dnsop-wg-materials/interim-2022-dnsop-02/interim-2022-dnsop-02-minutes.txt

# DNS Operations (DNSOP) Working Group
## interim-2022-dnsop-02


### Chairs
* Benno Overeinder [benno@nlnetlabs.nl](benno@nlnetlabs.nl)
* Suzanne Woolf [suzworldwide@gmail.com](suzworldwide@gmail.com)
* Tim Wicinski [tjw.ietf@gmail.com](tjw.ietf@gmail.com)

### IESG Overlord
* Warren Kumari [warren@kumari.net](warren@kumari.net)

### Document Status
* [Github](https://github.com/ietf-wg-dnsop/wg-materials/blob/main/dnsop-document-status.md)
* [Datatracker](https://datatracker.ietf.org/wg/dnsop/documents/)
* [Propose Slides](https://datatracker.ietf.org/meeting/interim-2022-dnsop-02/session/dnsop)


## Session interim-2022-dnsop-02

* Date: 26 September 2022
* Time: 15:00-16:00 UTC
* MeetEcho: [https://meetings.conf.meetecho.com/interim/?short=40f2f302-13a7-477b-b6fd-3d3d0754f05f](https://meetings.conf.meetecho.com/interim/?short=40f2f302-13a7-477b-b6fd-3d3d0754f05f)

* Jabber:  [dnsop@jabber.ietf.org](dnsop@jabber.ietf.org)
* Minutes: [https://notes.ietf.org/notes-ietf-interim-2022-dnsop-02-dnsop](https://notes.ietf.org/notes-ietf-interim-2022-dnsop-02-dnsop)


## Agenda

### Administrivia

* Agenda Bashing, Blue Sheets, etc,  5 min


### Current Working Group Business

*   DNS Terminology 
    - https://datatracker.ietf.org/doc/draft-ietf-dnsop-rfc8499bis/
    - Paul Hoffman and Kazunori Fujiwara, 55 min
    - Chairs Action:

Paul Hoffman(PH): Are we interept original document; or new definitions based on current DNS usage


### Definition of Bailiwick

* Number of proposals
* Definition of in-bailiwick
  - in-domain
  - sibling domain
  - out-of-bailiwick
 
Warren Kumari(WK): should not fully drop a definition, but note no longer using. 

PH: What should be saying on how its defined. 

John Levine(JH): don't try to define now

PH: Has been used in the past, but we don't really know anymore

Jim Reid(JR): drop the term in-baliwick we must mention not to define it

* Drop the term “in-bailiwick”
  - Only use “in-domain” and “sibling domain”?
 
**Action** Pull the formal definition and write a historical definition

* Scope of use of bailiwick
  - Strict use with A/AAAA: “needed for DNS resolution”
  - Other RR types for DNSSEC, SVCB, …

PH: in-baliwick consensus call Old vs Current; Current means future

Kazunori Fujiwara: I think the terms *bailiwick, in-domain, sibling are necessary for glue is not optional draft.
(in-domain glue is necessary, sibling glue is optional, out-of-bailiwick glue SHOULD be ignored)

### Definition of Glue

* Definition of glue (on DNSOP mailing list, plus some small amendments)
  - “Glue is non-authoritative data in a zone that is transmitted in the additional section of a referral response on the basis that the data might be necessary for resolution to proceed at the referred name servers.”

* Necessary vs. useful discussion
  - too broad definition?
  - use “glue for in-domain/sibling domain name servers” as a term (from draft glue-is-not-optional)

PH: definition of glue has changed over time

Duane Wessels: This definition does not say which RRTypes are used, we may need a registry
Narrow definition for now. 

List of RRTypes

Defintion use necessary or useful

**Action** wording on in-domain

### Definition of Sibling Glue

* Definition of sibling zone and glue 

* Sibling zones: two zones whose delegations are in the same parent zone.

* Sibling glue: addresses of nameservers that are in a sibling zone.

* Necessary vs. useful interpretation
  - Same as for in-domain glue?

sibling zone
sibling glue


```
Jim Reid 11:01
Is anyone speaking or is my audio bust?

Eliot Lear 11:02
I can hear you

Sean Turner 11:02
I can hear.

Warren Kumari 11:02
I can hear you

Suzanne Woolf 11:02
Benno sounds fine

Tim Wicinski 11:02
https://notes.ietf.org/notes-ietf-interim-2022-dnsop-02-dnsop?edit

Jim Reid 11:02
Thanks Benno. I hear you OK.

Tim Wicinski 11:02
I'll be taking notes today
But primary today is putting together wording on baliwick

Kazunori Fujiwara 11:07
draft-ietf-dnsop-glue-is-not-optional-06 uses "in-domain" and "sibling", However, it does not define these words and does not refer RFC8499...

Tim Wicinski 11:08
Correct - what should come out of this discussion is terminology that the glue-is-not-optional draft will use.
My opinion is that one goal of the terminology doc was to create current definitions
but we want to hear from y'all
I copied/pasted the 8499 text at the end of the HedgeDoc https://notes.ietf.org/notes-ietf-interim-2022-dnsop-02-dnsop?edit

Kazunori Fujiwara 11:15
"in-bailiwick" is often used in non-IETF documents.

Warren Kumari 11:15
"current definitions" + a note that the definition has changed over time so that people reading historic documents are not confused? Or jsut "this is what it means, be told" ?

Wes Hardaker 11:17
it exists in way too many docs and logs to not define it

PE 11:17
so you mean definition as more histrorical context than current best use of term?

Warren Kumari 11:17
@PE : Yah, kinda. "I jsut read a document with this term, but had no idea what it means" help...

Tim Wicinski 11:20
broken up on my end

Warren Kumari 11:20
Not just at Benos emd. Audio dropped out for me too

John Levine 11:20
Yes, and don not try to define it now

Roy Arends 11:21
audio dropped here too

John Levine 11:21
Sorry bad ipad audio
Only as a historical use

Suzanne Woolf 11:22
IMO it's okay to admit the term is widely used but ambiguous.

Paul Hoffman 11:24
I like "historical term"

Duane Wessels 11:24
agreed

Tim Wicinski 11:25
yes

Kazunori Fujiwara 11:26
For "in-domain" and "sibling", there was no clear definition of terminology before RFC 7719, but I thought it was necessary, so I borrowed terminology from Peter Koch's draft.

Jim Reid 11:27
@ Wes, I think it's impractical to come up with definitions for how that term as been used in all those non-IETF docs. IMO a "we chose not do that" is the right way forward.

Tim Wicinski 11:29
"Pull the formal definition of baliwick and write a historical definition"

Kazunori Fujiwara 11:29
I think the terms *bailiwick, in-domain, sibling are necessary for glue is not optional draft.

Tim Wicinski 11:29
I do agree

Duane Wessels 11:31
no audio, I'll use chat
is this slide really about use of glue?
rather than use of in-bailiwick?

Warren Kumari 11:32
I don't really care -- for my use case, I'm assuming people can solve their issue with: https://www.google.com/search?q=in+ballwick

Duane Wessels 11:32
ok, thanks
yes I think that would be best
I'd rather not define the terms in the glue is not optional doc
agreed

Tim Wicinski 11:35
strict definition or more ambigious historical?

Duane Wessels 11:36
sorry still no audio for me

Kazunori Fujiwara 11:39
(in-domain glue is necessary, sibling glue is optional, out-of-bailiwick glue is unnecessary)

Tim Wicinski 11:39
Thank You Kazunori !

Kazunori Fujiwara 11:40
(out-of-bailiwick glue SHOULD be ignored, sorry)

Tim Wicinski 11:41
I'm happy of not updating 2181

John Levine 11:42
Agree with Paul, most useful to describe fuzzy existing practice
then I hope say this is the preferred meaning

Kazunori Fujiwara 11:43
RFC 2181 5.4.1 Ranking Data seems to target older nameserver implementations that merge all the data.

Tim Wicinski 11:48
My feeling is in the future something like SVCB may become glue, we're not there yet

Kazunori Fujiwara 11:52
Is the EXCHANGE A/AAAA that comes with the MX RR glue ?

John Levine 11:52
Not as I've ever understood it

Duane Wessels 11:53
@kazunori I'd say no because those could be done as a separate query

John Levine 11:54
we had a big fight about sibling glue
since it only works sometimes

Kazunori Fujiwara 11:56
in-domain, sibling, out-of-bailiwick are exclusive.
in-bailiwick = in-domain + sibling

Warren Kumari 11:58
I dont.

Paul Hoffman 11:58
Please: no

Peter Thomassen 11:58
I'll have to leave

Warren Kumari 11:58
I have another meeting now

Eliot Lear 11:58
nah

Paul Hoffman 11:59
It's not a short conversation

Eliot Lear 11:59
^^^

Sean Turner 11:59
until next time

Eliot Lear 11:59
thanks chairs

Kazunori Fujiwara12:00
Thank you.

Suzanne Woolf12:00
Thanks all, this was a good session!