Re: [Idr] WGLC on draft-ietf-idr-as-private-reservation-00

[Jon Mitchell <[e-mail]>]

On Thu, Dec 20, 2012 at 08:19:29PM -0500, Jeff Wheeler wrote:
> On Thu, Dec 20, 2012 at 6:24 PM, Susan Hares <[e-mail]> wrote:
> > I'd be interested in you suggesting the situations would be valid to leak
> > the AS private path AS to a peer, before we downgrade to a "SHOULD".
> 
> This is happening in the DFZ today.  You don't have to take my word
> for it.  Log onto a router and check.  You can even find 65535 in DFZ
> AS_PATHs even though that is not a valid ASN!

I think we are all aware of this.

> 
> I think it's called a Private ASN and by its very nature, it ought not
> appear in a DFZ AS_PATH.
> 
> However, since some networks are actually doing this right now,
> perhaps it is worth asking them why.
> 

Isn't this like asking why folks leak RFC 1918 space into the global
routing table, /32's into the routing table, etc... ?  Mis-configuration
seems obvious, but I don't think a poll is a fruitful exercise.

I think because something happens today due to accidental (or poorly
designed networks that make fixing the issue difficult) doesn't mean we
should soften language that states operators (if they strive to be RFC
compliant) MUST or MUST NOT do something.  One of the reasons why it is
not fixed in these cases is likely that the impact is so low to the
originator (who may have a covering route) or the receiver.