[6lo] Comments on draft-ietf-6lo-btle-00

Dave Thaler <dthaler@microsoft.com> Wed, 26 March 2014 00:15 UTC

From: Dave Thaler <dthaler@microsoft.com>
To: "draft-ietf-6lo-btle@tools.ietf.org" <draft-ietf-6lo-btle@tools.ietf.org>
Thread-Topic: Comments on draft-ietf-6lo-btle-00
Thread-Index: Ac9IgprZdpBKHuPxR0ij8vNKky8yVw==
Date: Wed, 26 Mar 2014 00:15:31 +0000
Message-ID: <529718aa87a44aae8cd0124670f926ed@BY2PR03MB412.namprd03.prod.outlook.com>
Accept-Language: en-US
Content-Language: en-US
received-spf: None (: microsoft.com does not designate permitted sender hosts)
Content-Type: multipart/alternative; boundary="_000_529718aa87a44aae8cd0124670f926edBY2PR03MB412namprd03pro_"
MIME-Version: 1.0
Archived-At: http://mailarchive.ietf.org/arch/msg/6lo/YcK7_tFYFLDCk-cKLzwcEAtwEW4
Cc: "6lo@ietf.org" <6lo@ietf.org>
Subject: [6lo] Comments on draft-ietf-6lo-btle-00
Precedence: list

Here's my comments on this draft...

I find the lack of consistency between Figures 1 and 3 to be confusing.
For example, figure 1 has "Logical Link Control and Adaptation" but figure 3 has
"BT-LE L2CAP".  I assume these are the same thing but are using gratuitously
different labels.   Another example is that figure 1 shows "Host Controller
Interface" but figure 3 does not, and I cannot tell why.  I would expect the
bottom half of both figures to look identical, or there to be an explanation
in the text of why they look different.

Section 3.2.1:
> Alternatively, a randomly generated IID (see Section 3.2.2), MAY be used instead.

draft-ietf-6man-default-iids would make that MAY be a SHOULD, and it would be
good to not have contradicting documents between WGs. It would be good to find
wording that defers to 6MAN on the appropriate level. An example of such
wording might be:

"Alternatively, a randomly generated IID (see Section 3.2.2), can be used instead
as discussed in [I-D.ietf-6man-default-iids]."

Section 3.2.2:
> A BT-LE 6LN MUST register its address with the 6LBR by sending a
> Neighbor Solicitation (NS) message with the ARO option and process
> the Neighbor Advertisement (NA) accordingly.

Does that mean RFC 4429 (Optimistic DAD) is legal or illegal?

> NS with the ARO
> option SHOULD be sent irrespective of whether the IID is derived from
> the unique 48 bit BT-LE device address or the IID is a random value
> that is generated as per the privacy extensions for stateless address
> autoconfiguration [RFC4941].

Those aren't the only two possibilities.  What about CGAs (RFC3972)?
What about draft-gont-6man-stable-privacy-addresses?  What if you
need to use DHCPv6?   Reword to not attempt to enumerate possibilities,
such as simply "... irrespective of the method used to generate the IID."

> Although RFC 4941 [RFC4941] permits the
> use of deprecated addresses for old connections, in this
> specification we mandate that one interface MUST NOT use more than
> one IID at any one time.

Why?  This seems to be the biggest issue in the spec.  That would
seem to preclude the use of both temporary addresses and CGAs, and
hence introduce security/privacy issues that are missing from the
Security Considerations section.  The privacy problems are covered
in draft-ietf-6man-ipv6-address-generation-privacy.

Section 3.2.4:
> As described in the
> Section 3.2 the master will not forward link-local multicast messages
> to other slaves connected to the master.

This is unclear. Is the link presented to IP as a multicast-capable interface
or not?  If not, then "link-local" doesn't belong in the quote and of course a bunch of protocols just won't work
(just like they don't work on any other NBMA link).

It seems odd to say that it will forward multicast messages of
scopes greater than link local, but not forward link-local ones.

-Dave

[6lo] Comments on draft-ietf-6lo-btle-00 Dave Thaler
Re: [6lo] Comments on draft-ietf-6lo-btle-00 teemu.savolainen
Re: [6lo] Comments on draft-ietf-6lo-btle-00 Carsten Bormann
Re: [6lo] Comments on draft-ietf-6lo-btle-00 teemu.savolainen
Re: [6lo] Comments on draft-ietf-6lo-btle-00 Abdussalam Baryun
Re: [6lo] Comments on draft-ietf-6lo-btle-00 teemu.savolainen
Re: [6lo] Comments on draft-ietf-6lo-btle-00 Dave Thaler
Re: [6lo] Comments on draft-ietf-6lo-btle-00 Timothy J. Salo