Re: A common problem with SLAAC in "renumbering" scenarios

[Fernando Gont <[e-mail]>]

On 20/2/19 03:02, JORDI PALET MARTINEZ wrote:
> Of course, not acceptable. I'm not saying that is not a problem to be
> solved, just tried to make the data more precise. My reading of your
> text was like "dynamic is over static".

My text essentially was: "No matter what's your personal take on the
static vs dynamic debate, dynamic prefixes represent a significant part
of the deployed IPv6 Internet. Hence having such deployments fail is
unacceptable".

Much of what I've seen on this thread has been about who to blame: CPEs
for what they ship, ISPs because they employ dynamic prefixes, etc.

Reality is that there are engineering trade-offs for what they do.
There's a significant part of the IPv6 Internet that exhibit the problem
scenario we describe, and that we have contributed our bit for the
problem to exist.

I'd rather focus on improving the current state of affairs and making
the network more robust, rather than pretending the problem doesn't
exist or pretending that the folks that make our "words on paper" a
reality are just clueless people.


-- 
Fernando Gont
SI6 Networks
e-mail: [e-mail]
PGP Fingerprint: 6666 31C6 D484 63B2 8FB1 E3C4 AE25 0D55 1D4E 7492