Re: [Dime] Review of draft-ietf-dime-diameter-base-protocol-mib-00

[Mark Jones <Mark.Jones at bridgewatersystems.com>]

Hi Glen,

> -----Original Message-----
> From: Glen Zorn [mailto:gwz at net-zen.net]
> Sent: October 29, 2009 6:08 AM
> To: Mark Jones
> Cc: dime at ietf.org
> Subject: RE: Review of draft-ietf-dime-diameter-base-protocol-mib-00
> 
> Hi, Mark.  I've been reconstructing the latest (unpublished) version of
> th8is draft & going through the archives to make sure that the new rev
> will
> cover all the comments.  Anyway, at some point in the past you wrote "I
> have
> reviewed this draft and only have one comment: The dbpPeerVendorId
> should be
> Unsigned32 rather than enumerated INTEGER."  I see that I agreed to
> this
> change, but I'm wondering what the rationale might be...
> 

I assumed this was just an oversight as the other vendorId in the MIB are Unsigned32. 

I understand that the INTEGER type is recommended for representing named-number enumerations but it struck me as strange that the ID would define just four enumeration values for IETF, Cisco, 3GPP and Vodafone. The vendor name/id list is potentially very long and it didn't seem like the appropriate use of the enumerated INTEGER type. I'm not a MIB designer so this comment is largely based on how I've seen enums used in other MIBs. I now await the counter examples from the MIB experts ;)

Regards
Mark