Re: [DNSOP] A quick update on draft-ietf-dnsop-attrleaf / draft-ietf-dnsop-attrleaf-fix

[Adam Roach <[e-mail]>]

On 10/19/18 7:08 AM, Warren Kumari wrote:
> So, there were a few documents where I was not able to quickly figure 
> out which of the classes it should be placed in.
>

tl;dr: my analysis is that all four of the mentioned documents should be 
removed from the list. Details below.


> RFC3861 describes how to use SRV, but it is updated by RFC6121, which 
> largely says "Don't!" -- what do we do here? Update both? Just 
> RFC3861? Juast RFC6121?


That's sort of true. 3861 defines rules for _im._x and _pres._x (for 
arbitrary values of "x"), while 6121 says "don't do this for x = xmpp."  
For example, RFC 5509 builds on RFC 3861 for SIP, and it's technically 
still in effect (although the practical case is probably the same as 
XMPP, no one has produced a revision to 5509 that says as much).

All of that notwithstanding, 3861 most definitely does not deal with 
"Global" underscored node names as -attrleaf- defines that term. It 
deals only with labels further down the tree. I believe it should be 
removed from the list.


> I couldn't figure out RFC3404, and RFC6011.
> Clue appreciated.
>
> RFC3404 -- Dynamic Delegation Discovery System (DDDS) Part Four: The 
> Uniform Resource Identifiers (URI) Resolution Application
> Perhaps SRV? But it doesn't really seem to be underscore scoped...


This one is quite perplexing, and I don't know how it got caught up in 
Dave's net. The character "_" appears only 12 times in this document, 
all in variable names in the pseudo-code in its appendix. Neither of the 
English words "underscore" nor "underline" appear at all.

As far as I recall (and a quick skim seems to validate this), the only 
reservations DDDS makes are under "urn.arpa." and "uri.arpa." -- there 
aren't any global name reservations, with or without underscores.

I think this document was included in error.


> RFC6121 -- Extensible Messaging and Presence Protocol (XMPP): Instant 
> Messaging and Presence
> This updates RFC3921 and explicitly recommends against SRV.
> "Interoperability Note: RFC 3921 specified how to use the _im._xmpp 
> and _pres._xmpp SRV records [IMP-SRV] as a fallback method for 
> discovering whether a remote instant messaging and presence service 
> communicates via XMPP. Because those SRV records have not been widely 
> deployed, this document no longer specifies their use, and new 
> implementations are not encouraged."
> Should this be in this list?


RFC 6121, as you point out, only mentions _{im,pres}.xmpp as an 
historical note while deprecating its use. I believe it should also be 
removed.


>
> RFC3861 -- Extensible Messaging and Presence Protocol (XMPP): Instant 
> Messaging and Presence
> See above. It would be SRV, but was updated by RFC6121.


[see above]


>
> RFC6011 -- Session Initiation Protocol (SIP) User Agent Configuration
> I got confused here -- I cannot really see the underscore names here 
> as anything other than a target name.


6011 shouldn't be updated by this document. The only underscored node 
names in that document are in non-normative examples, and they're 
actually incidental to the purpose of the example. Presumably, they're 
included to demonstrate that querying for NAPTR records can return both 
UA Config records *and* other, unrelated records, and that 
implementations need to ignore the unrelated records.


Thanks for doing the heavy lifting on this. I think it has dramatically 
improved the document to figure out why each updated RFC is included, 
since it flushed out several that should not be.


/a