Ettiquette and rules regarding Hijacked ASN's or IP space?

Richard A Steenbergen ras at
Mon Jun 9 19:59:22 UTC 2003

On Mon, Jun 09, 2003 at 01:04:22PM -0400, Andy Dills wrote:
> On Mon, 9 Jun 2003, Christopher L. Morrow wrote:
> > excellent point :) the distinction between 'good' and 'bad' was just
> > non-abuser/abuser. Certianly ARIN's requirements for ASN ownership are
> > simple enough, be multihomed and have a 'unique' routing policy. If you
> > need an ASN likely you are already multihomed and have a 'unique' routing
> > policy, eh?
> It's not even THAT difficult...all you have to be is multihomed _or_ have
> a 'unique' routing policy.
> Being multihomed by itself is trivial and plenty of justification...does
> anybody have some examples of 'unique' routing policies, that require
> ASNs, that don't require or imply multihoming? For example, while
> anycasting is a good example of a potential use of an ASN without
> requiring multihoming, it's kind of implied that they're at least
> purchasing transit from multiple organizations (if not truly multihomed)
> and could easily justify an ASN without having to specify their unique
> routing policy.
> What sorts of 'unique' routing policies justify an ASN?

Anything weird, bizaare, or different. Like once every year when some
ip/colo provider decides they want to sell local peering routes or want to
give every datacenter an ASN, or when some route optimization company
decides they need a huge block of ASNs for...well...nevermind, or when
someone decides that they need a special ASN dedicated to acting as a
border between their reserved asn customers and the rest of the world...

Richard A Steenbergen <ras at>
GPG Key ID: 0xF8B12CBC (7535 7F59 8204 ED1F CC1C 53AF 4C41 5ECA F8B1 2CBC)

More information about the NANOG mailing list