The MPLS WG Archive

[Paul Langille <langille@crescentnets.com>]

    Hi (again!) Tom,


"Thomas D. Nadeau" wrote:

>          Hi Paul,
>
> >     Hi Tom. I agree with you (and Mani) that you need the triplet <
> > Tunnel index, Tunnel Instance and Source address> to uniquely identify a
> > tunnel. Good catch.
>
>          Thanks.
>
> >     So now I am wondering, do you need the tunnel cookie (since it is the
> > source address and the tunnel index)?
>
>          I believe that we are going to eliminate the cookies from the MIB.
> There was some discussion about this a while back.

Thanks. I don't recall the discussion. I will check the archives.

>
>
> >Could you also give some brief examples as to how this would work with
> >bi-directional tunnels. (I may be a bit confused by the
> >mplsTunnelDirection object.)
>
>          Do you still need the example now that you know that the
> cookie is deleted, or were you referring to the addition of the
> srcAddr to the index?

It is not clear to me how to represent a bi-directional tunnel with the MIB.
Some of my confusion may come from the fact that I don't understand the
mplsTunnelDirection object. For example, what does it mean to have this object
set to "bi-directional"? Does it mean that this row refers to a bi-directional
tunnel? Does the TunnelDirection (in, out) have any significance at the tunnel
midpoint? What role does the cookie (or the lack of one!) play in
bi-directional tunnels?

So rather than answer these questions one at a time, it might be faster to
catch them all with an example. I assume that you have a solution in mind but
you currently don't have the cycles to put it in the MIB. Have you implemented
bi-directional tunnels?

        Thanks Paul

>
>
>          --Tom

--
Paul Langille                e-mail: langille@crescentnets.com
Crescent Networks            phone: (978) 244-9002 x244
201 Riverneck Road           fax: (978) 244-9211
Chelmsford, MA 01824