Re: AD review of draft-ietf-rap-feedback-fr-pib-04.txt

>>>>> Wijnen, Bert (Bert) writes:

Bert> Blush... indeed it says so.  Mmm... so why would a TC like
Bert> Usage32/64 be any better?

I guess Keith just wanted to be very clear that PIBs are not for
monitoring and just for configuration.

But to do real things, you need to get a feedback loop running and so
once you have COPS-PR running for configuration and control, you also
want to get monitoring^H^H^H^H^H^H^H^H^H usage data.

In Keith world, you would use SNMP and the DiffServ MIB to retrieve
monitoring data and that is why it is crucial that MIBs and PIBs are
very closely related.

/js

-- 
Juergen Schoenwaelder    <http://www.informatik.uni-osnabrueck.de/schoenw/>