer is that rapid spanning tree
and classic spanning tree are compatible. However when a switch running rapid spanning tree
communicates with a switch running classic spanning tree all the Speedy Gonzales features
won’t work!
Last but not least, let’s talk about compatibility. The short answer is that rapid spanning tree 
and classic spanning tree are compatible. However when a switch running rapid spanning tree
communicates with a switch running classic spanning tree all the Speedy Gonzales features
won’t work!
In the example above I have my three switches. Between SwitchA and SwitchB we will run rapid
spanning tree. Between SwitchB and SwitchC we will fall back to the classic spanning tree.
Seen enough theory? In the next lesson I will show you the conᴀ밄guration and the debugs of
everything that you have learned so far.
Rate this Lesson:
« Previous Lesson
Spanning-Tree Backbone Fast
Next Lesson
Rapid Spanning-Tree Conᴀ밄guration »
Home › Forums › Rapid Spanning-Tree (RSTP)
This topic contains 19 replies, has 9 voices, and was last updated by Andrew P 3 weeks, 4
days ago.
Viewing 15 posts - 1 through 15 (of 19 total)

12→
Author
Posts | Subscribe
December 30, 2014 at 08:28 #11378 Reply
Raed A
Participant
Thank you very much
December 30, 2014 at 08:40 #11379 Reply


Rene Molenaar
Keymaster
You are welcome Raed!
January 25, 2015 at 15:41 #11380 Reply
Michael M
Participant
Wow. Excellent lesson. Very clear and to the point. I am deᴀ밄netly enjoying this site and learning.
Excellent refresher cource !
Please keep up the great work. BTW…Yes I am recommonding this site.
January 26, 2015 at 21:20 #11381 Reply
Rene Molenaar
Keymaster
Glad to hear you like it!
June 13, 2015 at 23:11 #11382 Reply
Ricardo V
Participant
Excellent explanation!
June 21, 2015 at 09:52 #11383 Reply
Srinivasan C
Participant
Hi Rene,
Very Good explanation on RSTP.
I think BPDU message ᴀ밄ags are reversed in the diagram.
Thanks,
Srini

June 21, 2015 at 12:11 #11384 Reply
Srinivasan C
Participant
Hi Rene,
what is the characteristic of blocking mode in RSTP? Can send and receive BPDUs?
In most of the scenario Alternate / Backup port directly transition in to forwarding ?
On what condition the port will be in Learnging stat?
Thanks,
Srini
June 21, 2015 at 16:05 #11385 Reply
Rene Molenaar
Keymaster
Hi Srini,
The port in blocking mode will only receive BPDUs, not send them. The same thing applies to
the root port. Only the designated port sends BPDUs since this is downstream (to other
switches).
The alternate port can go into forwarding when the switches loses its root port, the same thing
occurs when you use uplinkfast with PVST.
Each interface has to go through the blocking, listening and learning mode unless you enable
portfast:
https://networklessons.com/cisco/spanning-tree-port-states/
Rene
June 21, 2015 at 16:10 #11386 Reply
Rene Molenaar
Keymaster
Hi Srini, 
Which one do you mean?
Rene
June 21, 2015 at 16:21 #11387 Reply
Srinivasan C
Participant
Hi Rene,
Can you please explain bit more on TC while timer and Fd While Timer in RSTP?
Tc while timer 2xhello time = 4 sec , only 4 sec for ᴀ밄ushing the Mac address learned through
topology change action ports. Please correct my understanding on Tc while timer.
Thanks,
Srini
June 21, 2015 at 16:29 #11388 Reply
Srinivasan C
Participant
Hi Rene,
I am seeing BPDU Message Flags in reverse order.
1 TCN
2 Proposol
.
.
.
7 Agreement
8. TCA.
Please correct me If my understanding is wrong.
Thanks,
Srini
June 21, 2015 at 16:34 #11389 Reply

Srinivasan C
Participant
Hi Rene,
You are correct . In RSTP a special case , designated port in the non-edege port of a switch will
go in to blocking state and block all non-BPDU traᴀ甇c until it gets agreement from the neighbour
switch.
Thanks,
Srini
June 25, 2015 at 16:26 #11390 Reply
Rene Molenaar
Keymaster
Hi Srini,
When a non-edge or root port moves into the forwarding state then this triggers as topology
change in RSTP. When this happens it will start the TC while timer which 2x hello (4 seconds in
total) and it will ᴀ밄ush the MAC addresses that were learned on all non-edge interfaces.
When the TC while timer is active, all BPDUs that the switch will send will have the TC bit set.
Other switches that receive this BPDU will start their own TC while timer and they will ᴀ밄ush all
MAC addresses except the interface where they received the BPDU with the TC bit.
I hope this helps.
Rene
October 12, 2015 at 03:49 #18005 Reply
Frades
Participant
awesome lesson! very detailed!
i have a question. you said that RSTP uses uplinkfast by default, so no need to conᴀ밄gure it.
done some research on the internet, they are saying that RSTP doesnt use the BBfast and 
ULfast, but it has a similar feature like them which is set on by default. meaning, when you issue
the “show span summary” you’ll see the Uplinkfast and BackboneFast disabled, because RSTP
has a built-in feature like that. is this true? thanks!
October 12, 2015 at 09:46 #18012 Reply
Rene Molenaar
Keymaster
Hi John,
That’s correct, RSTP has features that are similar to uplinkfast and backbonefast, there’s no
need to activate these.
Rene
Author
Posts
Viewing 15 posts - 1 through 15 (of 19 total)

12→
Reply To: Rapid Spanning-Tree (RSTP)
b i link b-quote del img ul ol li code close tags
Please put code in between `backticks` or use the CODE button.

To place inline images, please use any image share service (such as TinyPic or Imgur) and use
the IMG button!
Notify me of follow-up replies via email
Maximum ᴀ밄le size allowed is 2048 KB.
Attachments:
Выберите файл Файл не выбран 
Add another ᴀ밄le
Submit
About NetworkLessons.com
Hello There! I'm René Molenaar (CCIE #41726), Your Personal Instructor of
Networklessons.com. I'd like to teach you everything about Cisco, Wireless and
Security. I am here to Help You Master Networking!
Read my story
Social Fans
Social Fans
  
14,267 7,929 1,589
FANS FOLLOWERS SUBSCRIBERS
Highest Rated Lessons
MPLS Layer 3 VPN Conᴀ밄guration

(25 votes)
VRF Lite Conᴀ밄guration on Cisco IOS

(23 votes)
Cisco Portfast Conᴀ밄guration

(20 votes)
IPv6 Address Types

(18 votes)
EIGRP Stub Explained

(17 votes)
New Lessons
Introduction to Cisco IOS XE 

ERSPAN Conᴀ밄guration on Cisco IOS XE
IGMP Filter
IGMP Snooping without Router
Cisco Group Management Protocol (CGMP)
Disclaimer
Privacy Policy
Support
Rapid Spanning-Tree (RSTP) written by Rene Molenaar average rating 4.8/5 - 12 user ratings

Rapid Spanning-Tree (RSTP)

Uploaded by

Copyright:

Available Formats

Rapid Spanning-Tree (RSTP)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Rapid Spanning-Tree (RSTP)

Uploaded by

Copyright:

Available Formats

22.06.2016 Rapid Spanning­Tree (RSTP) | NetworkLessons.

Cisco Switch Virtualization

Introduction to VLANs (Virtual LAN)

How to conᴀ밄gure VLANs

How to conᴀ밄gure a trunk between switches

Cisco DTP (Dynamic Trunking Protocol) Negotiation

802.1Q Tunneling (Q-in-Q)

Etherchannel over 802.1Q Tunneling

How to change the Native VLAN

VTP (VLAN Trunking Protocol)

Private VLANs (PVLAN)

Spanning-Tree Cost Calculation

PVST (Per VLAN Spanning Tree)

Spanning-Tree Port States

Spanning-Tree TCN (Topology Change Notiᴀ밄cation)

Spanning-Tree Backbone Fast

Rapid Spanning-Tree Conᴀ밄guration

MST (Multiple Spanning-Tree)

Spanning-Tree LoopGuard and UDLD

Cisco IOS SPAN and RSPAN

You are here: Home » Cisco » CCIE Routing & Switching

Rapid Spanning-Tree (RSTP)

Classic Spanning- Rapid Spanning- Port active in Learns MAC

Listening Discarding Yes No

Learning Learning Yes Yes

Forwarding Forwarding Yes Yes

Something else that is diᴀ洅erent is the BPDU, take a look:

Point-to-point (full duplex)

What about the fa0/16 and fa0/19 interface on SwitchB?

Rate this Lesson:

Viewing 15 posts - 1 through 15 (of 19 total)

December 30, 2014 at 08:40 #11379 Reply

January 25, 2015 at 15:41 #11380 Reply

January 26, 2015 at 21:20 #11381 Reply

June 13, 2015 at 23:11 #11382 Reply

June 21, 2015 at 09:52 #11383 Reply

June 21, 2015 at 12:11 #11384 Reply

June 21, 2015 at 16:05 #11385 Reply

June 21, 2015 at 16:10 #11386 Reply

June 21, 2015 at 16:21 #11387 Reply

June 21, 2015 at 16:29 #11388 Reply

Please correct me If my understanding is wrong.

June 21, 2015 at 16:34 #11389 Reply

June 25, 2015 at 16:26 #11390 Reply

I hope this helps.

October 12, 2015 at 03:49 #18005 Reply

October 12, 2015 at 09:46 #18012 Reply

Viewing 15 posts - 1 through 15 (of 19 total)

Please put code in between `backticks` or use the CODE button.

Notify me of follow-up replies via email

Maximum ᴀ밄le size allowed is 2048 KB.

Highest Rated Lessons

MPLS Layer 3 VPN Conᴀ밄guration

VRF Lite Conᴀ밄guration on Cisco IOS

Cisco Portfast Conᴀ밄guration

IPv6 Address Types

EIGRP Stub Explained

Introduction to Cisco IOS XE 

You might also like

22.06.2016 Rapid SpanningTree (RSTP) | NetworkLessons.