Re: [Idr] WGLC on draft-ietf-idr-as-private-reservation-00

[Jon Mitchell <[e-mail]>]

First up front, this is the text with the current range and sizing
suggested to IANA in the draft, which some people seem to have missed
based on other threads:

   [Note to IANA, NOT for publication: The IANA should update the "16-
   bit Autonomous System Numbers" registry to reference this RFC (when
   published) for the existing private use reservation.  Further, to
   maintain consistency from an operator standpoint, it is suggested
   that the end of the "32-bit Autonomous System Numbers" range be
   reserved for Private Use, and a size of 16777215 (value to replace
   TBD1 below) is suggested corresponding to the range of 4278190080
   (value to replace TBD2 below) to 4294967294 (value to replace TBD3
   below).]


Other comments inline...

Jon


On Sun, Dec 09, 2012 at 04:07:23AM -0500, Jeff Wheeler wrote:
> On Wed, Nov 28, 2012 at 4:26 PM, John Scudder <[e-mail]> wrote:
> > We have received a request for a working group last call on draft-ietf-idr-as-private-reservation-00. A URL for the draft is http://tools.ietf.org/html/draft-ietf-idr-as-private-reservation-00
> 
> I did some searching to try and find if there has been any discussion
> on the recommended range of values that IANA should reserve.  I didn't
> find any, but admittedly, my searching skills may have simply been
> poor.
> 
> I mentioned this draft to two colleagues, and our discussion was
> immediately that it would be nice if the range of newly reserved
> values were easier for humans to read.  If it simply extended from
> 4,000,000,000 until 2**32-2 this would be a little nicer.
> 
> If there has been discussion about this, could someone point me to it?

The draft originally was more human/decimal boundary friendly before it
was a WG doc.  There has been some on list, some on meeting (Vancouver)
and some out of meeting discussions I did before we moved from nice
decimal boundary to nice bit or asplain friendly boundary).  Here is one
of the emails:

http://www.ietf.org/mail-archive/web/idr/current/msg06468.html

In the Vancouver meeting the draft was presented, and on range structure
a show of hands poll was done, and from my memory (meeting minutes does
not capture result unfortunately) it was overwhelmingly for bit/asplain
friendly boundary (about 10:1).  Others I talked to subsequently
preferred this approach as well (I think Tony's comment earlier about
hex friendly numbers can be grouped similarly).  I'd also point out that
the existing range is more bit friendly than human friendly so this is
not a change in approach.

Sizing has also been discussed a bit on and off list with more
preferring we aren't having this debate again in 10 years due to some
new novel use of BGP, and based on that feedback the current 24 bit text
was implemented, at this point unless a wide number of folks have a
specific change we can all agree to, I'm inclined to leave both as is.
Although I agree the resource we are using is really large, it seems
that reserving over 6% of the total space for private use is a step too
far to me.

Jon

> 
> Operators care about what things look like in the CLI or the NMS.  A
> bunch of private ASN all beginning with 427xxxxxxx 428xxxxxxx
> 429xxxxxxx is a little confusing, especially because 4278190080 would
> be a private ASN but 4278190079 is not.
> 
> -- 
> Jeff S Wheeler <[e-mail]>
> Sr Network Operator  /  Innovative Network Concepts
> _______________________________________________
> Idr mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/idr