Dear Mohamed 

Many thanks for your kind review.

I’d like to address the DISCUSS first and the rest separately to be faster on 
that part.

Please see below :

> ----------------------------------------------------------------------
> DISCUSS:
> ----------------------------------------------------------------------
> 
> Bonjour Pascal,
> 
> Thanks for the effort put into this specification with many "surgical" changes
> that manipulate various pieces.
> 
> Special thanks for Section 3. Note that some of these considerations (e.g.,
> those discussing pre-requisites) are better moved to an operational
> considerations section.
> 

Good point I’ll see how I can do that.


> # DISCUSS
> 
> ## Consistency with 7400 and IANA registration
> 
> CURRENT:
>       0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
>      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>      |     Type      |   Length = 1  |   Reserved    |X|A|D|L|B|P|E|G|
>      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>      |F|                           Reserved                          |
>      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
> 
>                  Figure 4: New Capability Bit in the 6CIO
> 
>   New Option Field:
> 
>   *F:*  1-bit flag, set to 1 to indicate "Registration for prefixes
>      Supported"
> 
> ### Per
> https://www.iana.org/assignments/icmpv6-parameters/icmpv6-parameters.xhtml#sixlowpan-capability-bits,
> the request should indicate a bit position.

Yes. And actually we had that discussion with David (IANA) upon his early 
review.

Position 16 that you find in the figure is requested in section 13.2.

> 
> ### None of the entries in that registry updates 7400. Any specific reason why
> are we deviating from that practice?

I understand that you suggest to remove rfc 7400 from
the draft heading. I’m good with that.

> 
> ### The fields indicated as “Reserved” in Figure are marked as unassigned (not
> reserved) in RFC7400: “Bits marked by underscores in Figure 5 are unassigned
> and available for future assignment.”. Some consistency is needed here vs. 
> 7400
> 

The behavior described in section 3.4 of rfc 7400 is that of reserved fields 
(set to 0 and ignore upon receiving). I can remove the ´reserved’ in the 
picture and inherit the _ from rfc 7400 if that’s ok?

> ### Any reason why not any of the unassigned bits in the low range is used?
> 

David’s point. RFC 7400 makes those bits experimental.


> ## New EARO Prefix Length Field and F flag
> 
> CURRENT:
>      0                   1                   2                   3
>      0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
>     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>     |     Type      |     Length    |F|Prefix Length|    Opaque     |
>     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>     |r|C| P | I |R|T|     TID       |     Registration Lifetime     |
>     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
>     |                                                               |
>    ...                            ROVR                             ...
>     |                                                               |
>     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
> 
>            Figure 5: EARO Option Format for Use in NS Messages
> 
>   New and updated Option Fields:
> 
> ### Only PRT are defined in
> https://www.iana.org/assignments/icmpv6-parameters/icmpv6-parameters.xhtml#icmpv6-adress-registration-option-flags.
> Where the flags are defined?

R is unassigned. C is an oversight from RFC 8928 that we are fixing with 
https://datatracker.ietf.org/doc/html/draft-ietf-6lo-updating-rfc-8928-02

Both RFCs should be in the same cluster and have consecutive numbers.

> 
> ### RFC8505 says “MUST be set to 0 in NS messages”. How a legacy receiver will
> handle this updated EARO option? Will it be ignored? Rejected? Please consider
> adding some considerations to the backward compatibility section. Of course,
> adding a pointer to where this is already described would be sufficient. 
> Thanks.
> 

That is the point of using RFC 7400 above. The receiver has indicated that it 
supports the extension else the sender does not use it. Should I clarify that ? 
A new section seems overkill.

What would you suggest ?

Pascal



> 
> ----------------------------------------------------------------------
> COMMENT:
> ----------------------------------------------------------------------
> 
> # Consistency with Your Own Terminology on Amend/Extend
> 
> For example,
> 
> OLD: 4. Updating RFC 4861
> NEW: 4. Amending RFC 4861
> 
> # Section 5:
> 
> CURRENT:
>   5.  Extending RFC 7400
> 
> This is about associating a meaning with an unassigned value in a registry
> managed by IANA, not updating 7400.
> 
> The document says "[RFC7400] was already extended by [RFC8505]" but it does so
> without updating 7400.
> 
> # Section 7.1: Update 8505
> 
> The values are handled in an IANA registry:
> https://www.iana.org/assignments/icmpv6-parameters/icmpv6-parameters.xhtml#p-field-values.
> Nothing is updated in 8505. Also, this field is defined in rfc9685, not 8505.
> 
> Why is this subsection provided under “Update 8505”?
> 
> # Section 12
> 
> Consider adding some deployment considerations. For example, how the various
> extensions are expected to be added to a network that is composed of nodes
> compliant with existing RFCs?
> 
> # NITS
> 
> ## Abstract
> 
> ### nits, expand acronyms, etc.
> 
> OLD:
>   This document updates IPv6 Neighbor Discovery RFC4861 and the 6LoWPAN
>   extensions (RFC8505, RFC8928, RFC8929, RFC7400) to enable a node that
>   owns or is directly connected to a prefix to register that prefix to
>   neighbor routers.  The registration indicates that the registered
>   prefix can be reached via the advertising node without a loop.  The
>   unicast prefix registration allows to request neighbor router(s) to
>   redistribute the prefix in a larger routing domain regardless of the
>   routing protocol used in the larger domain.  This document extends
>   RPL (RFC6550, RFC6553, RFC9010) to enable the 6LR to inject the
>   registered prefix in RPL.
> 
> NEW:
>   This document updates IPv6 Neighbor Discovery (RFC 4861) and the 6LoWPAN
>   extensions (RFC 8505, RFC 8928, RFC 8929, and RFC 7400) to enable a node 
> that
>   owns or is directly connected to a prefix to register that prefix to
>   neighbor routers.  The registration indicates that the registered
>   prefix can be reached via the advertising node without a loop.  The
>   unicast prefix registration allows to request neighbor router(s) to
>   redistribute the prefix in a larger routing domain regardless of the
>   routing protocol used in that domain.  This document extends
>   Routing Protocol for Low-Power and Lossy Networks (RPL) (RFC 6550,
>   RFC 6553, and RFC 9010) to enable a 6LoWPAN Router (6LR) to inject the
>   registered prefix in RPL.
> 
> ### Large routing domain
> 
> What does that mean? Do we really need to mention that for injecting routes?
> 
> I would avoid that as this is a distraction at this stage.
> 
> # Introduction
> 
> ### Stimulated
> 
> CURRENT:
>   *  Unicast host to router operations stimulated by the host and its
>      applications.
> 
> What does "stimulated” mean here? Do we mean “triggered”?
> 
> ### (nits) There are many instance of 6LN/6LR
> 
> Consider making these changes:
> 
> OLD:
>      unicast Neighbor Solicitation (NS) and Neighbor Advertisement (NA)
>      messages between the 6LN and the 6LR.
> 
> NEW:
>      unicast Neighbor Solicitation (NS) and Neighbor Advertisement (NA)
>      messages between a 6LN and a 6LR.
> 
> OLD:
>      [RFC9010] to enable the 6LR to inject the anycast and multicast
>      addresses in RPL.  Similarly, this specification extends [RFC8505]
>      and [RFC9010] to add the capability for the 6LN to register
>      unicast prefixes as opposed to addresses, and to signal in a
>      routing-protocol-independent fashion to the 6LR that it is
>      expected to redistribute the prefixes.
> 
> NEW:
>      [RFC9010] to enable a 6LR to inject the anycast and multicast
>      addresses in RPL.  Similarly, this specification extends [RFC8505]
>      and [RFC9010] to add the capability for a 6LN to register
>      unicast prefixes as opposed to addresses, and to signal in a
>      routing-protocol-independent fashion to a 6LR that it is
>      expected to redistribute the prefixes.
> 
> There are other instances in the document that I think should be fixed as 
> well.
> 
> Cheers,
> Med
> 
> 
> 
_______________________________________________
6lo mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to