On Jan 22, 9:45am, Ian Mason wrote:

...

I don't think there are any sacred cows for any of these here - if we adjust the LINX policy to meet requirements of the root NS rather than the other way around, I don't think we break either of 1 or 2. I don't think it breaks 3. either, but this requirement will go away when we have a route server in any case.

If we opt for the root NS to have its own AS then there is no barrier to people offering identical routing policy for the LINX and the NS. If it shares the LINX AS then it will be impossible to have different policy for LINX and the NS. So, if there is a possibility that there will ever be a requirement to have different routing policy for the NS and the LINX it would seem wise to give the NS its own AS.

It may be moot in as much as it's probably desireable to achieve 'root server connectivity quality' for the LINX AS, but the seperate AS for the root server would preserve some flexibility.

I agree, but at the cost of an additional router. Due to the nature of the full routing we need for theses AS(es), a 64Mb router is a requirement eventually anyway, but it is not a LINX budget item at present, and while there have been lots of offers of server hardware for the root NS, there have been none to date of a router. I suggest we make the ASes the same for now, but split them when a clear requirement arises. At that point we will hopefully have a route server box to take over various of the collector functions. Keith