I'd consider that the non-controversial set are listed in Bill Manning's draft. Adding to that adds potential for controversy... Mind you, documenting in a generic sense what providers do might be a good idea for the sake of accurate implementation. It would have to be clearly stated that this is documenting current practice by some and is in no way a recommendation to follow suit. As I've said before, there are people who simply cut and paste configurations.... philip -- At 14:11 27/06/00 +0100, philip bridge wrote:
At 12:29 27.06.00 +0200, Lars Marowsky-Bree wrote:
On 2000-06-27T12:25:34, Fredrik Rosenbecker <fredrik.rosenbecker@ip-only.net> said:
Yes, such a document could be useful. We're using a prefix-list as shown below. We could use it as a starting-point to establish a standard prefix filter, or as Randy put it, a prefix filter technical coding advice.
What is also very helpful is to point out the usefulness of anti-spoofing rules - ie filter your own netblocks when they are announced to you, _including_ more specific routes ;-)
Hm. Yes. But this gets us into a controversial issue since filtering more specifics from your own netblocks means your multihomed customers don't get connectivity to you when the line between you and them is down.
I am a bit worried about expanding the scope of this to the point where we get bogged down in arguments about such practices. It seems to me that there is a large body of non-controversial practices that simply require collation and documentation in order to acheive a major effect. We should stick to those, at least in the first instance.
Phil
Sincerely, Lars Marowsky-Br�e <lmb@suse.de>
-- Perfection is our goal, excellence will be tolerated. -- J. Yahl
Philip Bridge Nextra (Schweiz) AG <www.nextra.ch> Tel: +41 031 985 88 06 / Mobile: +41 79 659 75 50 E-Mail: <mailto:bridge@nextra.ch> Disclaimer: <http://www.nextra.ch/signature_nw.html>