Re: [v6ops] WGLC: draft-ietf-v6ops-unique-ipv6-prefix-per-host-02 - multiple prefixes per device

[Alexandre Petrescu <[e-mail]>]

Hi,

Touchscreen and other WiFi, DSL and Cable Boxes of today rarely have
only one network interface.

As such, whereas it can make sense to dedicate an entire prefix to one
interface of such device (instead of sharing a prefix on the direct
link), the following also apply:
- should dedicate multiple such prefixes per device (not just one).
- in case only one such prefix is dedicated, it should not be /64 but a
   different prefix length.

This draft does neither.  It is called "unique prefix per Host"; and it
allocates a /64: "subscriber gets its own unique /64 IPv6 prefix".

If the text needs some example prefix len, please use /63 instead, not /64.

Maybe instead of "unique prefix per Host", would be better to say
"prefixes dedicated to a Host".

If the text needs to take a stance with respect to DHCPv6 PD, then
please make it a MUST.  The following is too ambiguos and delays PD
deployment: "if/when DHCPv6 prefix delegation support is desired".
Because it is not an if, and it is now.  This is really not good.

If this is not taken into account, I am afraid it will add to the pile
of RFCs mandating to allocate a /64 to an end host, use 64share, and
thus make it impossible to extend the network at the edge.

(once it agrees it should rather talk DHCPv6-PD.  Then we can discuss
again.)

(this draft could benefit from experience of earlier RFCs that also
dedicate a single prefix to an - albeit mobile - node - Proxy Mobile IP
RFC5213, Prefix Delegation for PMIP RFC 7248.)

Alex

Le 14/03/2017 à 15:40, Ron Bonica a écrit :
> Folks,
>
> This message initiates a Working Group Last Call (WGLC) for
> draft-ietf-v6ops-unique-ipv6-prefix-per-host-02. Since IETF is
> approaching, we will let the last call run a little longer than
> usual.
>
> Please submit comments to the list by April 11, 2017.
>
> Lee, Fred and Ron
>
> _______________________________________________ v6ops mailing list
> [e-mail] https://www.ietf.org/mailman/listinfo/v6ops
>