cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
7428
Views
5
Helpful
4
Replies

STP Topology change Notification

rrajmohan
Level 1
Level 1

Regarding TCN propagation towards the root switch, why shouldn't the switch sensing the topology change update directly all other switches in the STP instance.

What is the logic behind sending TCN upstream towards the root switch and root switch in turn helps in enabling TCAM updation of all other switches.

1 Accepted Solution

Accepted Solutions

Peter Paluch
Cisco Employee
Cisco Employee

Dear friends,

Please allow me to join.

To understand the motivation behind the two-phase TCN process in the 802.1D legacy STP, recall that:

  1. Configuration BPDUs are originated only by the root bridge. A non-root switch cannot and does not originate a configuration BPDU itself; rather, it only relays (and appropriately updates) a BPDU that was received on its root port. The TC flag that instructs switches to shorten their MAC table aging timer is a part of this BPDU - but a non-root switch that detects a topology change cannot originate a configuration BPDU itself.
  2. Even if a non-root switch was allowed to originate its own configuration BPDU, some of the switches in the topology would not accept this BPDU because it would be inferior to their own BPDUs. The only switch whose BPDUs are unconditionally accepted by all switches is the root bridge, so only the root bridge is in the position of sending a TC signal to all other switches in the topology.

RSTP significantly changed the rules on originating and accepting BPDUs, so its behavior is strongly different. The rules above apply only to legacy 802.1D STP.

Best regards,
Peter

View solution in original post

4 Replies 4

I understand thats the reason RSTP came in to picture :) 

But it came up with a handshake mechanism to avoid any temporary loops during convergence, which STP does not have by design.

Hope this helps

Carlos Villagran
Cisco Employee
Cisco Employee

Hello,

I understand this is because the same reason the DR is elected in OSPF access networks, to have an organized communication, if the BPDUs had to traverse the whole network then there would be even more problems of network downs because of STP instability than we have now.

Root Bridge can send the BPDUs easily reaching every switch in the topology via the root port.

Hope this was not really confusing.

BR.

JC

Peter Paluch
Cisco Employee
Cisco Employee

Dear friends,

Please allow me to join.

To understand the motivation behind the two-phase TCN process in the 802.1D legacy STP, recall that:

  1. Configuration BPDUs are originated only by the root bridge. A non-root switch cannot and does not originate a configuration BPDU itself; rather, it only relays (and appropriately updates) a BPDU that was received on its root port. The TC flag that instructs switches to shorten their MAC table aging timer is a part of this BPDU - but a non-root switch that detects a topology change cannot originate a configuration BPDU itself.
  2. Even if a non-root switch was allowed to originate its own configuration BPDU, some of the switches in the topology would not accept this BPDU because it would be inferior to their own BPDUs. The only switch whose BPDUs are unconditionally accepted by all switches is the root bridge, so only the root bridge is in the position of sending a TC signal to all other switches in the topology.

RSTP significantly changed the rules on originating and accepting BPDUs, so its behavior is strongly different. The rules above apply only to legacy 802.1D STP.

Best regards,
Peter

rrajmohan
Level 1
Level 1

Thank you Peter for the detailed clarification. I completely understand the logic behind that behaviour.

Review Cisco Networking products for a $25 gift card