TCN flooded between MST Regions

[martin.foote]

Hello

Can anyone explain why the following would occur, it seems like I'm missing something.

In the topology shown running MSTP with two different regions:

Can someone explain to me why when if I shut the link between SW2 and SW3 when it is re-enabled I recieve a TCN message on SW1?

My understanding is that this change and subsequent TCN/MAC table flush should be confined to Region 2 and not propogated to Region 1. What am I missing?

I see a similar question with no replies here https://supportforums.cisco.com/thread/2070140 but thankfully all my kit is Cisco

Thanks in advance.

Martin

[lgijssel · Accepted solution]

Likely it is not the original TCN sent to the root but the corresponding TCN ACK frame broadcast from the root switch (both SW2?) you are observing? This is how it works:

http://www.cisco.com/en/US/tech/tk389/tk621/technologies_tech_note09186a0080094797.shtml#topic1

Besides, as the TCN carries information about the STP domain (root-id) it originated from, it will not cause another region/domain to reduce max_age.

http://www.cisco.com/en/US/tech/tk870/tk136/tk885/technologies_tech_note09186a0080093cc6.shtml#formats

regards,

Leo

[martin.foote]

Hi Leo

Thanks for the response, I just accidently clicked correct answer... not to say you answer isn't correct but just that I don't understand it as yet. Enjoy the Kudos anyway.

I should have said SW1 is the Root Bridge in the topology, so I don't know why a TCN ACK or TCN would be sent to SW1 when SW2 experiences a change?

The priniciple being that each region should contain these changes within themselves, why does SW1 need to know?

Thanks

Martin