Minutes IETF104: uta
minutes-104-uta-00
| Meeting Minutes | Using TLS in Applications (uta) WG | |
|---|---|---|
| Title | Minutes IETF104: uta | |
| State | Active | |
| Other versions | plain text | |
| Last updated | 2019-04-01 |
minutes-104-uta-00
Agenda for UTA at IETF-104
Hotel Hilton Prague, Prague, Czech Republic
Tuesday, March 26, 2019
10:00 - 11:00, Karlin 3 Room
Chairs: Leif Johansson
Valery Smyslov
5 min. Administrativia (jabber scribes, note takers, blue sheets),
Note Well, Agenda bashing 5 min. WG documents status update 20 min.
draft-ietf-uta-tls-for-email-01 (Loganaden Velvindron) 20 min.
Discussion around IESG evaluation of
draft-ietf-uta-smtp-require-tls-07 10 min. Open Mic
draft-ietf-uta-tls-for-email-01 - Loganaden report:
- Leif made the point that once a draft becomes a working group document
that change control passes to the group - Alexey Melnikov asked as an
implementor that he'd like guidance on the cipher suites used. - Is the
focus on Client to Server or Server to Server. - Alexey: The document is 5
pages at the moment and ideally it should cover both Client to Server and
Server to Server but if only one topic is to be included it should be
Client to Server as the first preference.
Discussion around IESG evaluation
- Jim Fenton update on draft-ietf-uta-smtp-require-tls-07
- Two (2) DISCUSS positions
- Changes slated for -08
- RequireTLS and REQUIRETLS were different but not always picked up
- Header field to change from RequireTLS to TLS-Required. Will still have a
mandatory value of "No". - Many more instances of messages being re-originated
(beyond mailing lists) such as Vacation, SIEVE and may not be aware of
REQUIRETLS - Barry Leiba: Bounce messages may include the original message and
not requiring REQUIRETLS being propogated. - Alexey and Barry committed to
providing text regarding propogation of REQUIRETLS to SIEVE. - Jonathan
Hoyland conveyed comments from the Jabber stream - Ned Freed: SIEVE can be
executed from just before delivery until long after. - Security Consideration
- Barry: Error in 8.4 Policy Conflicts - "on the part of the sender not to use
TLS" should be "not REQUIRE the use of TLS". - Ben Kaduk: Not sure this
directive will be honoured. - Watson Ladd: "TLS-Required: No" is unintuitive.
- Joseph Salowey: The wording is unclear on the decision to not use TLS. -
Jim: The wording error has been pointed out by Barry. - Joe: The last sentence
is problematic. - Viktor Dukhovni: TLS-Required: No is not the same and
require no TLS. The bulk of SMTP is transmitted via TLS as shown by Google +
START-TLS everywhere. - Barry: use of "use" in that text is problematic. -
Ben: Not confused in the way identified by Viktor. This is a sender
preference. Do we honor the sender preference vs honor the receiver
preference. - Barry: Should look at picking the better of the two options
between sender/receiver preference. MTA-STS does require TLS in strict. -
Viktor: The word strict in MTA-STS and its interpretation is incorrect. It
only applies for every hop using MTA-STS. - Pete Resnick: The audience
understands the MTA-STS requirements. There isn't a conflict between
"TLS-Required: No" and the use of TLS isn't in conflict and maybe the wording
should be "TLS-Optional: Yes".
- Leif: Will this text be wordsmithed on the mailing list.
- Jim: Yes will be sent to the list for discussion prior to being in the -08
update. - Viktor: SMTP is infact a UDP datagram sent to a recipient. The
sender understands the context behind the options they set. - Leif: Thanked
Viktor for his work on the mailing list making people understand this.
https://tools.ietf.org/html/draft-tschofenig-uta-tls13-profile-01 - Hannes:
- Submitted to UTA because _DICE_/_ICE_ (missed that) doesn't exist anymore.
- Leif: Are you looking for this to be a working group document?
- Hannes: Yes. DTLS 1.3 is wrapping up.
- Leif: With only 1 reader of this document and the focus of this group to
be email at the moment it needs more reviews to work on this. - Barry:
Agreed to be a review volunteer. - Alexey: It is up to the chairs on
whether they should adopt this document. It is fine charter wise. - Leif:
It is good to know that the working group can complete. - Jonathan (Jabber
Scribe) that there is another reviewer in Loganaden.
Open Mic:
- No comments.
Meeting closed at 10:50