Yes, I think that your diagram is a good start. I hope your ASCII art
fu is better than mine :-)

For the "functional analysis" of I2RS discussion, I think that a
diagram or two would be a good idea as well - to make it clear what
aspects are interesting, in-scope, and generally likely to be safe.
Do you have time to draw something a bit more detailed up?

Of course, it's not going to fully describe all the aspects - but
things like mucking with data learned via routing (as opposed to
policy in processing that data or in specifying data "owned" by the
network element) is pretty clearly hazardous.

Thanks,
Alia

On Mon, Mar 4, 2013 at 5:42 PM, Russ White <[email protected]> wrote:
>
>> The diagram is a high-level conceptual picture mostly showing
>> connections from i2rs
>> to other aspects of the box.  The RIB Manager is, I think, the exception.
>
> Of course --my point was that to drive the data model, and better
> understand the interaction between I2RS and the devices we want to
> interact with, we need something that's more specific.
>
> Different purposes, but both are needed, IMHO.
>
> :-)
>
> Russ
>
>
_______________________________________________
i2rs mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/i2rs

Reply via email to