Re: [RAM] Preliminary thoughts on LISP 1.5

[Robin Whittle <rw at firstpr.com.au>]

Hi Vince,

I have read your email carefully at least three times and I
still don't have a clear understanding of what you are trying to
describe.

Can you or anyone else provide examples, with more concrete
descriptions, diagrams etc. for each point?  For instance, I
couldn't develop a clear example in my mind of any one of your
options for point 1.

I recognise point 6 as being similar to what I thought
distinguished my Ivip idea from LISP.  As far as I knew, apart
from maybe LISP 1.0 (which I understand is a conceptual stage
rather than something which helps reduce the size of the BGP
routing table), for all other LISP variants the (EID) addresses
mapped by the system (to ETR addresses) were not within any
prefix which was advertised by a BGP router.  But in this
message you raise the possibility of a new construct: an LTA
(LISP Translation Aggregator).  Point 6 mentions a (sub)set of
LTAs advertising "their aggregated EID-to-RLOC mappings" to the
existing global routing table.  Maybe that means something like
what I propose for Ivip ITRs.

Dino protested on 15 June (Re: ViP: Anycast ITRs in the DFZ &
mobile tunnels):

> Robin, this is LISP 1.5, don't you realize that? In LISP 1.5,
> when EIDs come out of PI space, they cannot be routeable via
> the BGP Internet, so they use another topology that does carry
> EIDs. This is exactly what you are describing above.

You and probably Dino have a much more concrete idea of LISP 1.5
than I am able to create in my own mind by reading what little I
have to go on.

I am not convinced Dino understood correctly what I had in mind
when I wrote my first piece.  Hopefully, seven long Ivip
messages since then will enable anyone with sufficient patience
to have a clearer idea of what I am trying to communicate.  I
plan to write a much more accessible I-D in August.  For now,
links to the longer messages I wrote are at:

  http://www.firstpr.com.au/ip/ivip/

I got no idea from your LISP I-D:

  LISP 1.5:  where EIDs are routable for bootstrapping
             EID-to-RLOC mappings; such routing is via a
             separate topology.

that LISP 1.5 involved any router advertising prefixes
containing EID addresses to the rest of the BGP system.

It is five months since you released the LISP I-D and I suspect
I am not the only one who is struggling to create a concrete
example in their minds of LISP 1.5.  Anything you can write by
way of examples would really help.

 - Robin


_______________________________________________
RAM mailing list
RAM at iab.org
https://www1.ietf.org/mailman/listinfo/ram