> Ok.  I think it would be very helpful to be very clear about these things, as
> there is some confusion.  Our primary discussion has been about leader
> election vs. leaderless.

I agree.

If there are concerns about centralized algorithm (as per RFC 9667) that is a 
separate topic and should be discussed in a separate thread.

The discussion here should be restricted to leader-based distributed mode (as 
per RFC 9667) vs leaderless (which clearly has to be distributed).

   Les


> -----Original Message-----
> From: Tony Li <[email protected]> On Behalf Of Tony Li
> Sent: Monday, November 25, 2024 9:15 AM
> To: Robert Raszuk <[email protected]>
> Cc: Tony Przygienda <[email protected]>; lsr <[email protected]>
> Subject: [Lsr] Re: A counter example
> 
> 
> Hi Robert,
> 
> 
> > > If your skepticism reaches to the centralized flooding topology algorithm
> >
> > It was the above.
> 
> 
> Ok.  I think it would be very helpful to be very clear about these things, as
> there is some confusion.  Our primary discussion has been about leader
> election vs. leaderless.
> 
> 
> > In a mixed vendor's network with a zoo of versions of operating system such
> algorithm would need to be uniformly understood unless you would be
> flooding per each dst node a complete flooding graph.
> 
> 
> Ok.  In a centralized algorithm, the result that is published is simply a 
> list of list
> of nodes: (A, B, C, D, A), (C, E, F, G, H, B), ….  While it is not impossible 
> to mess
> up interpreting this, I would think that a bug in the centralized algorithm 
> itself
> would be far more likely.
> 
> T
> 
> 
> _______________________________________________
> Lsr mailing list -- [email protected]
> To unsubscribe send an email to [email protected]
_______________________________________________
Lsr mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to