[c-nsp] Deploying MSTP

[Jon Harald Bøvre]

During your transition, or possibly for a long time dependent on where
the STP edge will be:
 
This is the statement from config guide which has made my headache:
-In a mixed MSTP and PVST+ network, the common spanning-tree (CST) root
must be inside the MST backbone, and a PVST+ switch cannot connect to
multiple MST regions.
 
A. have full controll of root (and STP edge) for all vlans
B. start migration on root switch
Any vlan outside MST region with lower STP priority than MST will be in
a Inconsistent state until this has been solved.
'show spantree root' on the root switch is a good starting advice
 
Easy to simulate this behaviour using only 2 switches
 
Jon Harald Bøvre
 
 
On 28. des. 2011 17:14 "Steve Dodd" <sdodd at syringanetworks.net> wrote:
> -----Original Message-----
> From: <cisco-nsp-bounces at puck.nether.net>
> [mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Chuck Church
> Sent: Wednesday, December 28, 2011 7:02 AM
> To: 'Frédéric Loui'; 'Jay Nakamura'
> Cc: 'cisco-nsp'
> Subject: Re: [c-nsp] Deploying MSTP
> Yep, definitely important to verify what VLANs are mapped to the
> instances. The Nexus for example reserves some VLANs that aren't
> obvious, so when you map them, they end up back in instance 0 with no
> warning. IOS didn't reserve those, so both my instance 0 and the one
> with high-numbered VLANs didn't match, so I had all kinds of problems
> till we discovered that. There is a checksum that you can get it to
> display that should match between all connected devices. Can't
> remember the command, but it worked well.
> Chuck
> show spanning-tree mst configuration digest
> will show you the hash and should match on all devices in the MST
> region.
> -Steve
> _______________________________________________
> cisco-nsp mailing list <cisco-nsp at puck.nether.net>
> <https://puck.nether.net/mailman/listinfo/cisco-nsp>
> archive at <http://puck.nether.net/pipermail/cisco-nsp/>