North American Network Operators Group|
Date Prev | Date Next |
Date Index |
Thread Index |
Author Index |
Re: Guts (Was: Drivel about BCP38, et al.)
- From: Paul Vixie
- Date: Mon Oct 11 22:00:30 2004
firstname.lastname@example.org ("Fergie (Paul Ferguson)") writes:
> The Internet has almost becoem unusable because of this poor
> state of affairs -- the dispruptors have beaten you?
> Reluctantly, sadly,
It's not the engineers. Those who still have jobs Really Want to do the
Right Thing. However, they're all pretty busy keeping their networks alive
after their eleventh-in-a-row quarterly RIF. So, "the CFO has beaten them."
OTOH, I agree. I didn't jump on Sean and smack him around when he whined
this time, because I think anybody who wants to see Sean get what's coming
to him can just check the archives. But truly, the whinage in here is ugly.
BCP38 comes in many forms (thanks, DTS, for s'plaining that uPRF is just one
of the forms it can take.) There's even partial compliance which is better
than no compliance. What's so damned difficult about an input-ACL on the
router that stands between a DSL cloud and the backbone, of the form "if
it's not among the pools this cloud allocates from, drop it on the floor" ??
(Season as nec'y for colo rooms, T1 hubs, wireless clouds, dialup, and so on.)
If someone complains to you that they need to be able to spoof, then cancel
their contract, refund their money, and let them go elsewhere. It's 2004,
and the small number of valid reasons to spoof can be accomplished on
research networks, and need not be done in last-mile end-user loops, and
being able to do it at all is an expensive luxury -- a dropping tide has
lowered all boats.