Re: [6lo] Draft applicability for 6775bis

Any issue with a standards track document having a normative reference to a BCP?
If so, then could reword so that the MUST NOT is not in the same sentence as the BCP reference, e.g.

> However, the ability to return errors to address registrations MUST NOT be
> used to restrict the ability of hosts to form and use addresses.  See [RFC7934]
> for further discussion.

If there is no issue in having a normative reference to a BCP then I think Erik's text is fine.

Dave

> -----Original Message-----
> From: 6lo [mailto:6lo-bounces@ietf.org] On Behalf Of Erik Nordmark
> Sent: Wednesday, March 29, 2017 10:50 AM
> To: Lorenzo Colitti <lorenzo@google.com>; huitema
> <huitema@huitema.net>
> Cc: 6lo@ietf.org
> Subject: [6lo] Draft applicability for 6775bis
> 
> 
> Here is an attempt at an applicability statement based on what we talked
> about today.
> It is sufficiently strong?
> Other RFCs or drafts that we should reference?
> 
>     Erik
> 
> ----
> 
> Applicability
> 
> The purpose of the ARO and EARO is to facilitate duplicate address detection
> for hosts and pre-populate NCEs in the routers to reduce the need for
> sending multicast neighbor solicitations and also to be able to support
> backbone routers.
> 
> In some cases the address registration can fail or be useless for reasons other
> than a duplicate address. Example are the router having run out of space, the
> host having a stale sequence number, or the host is using an address which
> does not match the prefix(es) for the link. In such cases the host will receive
> an error to help diagnose the issue and retry.
> 
> However, the ability to return errors to address registrations MUST NOT be
> used to restrict the ability of hosts to form and use addresses as specified in
> [RFC7934]. In particular, this is needed for enhanced privacy, which implies
> that each host will register a multiplicity of address as part mechanisms like
> [RFC4941]. This implies that a 6LR or 6LBR which is intended to support N
> hosts MUST have space to register at least on the order of 10N IPv6
> addresses.
> 
> ---
> 
> _______________________________________________
> 6lo mailing list
> 6lo@ietf.org
> https://www.ietf.org/mailman/listinfo/6lo