Rikki,
i couldn't completely grasp the meaning of two lines below.
could u elaborate?

>The SPF nodes are completely distinct with the once difference that routes
>derived from them will have the same prefix and prefix length.

thanx,
Vivek.

-----Original Message-----
From: Mailing List [mailto:OSPFdiscuss.microsoft.com]On Behalf Of Rikki
Nguyen
Sent: Wednesday, 15 August 2001 12:55 AM
To: OSPFdiscuss.microsoft.com
Subject: Re: DR/BDR recalculation doubt

There are two scenarios. The first being that this segmentation
also creates two disjoint autonomous systems. The effects are that
of having two completely different autonomous systems.

The more interesting scenario is when the autonomous system is still
intact but the network is disjoint. The SPF calculations still work
correctly whether there are 2 network-LSAS or 10 network-LSAS.
Remember that checks for reverse links must also be done. The SPF
nodes are completely distinct with the once difference that routes
derived from them will have the same prefix and prefix length.

Even though SPF is correct, routing could be screwed. You could end
up with assymetric routing, things being black holed, or maybe routing
loops.

On 14 Aug 2001 18:09:36 +0000, Sam Smith wrote:
> Rikki:
> In the case of segmentation, there will be one DR/BDR per segment. This
> would result in two network LSAs for the same physical network(although
> segmented) with different LinkStateIDs. How is this handled by the
protocol
> and its effects on SPF?
>
> -Sam

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]