Re: [Isms] Last Call: rfc5953 (Transport Layer Security (TLS) Transport Model for the Simple Network Management Protocol (SNMP)) to Draft Standard

[Sean Turner <[e-mail]>]

On 5/1/11 12:41 AM, SM wrote:
> At 14:14 19-04-2011, The IESG wrote:
>> The IESG has received a request from the isms WG (isms) to consider the
>> following document:
>>
>> - 'Transport Layer Security (TLS) Transport Model for the Simple Network
>> Management Protocol (SNMP) '
>> RFC 5953 as a Draft Standard
>>
>> The IESG plans to make a decision in the next few weeks, and solicits
>> final comments on this action. Please send substantive comments to the
>> [e-mail] mailing lists by 2011-05-03. Exceptionally,
>> comments may be sent to [e-mail] instead. In either case, please
>> retain the beginning of the Subject line to allow automated sorting.
>>
>> This specification contains eight normative references to standards
>> track documents of lower maturity: RFCs 1033, 3490, 3584, 4347, 4366,
>> 5246, 5280, and 5952.
>
> In Section 7:
>
> "A hostname is always in US-ASCII (as per [RFC1033]);
> internationalized hostnames are encoded in US-ASCII as domain
> names after transformation via the ToASCII operation specified
> in [RFC3490]."
>
> As a quick comment, RFC 1033 is a down-ref. A better reference is RFC
> 1123. As it is part of STD 3, a down-ref is no longer needed. The
> reference to RFC 3490 could be updated to RFC 5890. That also avoids a
> down-ref in a Draft Standard to a document that has an Obsolete status.

As SM noted, the reference to RFC 3490 needs to be updated.  In fact, I 
know the Apps ADs will place a DISCUSS on this particular point based on 
some exchanges about a DKIM draft on a similar topic.  The change is 
slightly more complicated than replacing the references because ToASCII 
went bye-bye in RFC 5890.  After some exchanges with an Apps AD, the 
following change would "work" (this very similar to the changes proposed 
by DKIM to their draft) - and "work" means won't force a recycle at DS:

OLD:

   A hostname is always in US-ASCII (as per [RFC1033]);
   internationalized hostnames are encoded in US-ASCII as domain
   names after transformation via the ToASCII operation specified
   in [RFC3490].  The ToASCII operation MUST be performed with the
   UseSTD3ASCIIRules flag set.  The hostname is followed by a
   colon ':' (US-ASCII character 0x3A) and a decimal port number
   in US-ASCII.  The name SHOULD be fully qualified whenever
   possible.

NEW:

   A hostname is always in US-ASCII (as per [RFC1033]);
   internationalized hostnames are encoded as A-labels as specified
   in [RFC5890].  The hostname is followed by a
   colon ':' (US-ASCII character 0x3A) and a decimal port number
   in US-ASCII.  The name SHOULD be fully qualified whenever
   possible.

FYI I also asked about setting the UseSTD3ASCIIRules flag and Pete said 
it was completely unnecessary as far as he could tell.  That's why I 
just dropped it.

Does anybody object to this change?  (It's perfectly okay to object to 
these words.  We'll just need to work some other text out with Pete/Peter).

What do others feel about the reference change from RFC 1033 to 1123. 
Are there any technical reasons to not make this change?

As a result of the change for the 3490 reference (and maybe 1033 
references), either a new draft is needed or I need to enter an RFC 
editor note.*  I'm willing to enter these changes as RFC editor notes. 
  Just let me know which path you'd like to follow.

spt

*  The new draft or RFC editor note will need to include the editorial 
errata filed against RFC 5953.