Re: automating the network: the big picture...

[Erik Nordmark <Erik.Nordmark@sun.com>]

> i think erik was looking at some of this, but am not sure i
> remember the details well enough, so will let him speak for
> himself.

I was merely trying to encourage the folks that said "we need 3rd party free
robust configuration/discovery" to focus their energy on writing up a problem
statement and having a BoF. Nothing happened AFAIK.

> steve deering wrote in the margin of a book that he had solved this
> problem, but did not have time or space to write the complete
> design down before he left for the airport.

He had the above problem statement in his head and it made sense to him.

I think there are some tradeoffs to take into account between e.g. 
securing a single channel for configuration information (between the client 
and the DHCP server) with the DHCP server being a single point of failure,
and securing N channels to N entities providing configuration information
co-located with the actual service (with fate sharing between the
configuration entity and the sevice itself).
I'm sure there are other tradeoffs as well.

   Erik