Re: [manet] I-D Action: draft-ietf-manet-dlep-multi-hop-extension-01.txt

[Rick Taylor <[e-mail]>]

Comments inline...

On Wed, 2017-11-08 at 08:36 -0500, Lou Berger wrote:
> Henning,
> 
> Thank you for the comments! see below for in-line response.
> 
> On 11/8/2017 2:21 AM, Henning Rogge wrote:
> > Hi,
> > 
> > First, why do you need the "reserved" bytes at the end of the
> > TLVs...
> > DLEP already contains TLVs with a length not dividable by 4.
> 
> I'm fine with byte alignment, but alos thought it would be nice to
> leave
> room for more flags definition without requiring a new data item. 
> Bits
> are cheap in the control plane.  That said, I'm okay with
> reducing/eliminating them if that's WG consensus.
> 
> > I am a bit concerned about the "hop control" data item, it packs A
> > LOT
> > more complexity than the "hop count" one.
> 
> Agreed.  Note that the modem controls if a router can do any control
> via
> the Hop Count Data Item P bit.
> 
> > I also wonder if "hop control" should be part of a "request link
> > characteristic" message.
> > 
> > Thoughts about it?
> 
> I'm very interested in such a message, but also like small and
> focused
> function increments from both standards and implementation
> perspectives.  My inclination is to move forward with this one and
> start
> thinking and working on the details of such a request message/DI. 
> What
> do you think?

Lou, the Link Characteristics Request message already exists in
RFC8175, and I think Hop Control data items should probably live there
instead.

Rick