[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Methods in the NIM requirements

>>>>> Andrea Westerinen writes:

Andrea> I know that Juergen said that he didn't want to discuss this -
Andrea> but why bring up an example and then say "don't discuss this"?
Andrea> The DMTF teams (both Device and Networks) really did give some
Andrea> thought to the Speed property in CIM_NetworkAdapter.  Here is
Andrea> the background:

Andrea> 1.  "Speed" was modeled as a uint64 since it was discussed
Andrea> that maximum growth range should be allowed and that units
Andrea> should stay "bps" for consistency with the IETF MIB.  2.  If
Andrea> the Speed value actually fits in a uint32, then the mapping is
Andrea> clear.  If the Speed is greater than a uint32 with units of
Andrea> bps, then reality can't be reflected in RFC1213's attribute.
Andrea> 3.  The goal was to map both the MIB and DMI, where DMI
Andrea> defined the property with units of Mbps.

So the question which may be important in this context is what an
expression like

>          MappingStrings {"MIB.IETF|RFC1213-MIB.ifSpeed",
>                "MIF.DMTF|Network Adapter 802 Port|001.5"}

really means if both referenced definitions are not really the same as
the CIM definition of Speed. (BTW, IF-MIB (right now RFC 2233) defines
ifHighSpeed which is again different from CIM's definition of Speed
(mostly due to SMIv2 shortcomings).

Once again, it was not my goal to discuss whether the CIM definition
of Speed makes sense or not. What I was trying to point out is that
mappings even in this simple case turn out to be not that simple.
(And they may actually mislead implementors who do not carefully study
the subtle differences to do the wrong thing.)