[rpd] Factors affecting in-region utilization - way forward?

[Stephen Wilcox]

On 22 July 2014 23:56, Mark Tinka <mark.tinka at seacom.mu> wrote:

> On Monday, July 21, 2014 02:34:31 PM Stephen Wilcox wrote:
>
> > My point is being accepted by an upstream != globally
> > routable. And internal use of long prefixes (which is
> > what I think you are seeing in route-views) doesn't
> > count. What I see is a lot of examples of (very bad
> > netiquette) deaggregation being performed by the same
> > handful of ASNs...
>
> And my point was if a "Really Big Global Carrier" is
> accepting and routing longer-than-/24's and /48's in their
> network (which reasonably well-connected and peers with
> other large carriers who have a similar persuasion), there
> is a higher (but still remote) chance that some portion of
> those longer subnets could be routable.
>
> Not very globally, but the intermittent performance would
> certainly raise enough frustration for some operators to
> seriously consider this.
>
> What I'm saying is - don't even do it. Doing it half-way is
> worse than not doing it at all, because it sets expectations
> that are difficult to undo.


Doing it 80% is enough to perceive a massive problem with your Internet
access.. it needs to be 99%+ or its not worth considering...

Steve
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.afrinic.net/pipermail/rpd/attachments/20140722/f11caed6/attachment.html>