<div dir="ltr"><div dir="ltr"><br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, Mar 22, 2022 at 4:17 PM John Santos <<a href="mailto:john@egh.com">john@egh.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">We are doing exactly that: exchanging routing information with one of our <br>
customers using BGP routing over a private network (multiple, redundant VPN <br>
circuits using different ISPs at our end and two different geographically <br>
dispersed sites at our customer's end.) We are using one of the private ASNs, <br>
which works fine for our purposes, but if multiple customers wanted to do the <br>
same thing, we might need to avoid colliding the private ASNs to prevent <br>
route-through to our customers' other vendors or from one customer to another.<br>
<br>
Only one customer has dynamic networking, with multiple possible routes; all our <br>
other customers are configured with static routing.<br>
<br>
So right now we don't need an assigned ASN, but we might someday.<br></blockquote><div><br></div><div><br></div><div>Which means there's a plausible use case for avoiding private ASN's much like avoiding using RFC 1918 (or squatting). <br></div><div> <br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<br>
There is no "Upstream" involved. the networks are peer-to-peer. I think this is<br>
(or would be if there were more peers) an example of what Andrew is describing.<br>
<br>
I think Owen is saying the two items (first and third of the reasons to justify <br>
a public ASN) are equivalent and the third is fewer words and easier to <br>
understand. The 1st reason doesn't apply if there is no Upstream, so they <br>
aren't equivalent. However, I'm not sure the first reason is a proper subset of <br>
the third. (If it were, then the first reason would be redundant.)<br>
<br>
Don't know if I'm clarifying the question or muddying the waters.<br></blockquote><div><br></div><div>Owen has been in the room for the "Is it partial transit, paid peering, on-net peering, selective peering or partial or full transit" arguments. 'Peer' encompasses all of them as I understood what he wrote fully.<br></div><div><br></div><div>If one really understands the business they know that every ISP, CDN, cloud and other network _gladly_ accepts* any single-homed network that is willing to engage in the service.<br></div><div><br></div><div>Warm regards,</div><div><br></div><div>-M<</div><div><br></div><div>*please don't go down a rat hole of "any". If you know, you know.<br></div><div><br></div><div><br></div><div> <br></div></div></div>