> The harm from persisting the data is in terms of robustness, not in terms
of
> interoperability.  With multiple interacting clients, and allowance for
lowered
> error checking, things can go wrong.

And in terms of what I2RS is setting out to do -- we're not trying to
configure the box in a general sense. We're trying to provide an interface
to the "routing system," which doesn't mean the entire router (which seems
to be a lost point in much of the ongoing shuffle). The "routing system"
generally means the control plane, which manages "ephemeral state" in its
normal mode of operation (I know there are exceptions, but building rules
based on exceptions is always a recipe for bad rules). 

Hence, the original concept was to stay away from anything that smells like
a configuration. 

> The fallback in the case of I2RS is that reboot will drop all the state,
so no
> persistent harm can be caused.

Right -- which means if you mess with the control plane in such a way as to
make the box unreachable, reboot it, and things rebuild to a known state.

:-)

Russ

_______________________________________________
i2rs mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/i2rs

Reply via email to