On May 13, 2014, at 2:22 PM, Jeffrey Haas <[email protected]> wrote:

> Wes covered most of the points I would raise - missing security
> considerations in particular.  Basically, if you filter using IRR and the
> IRR has issues (perhaps maliciously induced), you'll have routing issues.
> The entire first part of the document does very well at outlining where the
> issues are and some of the paths for fixing them.

That makes sense to me, there are external dependencies and if you use them you 
better have some sort of buffer or things could be really ugly.  Are there 
other things you’d like to see included here?

> In general, once the security considerations are addressed, this makes a
> fine Informational document and I think it's ready to go.  What's
> frustrating is that the state of the art hasn't changed much since I was
> doing IRR related work in 2000.

Or me in 94/95, for that matter — wohoo, two decades!  Except that the routing 
system was more secure then, perhaps, because it was less densely 
interconnected and fewer meatheads had keyboard access to BGP speaking routers..

-danny

Attachment: signature.asc
Description: Message signed with OpenPGP using GPGMail

_______________________________________________
GROW mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/grow

Reply via email to