[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [GLOBAL-V6] PI space under IPv6
On 2007-01-17 17:42, Gert Doering wrote:
> Hi,
>
> On Wed, Jan 17, 2007 at 05:38:14PM +0100, Brian E Carpenter wrote:
>>> Calling it "PI" or calling it "we open the policy so everybody can get
>>> their own independent address block" is technically pretty much the
>>> same.
>> Yes, technically it's called the swamp.
>
> No. The swamp is "everybody just gets address space, without clear
> rules who and why".
Why do RIR's then allocated multiple blocks to 1 single company!?
(below a 'random' but rather good example')
2001:588::/32
2001:600::/32
2001:4218::/32
2001:4440::/32
2001:4441::/32
2001:4442::/32
2001:408::/32
2001:506::/48
2404:e0::/28
2600:800::/27
That is 1 company that got these blocks. Yes they are a pretty huge
company (and most likely they might split at some point causing
deaggregation anyway if they where allocated one block). But those
2001:444x::/32 is completely silly on APNIC's part.
What use does it have to call that "PA" when the RIR already breaks it up ?
Especially the ARIN "Micro-allocations for internal infrastructure" is
funny in this context, why the *PEEP* does one need a single /48 when
one has 3x /28 and 7x /32 available?
These allocation procedures do match what I say about all of it: Give
address space to organizations that need it. It does FAR from conserve
routing table slots though, which indeed is not something the RIR's are
involved with, but they should of course try a little IMHO.
Greets,
Jeroen
Attachment:
signature.asc
Description: OpenPGP digital signature
