Hi Nick,

The intent is to document observed behavior and its consequences and to provide 
operational guidance. The references to prepending beyond 5 ASNs reflect 
commonly observed practice and outcomes rather than a hard threshold. 

If there isn't consensus to publish as a BCP perhaps we should move it to 
Informational.

Thanks,
mike


-----Original Message-----
From: Nick Hilliard <[email protected]> 
Sent: Wednesday, April 15, 2026 3:19 PM
To: David Farmer <[email protected]>
Cc: Doug Madory <[email protected]>; [email protected]
Subject: [GROW] Re: I-D Action: draft-ietf-grow-as-path-prepending-19.txt

David Farmer wrote on 15/04/2026 20:45:
> Given the prevalence of seemingly excessive numbers of AS prepends on 
> the Internet, this guidance seems appropriate.

This is the point: how would the WG define "excessive"? A baseline of having no 
prepends won't stop hijacking from happening, and also won't stop the hijacking 
from having an impact on traffic destined to the network in question. Even 
adding a single prepend will have an additional impact, and there's an argument 
to be made that even a single prepend is in excess of what you would want to 
aim for, if it causes additional service impact.

Fundamentally, wide-area distance vector routing algorithms work on the basis 
that external networks trust the total distance, so anything that adds to an 
already low single-digit metric should be considered harmful. 
But I don't see this as being useful advice to give in an RFC.

That said, all this takes the focus away from the reality that it's not the 
prepending that causes the problems: it's the hijacking, and whether you 
prepend or not, hijacking is going to have an impact on service.

Nick

_______________________________________________
GROW mailing list -- [email protected]
To unsubscribe send an email to [email protected]
_______________________________________________
GROW mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to