Hi,
> While it was Jeff Case who suggested adding 'power' as one of the values
> of chasEntityFunction, I think Dan's "would allow" phrasing is what is
> intended, i.e., if the power supply is on a card in a regular slot, then
> it's important to reflect that in the chasEntityTable/chasSlotTable;
> if it's not on a card in a slot, then it need not be in the
> chasEntityTable/chasSlotTable.

Exactly!

> Since you raise the issue of Parties, IP Addresses and Community Strings,
> with the progression of SNMP Security RFCs, and the evolution of
> S[N]MP, does anybody think we need to retain Community Strings and
> IP Addresses in the chasEntityTable ??

YES!
I think that one of the biggest problems of SMP is the dependence on 
implementation of parties, even if on most applications security is
not needed and no benefit is visible. Some subset of the Party concept
not implying all the security stuff would have been enough (Sorry, this
is not the S[N]MP discussion !!).
My bet is that we will run for quite a long period in paralel S[N]MP
(to get some of the many advantages it brings) and 'classic' SNMP for
simple applications and back compatibility.
Not retainig Community Strings and IP Addresses in the chasEntityTable
would mean that all existing agents already running in boxes should be
upgraded to support parties. It is not always possible and for sure 
not desirable.

dan@lannet.com

Dan Romascanu
LANNET Data Communications Ltd.
Tel Aviv, Israel
Voice: 972-3-6485414
Fax:   972-3-5447146