[ppml] Technical reason why /52,/56,/60,/64 are bad

[Azinger, Marla]

Robin-  

This proposal (IPv6 Guidelines)is not saying that PA space will be re-assigned or re-allocated by a provider for bgp routing at these subnet sizes.  This is only attempting to provide a clearer "how to decide what subnet size to re-assign or re-allocte".

What subnet size should be permitted to route bgp and enable multihoming and TE is a seperate subject.  It is easily confused within the discussions since PI space is registered within a "special block" that some people have willingly opened up filters to these smaller/more specific subnets that fall within the "PI special block range".  And so far in regards to PA space, the perception that filters should only allow /32 through for bgp routing is being accepted.  This difference between PA and PI is where alot of the confusion comes from.

Hope that helps clarify it for you.
Cheers!
Marla


-----Original Message-----
From: ppml-bounces at arin.net [mailto:ppml-bounces at arin.net]On Behalf Of
Robin Whittle
Sent: Sunday, August 19, 2007 6:30 PM
To: ppml at arin.net
Subject: Re: [ppml] Technical reason why /52,/56,/60,/64 are bad


Michael Dillon quoted something I wrote on the IETF list, in
response to the PPML message:

   http://lists.arin.net/pipermail/ppml/2007-August/008521.html

It was my initial impression that these longer prefixes, to /64,
were being assigned (allocated? - I get confused with the
terminology) by RIRs as PA space for end-users.  That would mean
that there would be BGP advertised prefixes of this length, with the
consequent need for all BGP routers to process 64 bits of
destination address of some packets in their FIBs.

After I wrote my first message, which Michael quoted, I read the
PPML message more carefully and saw that it concerned LIRs
(effectively ISPs, I think) providing PI space for their customers.
 This would mean that the /64 prefixes would not be advertised in
BGP.  So I wrote a retraction to the IETF list:

  http://www1.ietf.org/mail-archive/web/ietf/current/msg47249.html

Can someone confirm my second understanding is correct?

 - Robin

_______________________________________________
PPML
You are receiving this message because you are subscribed to the ARIN Public Policy
Mailing List (PPML at arin.net).
Unsubscribe or manage your mailing list subscription at:
http://lists.arin.net/mailman/listinfo/ppml Please contact the ARIN Member Services
Help Desk at info at arin.net if you experience any issues.