[AfriNIC-rpd] Opposition to AFPUB-2010-v4-005-draft-01

[Hannigan, Martin]

Andrew,

True, true, and true. But network operators will ultimately decide what the minimum prefix length will be related to routing table performance, not Afrinic or any other RIR. From a policy perspective, this is not relevant. Some networks already accept prefixes longer than /24 from peers today. From a practicality standpoint, it’s relevant but I would argue that it’s somewhat premature, at least in the Afrinic region, to start talking about assigning anything longer than /24’s.

Best,

-M<






On 2/28/11 5:54 PM, "Andrew Alston" <aa at tenet.ac.za> wrote:

Just one other thing I thought about as I was climbing into bed :)

In a world where /24 is the minimum allocation size and that’s whats globally filtered on, when someone hijacks space out of a larger block, they announce a more specific and if their announcement is accepted globally, they have pretty much successfully hijacked the space for a while (similar to what happened to youtube a while back).

Now, at the moment, if you deaggregate to /24s, it makes hijacking your space a little more difficult, since its equal prefix length size, the guy hijacking may get partial reachability, but the routes of equal length will be competing.

Drop the minimum allocation size to a /27, and to achieve the same effect, you’re gonna have to deaggregate to /27 size rather than down to /24 size.  Welcome to yet another reason this clause will bloat the routing table.

Thanks

Andrew



From: Daniam Henriques [mailto:daniam.henriques at gmail.com]
Sent: Tuesday, March 01, 2011 12:45 AM
To: Andrew Alston
Cc: McTim; Stacy Hughes; rpd at afrinic.net
Subject: Re: [AfriNIC-rpd] Opposition to AFPUB-2010-v4-005-draft-01



On Mon, Feb 28, 2011 at 11:52 PM, Andrew Alston <aa at tenet.ac.za> wrote:
Hi Tim,

As per my previous email, yes, it does, clause 3.5.2 lowers the minimum allocation size to a /27, which I think is hugely problematic.

Thanks

Andrew



-----Original Message-----
From: rpd-bounces at afrinic.net [mailto:rpd-bounces at afrinic.net] On Behalf Of McTim
Sent: Monday, February 28, 2011 11:22 PM
To: Stacy Hughes
Cc: rpd at afrinic.net
Subject: Re: [AfriNIC-rpd] Opposition to AFPUB-2010-v4-005-draft-01

Dear IP Goddess,

On Tue, Mar 1, 2011 at 12:07 AM, Stacy Hughes <ipgoddess.arin at gmail.com> wrote:
> Esteemed Colleagues,
> I must speak in opposition to this proposal.
> First, I am philosophically opposed to soft landing proposals in general.
>  When the party is over, it's time to go home.  We don't get 5 more minutes
> or more birthday cake.

Have you ever been to a 1st graders birthday party???....they get the
5 more minutes AND more cake...I experienced this first hand a few
days ago ;-/

  Incorporation of and transition to IPv6 is the way
> forward, and necessary for all of us.


Full ACK

>
> I especially disagree with direct assignments or allocations of IPv4 space
> in subnets of longer prefix lengths than /24.

Does this proposal do that?  If so, I must have missed that in this iteration.

Personally, I too would disagree with the allocation of prefixes that are greater than /24 in length.

While one could argue that filtering could be amended to facilitate such a requirement, a number of technical issues, some of which have been raised by Andrew Alston, exist which would mean that such a decision would have far reaching consequences related to current hardware and software limitations.

Individual and regional providers may be convinced to update their filters, however this is something which would require global buy-in with the major service providers and transit providers of the world. Without service providers agreeing to accept these longer prefixes, those that have been assigned these prefixes will be unable to make use of them in any meaningf

Additionally allowing /27 or greater subnets to be received within the global BGP table and to view such advertisements as acceptable will most likely promote further de-aggregation of the global table as various providers de-aggregate their existing subnets to achieve traffic engineering.

--
Cheers,

McTim
"A name indicates what we seek. An address indicates where it is. A
route indicates how we get there."  Jon Postel
_______________________________________________
rpd mailing list
rpd at afrinic.net
https://lists.afrinic.net/mailman/listinfo.cgi/rpd
_______________________________________________
rpd mailing list
rpd at afrinic.net
https://lists.afrinic.net/mailman/listinfo.cgi/rpd


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.afrinic.net/pipermail/rpd/attachments/20110228/a1dce7ea/attachment.html>