Re: [Status] SPRING Charter

Hi Jari,

> And I fear that somewhere down the line someone would decide all we need is
> inside/outside security model, and I do not think that helps enough in this case

For one I am not convinced that inside/outside security model is not
good enough for the problem at hand.

Is your point perhaps more focused towards clear placement of such
boundary ? If so I think it's get more involved indeed as deployment
models may vary. But once such boundary is set I am not sure why it
would not be sufficient for SR architecture.

Specifically we already have defined IGP as the carrier of SIDs. You
normally do not run IGP (passive interfaces excluded) outside of your
domain boundary. That plus ACL towards/against infrastructure
addresses IMHO should provide sufficient isolation.

Best regards,
R.

On Wed, Oct 16, 2013 at 8:38 PM, Jari Arkko <jari.arkko@piuha.net> wrote:
> I'd suggest that the distinction between trusted and non-trusted or host and router is not as useful as we might expect.
>
> What I was trying to suggest with my text was specifying a requirement for a technical mechanism that helps prevent inappropriate headers from being believed. Trusted/non-trusted in itself would not be as useful, until it results in a similar specific requirement. And I fear that somewhere down the line someone would decide all we need is inside/outside security model, and I do not think that helps enough in this case.
>
> But I don't mind the new text if the old text is still also there….
>
> Jari
>
> _______________________________________________
> status mailing list
> status@ietf.org
> https://www.ietf.org/mailman/listinfo/status