Re: [netext] Review of I-D: draft-ietf-netext-pd-pmip-01

Hi Basavaraj,

I have a question:

Can we at Netext work on specs that require modifying MN?

Reading the charter:

The work in this charter is entirely internal to the network and does
  not affect host IP stack operation in any way (except perhaps through
  impacting packet forwarding capacity visible to the hosts).

It seems to me the answer is no?

Regards,

Behcet
On Mon, Nov 7, 2011 at 2:17 PM, <Basavaraj.Patil@nokia.com> wrote:

>
> A few quick comments:
>
> 1. Abstract should have no references. And the description can be
>   improved. It really is hard to understand what this I-D is trying
>   to accomplish.
>
> 2. The introduction section can be improved. Maybe you should include
>   a problem statement section to indicate what is lacking in PMIP6 as
>   per RFC5213.
>
> 3. In Sec 3.1 it states:
> "
>   o  The MR (as a RR) MUST either obtain the Home Network Prefix (HNP)
>      before initiating the DHCPv6-PD procedure or in case of stateful
>      address configuration simultaneously while configuring the Mobile
>      Node Home Address (MN-HoA).
> "
>
>   So does this imply that the MR (PMIP6 MN) obtains an HNP as per
>   RFC5213 for its own use prior to the MR requesting via DHCPv6
>   another set of prefixes to be used for delegating to downstream
>   nodes?
>
> 4. If the prefixes (delegated) are provided by a DHCP server (and not
>   the LMA), how does the LMA get informed about these? In the PBU?
>   How do you ensure that
>   "All the mobile network prefixes managed in the DR MUST be
>   reachable via local mobility anchor (LMA)" when they are not
>   assigned by the LMA?
>
> 5. Sec 3.1 also states:
>   "The delegating router (DR) can be located either at LMA or some
>      other device in the PMIPv6 domain"
>
>    What is this some other device?
>
> 6.  In Sec 3.2 the I-D states that the profile needs to indicate
>    whether an MN can be assigned prefixes for delegation. Does the
>    profile indicate whether the host is a mobile router? Is this
>    needed? Any MN should be able to act as an MR.
>
> 7. In Fig 1, you have a DR shown there. How does the DR indicate the
>   assigned prefixes to the LMA?
>   - Step 5 seems disconnected from the rest of the process. Maybe it
>   is better to split that aspect and show it in another figure.
>
> 8. Does the lifetime of the delegated prefix (assigned by a DR) the
>   same as the lifetime assigned by the LMA?
>   Why not have Sec 3.3.3 to describe how a prefix is revoked or
>   deleted? Text is currently in 3.3.2.
>
> 9. In sec 3.4.2:
>   "Other considerations from Section 6.10.5 or [RFC5213]"
>
>   Sec 6.10.5 of this I-D? There is none.
>
> 10. Sec 3.5.2 states:
>    "In order to receive
>      those packets, the LMA MUST advertise a connected route into the
>      routing infrastructure for the MR's MNP(s)."
>
>    - Is it enough from a specification standpoint to state that the
>      routes need to be injected into the routing infra?
>      Handwaving????
>
> 11. Are there any new threats or security issues that arise from
>    assigning prefixes to be delegated downstream in the case of
>    PMIP6? Security section seems pretty light.
>
>
> -Raj
>
> _______________________________________________
> netext mailing list
> netext@ietf.org
> https://www.ietf.org/mailman/listinfo/netext
>