The Routing Over Large Clouds Mailing List Archive by date

[yakov@watson.ibm.com]

Ref:  Your note of Wed, 08 Mar 1995 12:18:05 -0800


Bruce,

>My point is that there are simpler networks, for which NHRP can live
>in loop freedom.

And my point is that NHRP doesn't have any mechanism(s) to distinguish
such "simpler networks" from other cases. So, I want to understand
what mechanism(s) should be used to constrain NHRP application to
only "simpler networks". There is also a need to define the exact meaning
of "simpler networks" -- the adjective "simpler" seems to be rather vague.

>My other point was that I don't hink we can conclude whether or not
>NHRP is useful for more than just address resulution until we decide
>what kinds of NBMA networks are really to be deployed.

I don't think it is within the scope of ROLC WG to "decide what kinds of
NBMA networks... to be deployed". So I am confused by your statement about
"we decide...".

>Enforced by routing. As long as the station has a misassigned IP address,
>it is unreachable, as routing directs IP traffic for the CIDR block
>to the NBMA network.

The routing may not help (an example can be provided on request).

>Now in Joel's example, one could limit the NHRP speaking routers to only
>use NHRP to resolve IP destinations within their respective administrative
>control. Either through ACLs, or by building multiple logical NBMA networks
>on top of the physical NBMA network. I believe this would solve (or at least
>help solve) the example looping problem.

Certainly more complexity can be added to NHRP (e.g. by adding ACL).

Certainly more complexity can be added to the deployment requirements
(e.g. building multiple logical NBMA networks on top of the physical
NBMA network).

I've yet to see how all this would solve the looping problem.

Yakov.