Re: Proposed change to draft-ietf-6man-multi-homed-host after IESG review

[Bob Hinden <[e-mail]>]

Brian,

While I don’t have a strong objection to the proposed change, I am uncomfortable with what I read is a request to not even mention source-destation routing.  I agree it should not be a requirement in this document, but I don’t see anything wrong with a forward looking pointer.  Perhaps something like this (based on the NEW text):

  Network designs exist that can usefully limit themselves to static
  routing (such as a simple tree network) or may internally use no
  routers at all, such as a single LAN with two CE routers, each of
  which leads to a different upstream network.  However, the
  mechanism specified in this document may require some form of
  support from the routing protocols used in some multihomed networks.
  The details of such support are out of scope for this document.
  One approach that is being investigated to provide the requisite support
  in routing protocols is described in [I-D.ietf-rtgwg-dst-src-routing].

Or similar.

Bob


> On Sep 15, 2016, at 1:53 PM, Brian E Carpenter <[e-mail]> wrote:
> 
> Hi,
> 
> This message concerns
> https://tools.ietf.org/html/draft-ietf-6man-multi-homed-host-09 .
> 
> There is one remaining issue in the draft before it can get IESG approval.
> One of the routing ADs proposes the following change. It does not affect
> the normative content of the draft, but it is a technical change (deletion)
> so we'd like to hear if there are any objections from the WG. Please comment
> within a week (silence means no objection to this change). Some more
> background is below the proposed change.
> 
> OLD:
> 
> 2.2.  Expectations of multihomed networks
> 
>   The mechanism specified in this document requires some form of
>   support from the routing protocols used in multihomed networks.  One
>   such way of providing the requisite support in routing protocols is
>   described in a routing protocol independent fashion
>   [I-D.ietf-rtgwg-dst-src-routing].  Network designs exist that can
>   usefully limit themselves to static routing (such as a simple tree
>   network), or may internally use no routers at all, such as a single
>   LAN with two CE routers, each of which leads to a different upstream
>   network.
> 
> NEW:
> 
> 2.2.  Expectations of multihomed networks
> 
>   Network designs exist that can usefully limit themselves to static
>   routing (such as a simple tree network) or may internally use no
>   routers at all, such as a single LAN with two CE routers, each of
>   which leads to a different upstream network.  However, the
>   mechanism specified in this document may require some form of
>   support from the routing protocols used in some multihomed networks.
>   The details of such support are out of scope for this document.
> 
> You can see Alia's original DISCUSS comment dated 2016-08-02 at
> https://datatracker.ietf.org/doc/draft-ietf-6man-multi-homed-host/history/ .
> The "SHOULD" mentioned there has already been removed, but her considered
> opinion is that the routing solution is simply out of scope for this draft.
> 
> Regards
>   Brian Carpenter
> 
> 
> --------------------------------------------------------------------
> IETF IPv6 working group mailing list
> [e-mail]
> Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
> --------------------------------------------------------------------