RE: Continuation...
- Date: Tue, 27 May 2003 10:09:11 +0100
At 10:17 27/05/2003 +1000, Nurani Nimpuno wrote:
>What about the following way to tackle it:
>
> - we change the PA policy back to "no initial usage to be documented"
> [- eventually reducing initial allocation size to a /21]
Hm. I'm not sure if I follow you Gert. In what way is it better to give
an organisation a PA allocation rather than a PI assignment? They both
add the same amount of routes to the routing table. The only difference
is that people might get more than they need if they get an allocation.
If an organisation want's to assign IPs to it's customers, then it
cannot have PI space though.
How many people apply for PA allocations who don't also
have an AS number and are not RIPE LIRs and are not
running BGP networks with many customers hanging off
them? I'd guess 0.
If someone wants about 500 "portable" addresses, only for their own
network, that they can multihome with, isn't it better that they get a
/23 PI assignment than a /20 PA alloction...? After all, PI is *in
itself* not bad for the routing table. It's the concept of everyone
wanting to be independent and multihome (as well as people chopping up
their PA allocations) that adds to the growing global routing table.
But of course, everyone wanting to be independent & multihome
is a good thing, not a bad thing :-)
The other problem is that:
- pi space is for your own machines
- pa space can be allocated to your customers machines.
But what happens if you don't meet the requirement for a /20 PA
allocation but do want to allocate address space to other people?
Currently, RIPE suggest that we apply for PI space, as do our
customers, which is silly, given that most customers wouldn't
know a bit of ethernet if it jumped up and electrocuted them.
So, particularly for PA space, since the people applying for PA
space are likely to be applying for it as part of an overall new-LIR
registration so they can get an AS, break BGP, allocate to
customers, etc. there is a need to permit smaller PA assignment
windows, because not everyone has the requisite number of
computers to justify the current PA requirements.
(In my orgs case) we were just over the PA assignment window
and if we weren't, then we would have had about 10 small
PI allocation and a couple of big PI allocations and thus the
world would have had 12 routes (of unrouteable space due to
filtering) rather than just 1 big route.
In this instance, I would argue that it is much better to have
one PA assignment than multiple PI allocations.
I think that Kurts suggestion, perhaps combined with a minimum PI
assignment size of /24 or /23 (to ensure routeability) would be very
helpful.
cheers
peter
