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
