<<< Chronological >>> Author Index    Subject Index <<< Threads >>>

Re: Changes to PI Policy?

  • From: leo vegoda < >
  • Date: Thu, 17 Apr 2003 17:29:10 +0200

Hi Peter,

Peter Gradwell peter@localhost writes:
At 20:17 15/04/2003 +0200, you wrote:
Kurtis has stated that he is worried that the policy requires the
RIPE NCC to assign address space that can be "useless". Gert has
stated that the concept of PI space is broken.
The concept of PI space is definitely not broken. There is a
very real need for organisations, large and small, to be able
to connect to multiple Internet providers
Multi-homing can be done without PI address space. This is something
we point out to everyone requesting PI address space. The specific
text requesters are sent is:

1. Why is PI space required rather than PA space?

- If PI is requested for multi-homing please explain why the
second provider cannot route PA space as a more specific
route (with the PA block holder adding a more specific route
too).

ARIN actually have a policy just to cover cases like this. "Ratified Policy 2001-2: Reassignments to multihomed downstream customers" was approved in the ARIN region last year. The core policy sentence is:

This policy allows a downstream customer's multi-homing requirement
to serve as justification for a /24 reassignment from their upstream
ISP, regardless of host requirements."

This is obviously much less convenient for the registrant when changing ISPs. It means that they may feel 'tied' to the ISP whose address space they use. The setup is likely to be quick(er) and eas(y/ier).

That being said, networks use PI space for a number of reasons. Apart from multi-homing, people often state that they just don't want to have to renumber if they switch ISP.

Finally, we sometimes get requests from organisations that are creating large internets that don't connect to "The Internet". They use unique addresses to avoid RFC19181 address clashes.

Have a great holiday weekend, y'all,

--
leo vegoda
RIPE NCC
Registration Services




  • Post To The List:
<<< Chronological >>> Author    Subject <<< Threads >>>