Re: [manet] Mail regarding draft-dlep-lid

[Stan Ratliff <[e-mail]>]

On Mon, Jun 5, 2017 at 8:01 AM, Rick Taylor <[e-mail]>
wrote:

> Stan,
>
> I think you're over-complicating this.  Given a Data Item is a TLV,
> then as long as the lengths are consistent on a per-session basis, I'd
> be happy letting the modem implementer decide on the length.  As far as
> the router is concerned, it really doesn't matter.  Zero-padding fixed
> length fields seems a bit unwieldy...
>

I think some more is required. The consistent length is important, because
I can see a router implementation wanting to use the LID as a search key
in, for example, an AVL tree. Fixed-length keys are required there. And
let's consider a failing case - At Destination Up for the first "link", the
modem builds a TLV with length 4 and hands the router an integer. At
Destination Up for a second link, the modem builds a TLV with length 6, and
hands the router a MAC address...

What then? If you "kill" the Destination, then which one? Or, in other
words, which LID was "right" - the integer LID, or the MAC-based LID? Or,
should the entire Peer Session be brought down (which, as we've already
agreed, is an awfully big hammer)?

My thinking was (and still is) - if you're going to call for "consistent"
LIDs (length-wise), then you either have to explicitly state what that LID
size is (and possibly be prepared to negotiate the LID size), or allocate
sufficient "real-estate" to cover all the possibilities.

Regards,
Stan



>
> Rick
>
>
>
> On Sun, 2017-06-04 at 12:15 -0400, Stan Ratliff wrote:
> > Hi,
> >
> > So, what about the notion of allocating a 64-bit quantity for the
> > LID? That way, a MAC address would fit, and by definition, all LIDs
> > would be the same length. We can specify the characteristics of the
> > data in the LID (right-justify, fill character 0x00, etc). Or, we
> > could even use the first octet of that 64-bit quantity as a set of
> > bit flags to indicate what's in the LID (e.g., is it a MAC address,
> > or an integer, or...)
> >
> > Regards,
> > Stan
> >
> >
> > On Fri, Jun 2, 2017 at 3:58 AM, Morgan, Keith <[e-mail]>
> > wrote:
> > > Hi,
> > >
> > > We have to provide the MAC address in the current protocol so we
> > > think it makes sense to use it if Link ID's are accepted.
> > >
> > > I think all Link ID's should be the same length in a session, to
> > > keep it simple.
> > >
> > > Regards
> > >
> > > Keith
> > >
> > > Keith Morgan
> > > Director of Engineering
> > > COMMUNICATION SYSTEMS / HARRIS DEFENCE LIMITED
> > > Office: [phone] / Mobile: [phone]
> > > www.harris.com / [e-mail]
> > > Jays Close, Viables Estate, Basingstoke, Hampshire, RG22 4BA,
> > > United Kingdom
> > >
> > >
> > >
> > >
> > >
> > > Registered in England No. 2803090 Harris Defence Limited a UK
> > > Subsidiary of Harris Corporation, USA
> > > CONFIDENTIALITY NOTICE:  This email and any attachments may contain
> > > material that is "Harris Proprietary Information" for the sole use
> > > of the intended recipient.  Any review, reliance, distribution,
> > > disclosure, or forwarding without expressed permission is strictly
> > > prohibited.  If you are not the intended recipient, please contact
> > > the sender and delete all copies without reading, printing, or
> > > saving in any manner.  Thank You.
> > >
> > >
> > > -----Original Message-----
> > > From: Rick Taylor [mailto:[e-mail]]
> > > Sent: 31 May 2017 15:57
> > > To: Morgan, Keith (Non U.S.); [e-mail]
> > > Subject: Re: [manet] Mail regarding draft-dlep-lid
> > >
> > > Hi Keith,
> > >
> > > Comments inline...
> > >
> > > On Wed, 2017-05-31 at 13:30 +0000, Morgan, Keith wrote:
> > > > Hello,
> > > >
> > > > I have been reviewing the proposed Link Identifier Extension to
> > > DLEP
> > > > and have the following comments:
> > >
> > > Thanks for the quick review!
> > >
> > > >
> > > > The focus of the proposal appears to be to enable Routers to
> > > identify
> > > > where they cannot reach Destinations via the MAC address provided
> > > in
> > > > the Destination Up message. This would be the case with our
> > > Modem.
> > > >
> > > > Our current implementation does use “sleight-of-hand” and will
> > > report
> > > > both the MAC address and IP address in the Destination Up
> > > message,
> > > > however the Destination is only reachable via its IP address.
> > >
> > > This is one of the use-cases I was trying to fix, so I'm glad it
> > > helps.
> > >
> > > >
> > > > The concept proposed appears sound; my immediate suggestion would
> > > be
> > > > to recommend a length of 6 bytes for the identifier, this would
> > > allow
> > > > the Modem to use the MAC address of the destination which should
> > > > ensure uniqueness while allowing the router to determine how to
> > > > address packets to the Destination based on the use of the Link
> > > > Identifier Data Item.
> > >
> > > 4 is only RECOMMENDED.  As a variable length field, Link Id's can
> > > be any length, so if you want to use 6, that's fine.  I only picked
> > > 4 as I imagined people had integer values and felt I had to
> > > recommend something to point people away from using SHA1 hashes or
> > > something huge (although they still can).
> > >
> > > It's interesting that your Link Ids actually are MACs, but by using
> > > Link Ids not MAC Address Data Items you are being explicit that the
> > > MACs can't be used in frames.  I hadn't thought of this use-case,
> > > but I'm glad it works for you!
> > >
> > > One thing that is probably missing is a requirement that all Link
> > > Ids should be the same length during a session.  Do people think
> > > this is a valid restriction?
> > >
> > > Cheers,
> > >
> > > Rick
> > > _______________________________________________
> > > manet mailing list
> > > [e-mail]
> > > https://www.ietf.org/mailman/listinfo/manet
> > >