Re: Last Call: <draft-ietf-lisp-eid-block-03.txt> (LISP EID Block) to Informational RFC

[SM <[e-mail]>]

Hi Luigi,
At 06:32 15-11-2012, Luigi Iannone wrote:
>  thanks for the comments. Few answers inline.

Thanks for the response.

>Well, if we go along that road we should put the whole document in a 
>single "IANA Considerations" Section. ;-)

Yes. :-)

>Actually the current IANA Considerations section states the same 
>request but does not specify "/12".
>You are right that it should be clearly stated, to make the document 
>coherent. Will fix.

Ok.

>It refers to the IANA allocation policies. May be it could be 
>changed in the following way:
>
>         If in the future there will be need for a larger EID Block the
>         address space adjacent the EID Block could be allocate by IANA
>         according to its current allocation policies."
>
>Would that work?

Not yet. :-)  You did not answer my question about which IANA 
allocation policies the draft is referring to.

>I agree that this point has been not discussed thoroughly, the idea 
>is not to create any new "manager", rather to make ISPs (or whoever 
>interested in deploying LISP) to request an EID address 
>sub-block  as they do with usual prefixes.

I'll comment below.

>Well, this is standard, to have a reserved space we have to go 
>through the (now called) "IETF Review", which is what we are doing ;-)

What the draft is doing is reserving address space.  According to 
Section 10 address blocks from that reserved address space (the /16) 
will be assigned through IETF review.  I read the previous comment as 
meaning that the EID address block will be assigned to ISPs by 
RIRs.  There isn't any mention of that in the draft.  Even if it was 
mentioned it is doubtful that ISPs would be able to get that address 
space assigned through RIRs at present.  The issue was mentioned in 
the AD review [1].  I didn't find any discussion of that in the LISP 
mailing list archives.

According to the LISP Charter the document is to request "address 
space to be used for the LISP experiment as identifier space".  The 
draft is reserving a /16 and there is scope to have that extended to 
a /12 in future.  This goes beyond the usual experiments.  There 
isn't any discussion of how the ip6.arpa delegation will be 
handled.  There isn't any discussion of how long the experiment will 
last.  I understand that IPv6 is not a scare resource.  However, that 
is not a reason for handing out address space and leaving it to 
someone in future to figure out what to do if this becomes a problem.

I haven't seen anyone asking why the document is not a BCP.  If the 
aim is to have:

   "Routers in the Legacy Internet must treat announcements of prefixes
    from the IPv6 EID Block as normal announcements, applying best
    current practice for traffic engineering and security."

I think that the document might have to be a BCP.  It would be good 
to be clear about whether the address space should be listed in the 
IANA IPv6 Special Purpose Address Registry.

Section 5 mentions that "The working group reached consensus on an 
initial allocation".  Could the document shepherd upload the write-up 
and provide some details about that?

Regards,
-sm

1. http://www.ietf.org/mail-archive/web/lisp/current/msg03848.html