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

Re: Changes to PI Policy?

  • To: Peter Gradwell < >
  • From: Kurt Erik Lindqvist < >
  • Date: Sat, 19 Apr 2003 09:56:06 +0200
  • Cc: leo vegoda < >


On torsdag, apr 17, 2003, at 18:09 Europe/Stockholm, Peter Gradwell wrote:

At 17:29 17/04/2003 +0200, leo vegoda wrote:
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.
But only if both providers agree, and both providers will not agree
or cooperate because it causes them to completely bugger their
network configurations.

ISPs thus suggest that $customer gets PI space.
From what I know this is actually pretty common practice - which is why I suggested it as a quick start for IPv6 multihoming.


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).
Absolutely. If an organisation wants to be connected to multiple upstreams then they presumably want to be provider independent. So, telling them to poke holes in one ISP's fabric is not going to help.
No. People could also connect to multiple providers in order to get redundancy. Actually I that is the most common reason.


The solution has to be to look at it from the other perspective and make everyone provider independent. In the interim, I think we should just look at stretching out the current address space so it supports legitimate uses and predicted capacity requirements.
This is research and again multi6 or IRTF.


- kurtis -





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