Re: Progressing: draft-farrel-rtg-manageability-requirements-01.txt

In message <00dd01c5ecf0$045f0750$fb849ed9@Puppy>
"Adrian Farrel" writes:
>  
> Attention ADs.
>  
> I wouldn't claim overwhelming support, but I would also note that the
> only issues raised so far were by you at the meeting in Vancouver.
>  
> So where do we go from here?
>  
> We originally positioned this as Informational, but making firm
> requirements.
>  
> My suggestion, in response to your concerns in Vancouver was to keep
> this as Informational, but drop the requirements making the
> Manageability Concerns section advisory.
>  
> The small email exchange on the list, however, is requesting BCP
> status with firm requirements.
>  
> Do you want to give some guidance and enter into a discussion?
>  
> Cheers,
> Adrian

Adrian,

It makes no sense to try to mandate something in an informational
draft.

If you mandated this structure most routing protocols would have to
cite the same link liveness and host liveness protocols and how they
interact plus the mechanisms of the protocol itself.

If there is a set of routing protocol for which management is not
widely enough understood, having a separate informational draft would
be good.  Otherwise to be complete this would be a very big section.

For example, in addition to MIB you missed XML.  Do you cite the DTD
and a sample XLST?  Is there such a things as a common XML DTD across
vendors (I think not)?  If a routing protocol was written prior to the
use of XML for management would that preclude using XML or require a
new iteration of each routing protocol?

I think it is a mistake to put in a "how to run your network" tutorial
in each routing protocol.  So far operations practices have changed
independently of iterations of the routing protocol itself.  Even
implememtation changes such as coupling with link liveness and
neighbor liveness detection occur independently of the routing
protocol itself.  Coupling the routing protocol with management of the
routing protocol by putting it in the same document would be a
mistake.

Curtis

_______________________________________________
routing-discussion mailing list
routing-discussion@ietf.org
https://www1.ietf.org/mailman/listinfo/routing-discussion