RE: comments

["Budd, Fred" <Fred.Budd@wcg.com>]

Where I was really heading was expanding the proposed requirement
to include the explicit definition of local (e.g. directly attached subnet) and remote, and corresponding functional behavior for
inline and remote access. This could be done either in the
requirement, the examples, or by adding a new informational
section. There have been quite a few explanations of the assumed or
implicit operational context surrounding the initial requirement.
That context really provides the logical framework that a consumer
(and vendor) needs to evaluate the importance of the requirement. 

-Fred Budd

> Robert Drake (rdrake@sprint.net)
> > George Jones(gmj@pobox.com) 
> > There are basicly two options: OoB management and non OoB mgt.
> > OoB management can be serial port, dedicated ethernet, whatever.
> > Inband can be whatever it is.
>
> It might be useful to differentiate between local and remote
> management as well as OOB and inband.  
> 
> > If OoB then initial config has to be done via OoB port.
> > If not OoB then initial config has to "inline".

> Here I would say if no OoB then initial config has to be local
> (and inline, but thats implied)
 
> > A decision tree needs to be worked up and the profiles ajusted
> > accordingly.