>>>>> C M Heard writes:

>> The problem is that mgmt applications either understand the
>> semantics of a particular implementation or they will have to guess
>> or ignore some rows.

Mike> I undestand that a management application may not have full
Mike> understanding of route table entries without understanding the
Mike> nature of the opaque identifier.  But I don't follow why that
Mike> implies that a management application has to ignore some rows --
Mike> at least if all it's doing is inspecting rather than modifying
Mike> the rows.  The proposed index components are destination prefix,
Mike> opaque identifier, and next hop.  I can still determine how many
Mike> routes to a given destination via a particular next hop exist
Mike> (and query their other properties) even if I don't know what the
Mike> opaque identifier is.

Well, it depends what you expect from a management application. If
your manager is a MIB brower which just shows the data to a human,
things are fine. If however the management application want to reason
about the data (e.g. which route will a packet with destination A
take), then a totally opaque value basically means that the
application can not answer this question.

/js

-- 
Juergen Schoenwaelder    <http://www.informatik.uni-osnabrueck.de/schoenw/>
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page:                      http://playground.sun.com/ipng
FTP archive:                      ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------

Reply via email to