Re: 3177-bis

[Thomas Narten <narten@us.ibm.com>]

Mark Smith <ipng@69706e6720323030352d30312d31340a.nosense.org> writes:

> I see the single /48 allocation size in that light. Yes, too big for
> most people, most of the time. However, /48 for nearly every end-site
> might be providing more indirect benefits to all end-users/end-sites as
> a whole rather than direct benefits to the individual. I know that
> managing two end-site prefix lengths /48s and /56s will probably double
> my employer's end-site addressing management costs. Two end-site prefix
> lengths also creates twice as many opportunities for mistakes as a
> single prefix length. A single prefix length means reduced end-site
> addressing costs, which will likely be a fairly significant indirect
> benefit that can (or may, depending on the business people :-) ) be passed
> on to all end-sites.

I think having 2 different end site prefix lengths (for a given end
site) is (generally) needlessly complicated and should be
discouraged. That is why the bar for getting more address space needs
to be reasonably low. Indeed, 3177bis does say:

   Looking back at some of the original motivations behind the /48
   recommendation [RFC3177], there were three main concerns. The first
   motivation was to ensure that end sites could easily obtain
   sufficient address space without having to "jump through hoops" to do
   so. For example, if someone felt they needed more space, just the act
   of asking would at some level be sufficient justification.

And:

     - assigning a longer prefix to an end site, compared with the
        existing prefixes the end site already has assigned to it, is
        likely to increase operational costs and complexity for the end
        site, with insufficient benefit to anyone.

But, that needs to be balanced against someone asking for a /48 when
they clearly only have a handful of subnets today (if even more than
one) and there is no credible scenario for them needing even a hundred
for the next ten years.

The other part that needs to be remembered is that getting an end site
assignment today does NOT lock one into that size forever. RIR
policies need to be (and currently are) flexible enough that if one
really can justify a need for more space, additional space can easily
be obtained.

We should not be looking at the end site assignment that one gets
today as what an end site will be stuck with forever.

Thomas