Re: Where do we go from here?

[Rob Austein <sra+xmlconf@hactrn.net>]

At Mon, 12 Aug 2002 09:50:07 -0500, Keith Allen wrote:
> 
> One of the issues we keep talking about here at SBC every time the subject
> of operator requirements comes up is the difference in requirements among
> operators.  One example is scale.  As mainly an access provider, we deal
> with tens of thousands of network elements, whereas a backbone provider
> might have only hundreds of very high throughput boxes.  Two orders of
> magnitude can have a significant impact on the way you manage your network.

Good point.

> The post I am responding to points out another difference.  We want our
> boxes to never reboot.  (Sadly, this is still a goal, not reality.)  Still,
> if the xmlconf group focuses on defining config files to be loaded on
> reboot, we will hope to never make use of it, and will still need a good
> management protocol.

Feel free to sustitute "whack all configurable parameters of an
element into a known and consistant state" for "reboot" and "complete
settings for all the configurable parameters" for "config file" if
that works better for the kind of equipment you deal with.  I do
understand that there may be kinds of equipment for which the config
file model is totally wrong, I just don't want to get hung up on the
baggage associated with words like "reboot" and "file".

I do also understand the goal of having the boxes never need to
reboot, but I suspect that there's a corollary of the laws of
thermodynamics that says you'll never get there.  Past a certain point
I'd guess that one simulates getting there by having enough redundancy
in the system that one can whack individual elements without anybody
outside the NOC needing to care that one has done so.  Right?

--
to unsubscribe send a message to xmlconf-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/xmlconf/>