FW: Comments on draft-ietf-ops-rfc2851-update-05.txt

["Dave Thaler" <dthaler@windows.microsoft.com>]

[correcting the email alias]

-----Original Message-----
From: Dave Thaler 
Sent: Monday, November 05, 2001 5:34 PM
To: 'mibs@ietf.org'; 'iesg@ietf.org'
Subject: Comments on draft-ietf-ops-rfc2851-update-05.txt

Background:
> From: Juergen Schoenwaelder [mailto:schoenw@ibr.cs.tu-bs.de]
> Sent: Monday, November 05, 2001 3:11 PM
>
> Dave> Situation: a table INDEX'ed by an InetAddressType object and an
> Dave> InetAddress object that are not in the table itself.
> 
> Dave> My reading of the section 4 is that if an InetAddressType object
> Dave> occurs in the INDEX, but that object does not have an OID in the
> Dave> table itself, then the table MUST still have another
> Dave> InetAddressType column before any InetAddress columns in the
> Dave> table.  (And presumably its DESCRIPTION would typically say the
> Dave> value MUST be equal to the value of the InetAddressType object
> Dave> in the INDEX.)
> 
> Dave> Is this the intent?
> 
> Not really. For me, the INDEX components are kind of "logically
> registered" before any other columns of a table. The big question is
> how to describe this in an understandable way.
> 
> Since we are in IETF last call mode, I would like to ask you to bring
> this issue up on the <mibs@ietf.org> and the <iesg@ietf.org> or
> <ietf@ietf.org> mailing lists.

The problematic text is (from Sect 4, 3rd para):
> The InetAddressType object must be registered before the InetAddress
> object(s) or InetAddressPrefixLength object(s).  In other words, the
> object identifiers for the InetAddressType object and the InetAddress
> object MUST have the same length and the last sub-identifier of the
> InetAddressType object MUST be less than the last sub-identifier of
> the InetAddress object.

Here's some suggested replacement text:
The InetAddressType object must be registered before the InetAddress
object(s) or InetAddressPrefixLength object(s).  In other words, either:
a) the object identifiers for the InetAddressType object and the
InetAddress
object MUST have the same length and the last sub-identifier of the
InetAddressType object MUST be less than the last sub-identifier of
the InetAddress object, or
b) the InetAddress object MUST be a columnar object in a table with an
InetAddressType object in the INDEX (preceding the InetAddress object if
it is also in the INDEX).
Furthermore, when an InetAddress object appears in the INDEX of another
table, it MUST be preceded in the INDEX of that table by an
InetAddressType object.


(I believe the above text covers all the cases where an INDEX can
contain a Type from the same or a different table, and perhaps an
Address from the same or a different table.)

-Dave