Re: Mib questions

[[e-mail] (Keith McCloghrie)]

Dan,

>                                          I would like to answer these
> questions, not in order to give an authoritive answer but rather to
> present how we understand what has been proposed untill now and how we 
> try to implement it. 

Good idea.  I agree with most of your answers.  Here's where we differ:

> > 1) What is the scope of the chassis MIB?  Is it limited
> >    the the devices covered in the Overview (section 5)?
> Yes, I think we should limit the scope to 'networking devices'
> as defined in the Overview. It is very tempting to extend the
> scope ('The sky is the limit') but in order to be effective we
> should limit now the scope and maybe later, when we will have
> some implementation experience try to solve other problems of
> mankind.
 
If extending the scope were to require more or different MIB objects,
then I'd agree, but if someone wants to apply the same MIB to a more 
general case, then I don't see any need to prohibit it.  I think this
can be handled by appropriate text to this effect in the Overview.

> > 2) Do the terms "functional module" and "logical device"
> >    mean the same and are they things like routers, bridges,
> >    terminal servers, etc?
> Yes, I think so.
 
As Bob said, a "module" is a physical entity in the Chassis, whereas
a logical device is something which has an agent.  These can be the
same but are not necesarily.

> > A fundamentation question is just how is an agent that implements
> > the chassis MIB suppose to get all the information about the other
> > "logical devices" - 1) through SNMP requests to the other "logical
> > devices", 2) through a private network protocol, 3) through a backplane
> > bus using shared memory or messages between CPUs,  or 4) through some
> > "static" configuration file that is loaded at agent boot time (and
> > maybe refreshed during running)?
>
> I really do not think that the Chassis MIB standard must address this 
> question which is a matter of implementation. Anyway, in my opinion:
> 1) does not work in the SNMP framework, as all "logical devices" are
> agents.
> 2) could work
> 3) the best
> 4) could work, but it is the least efficient.

I agree with all but your answer to 1), in that some (but not all) of
the information could be obtained through SNMP.  (Not that I recommend
that.)

Keith.