ietf-openproxy
[Top] [All Lists]

Re: Draft minutes from IETF 58

2003-11-17 07:36:07

Hi,

since there were no further comments, these minutes are final now and have been forwarded to IETF secretariat.

-Markus


Markus Hofmann wrote:

Folks,

attached the *draft* minutes from our meeting at IETF 58. Thanks to Eric for taking the minutes, and thanks to Marshall for taking care of the Jabber.

In case you've comments, please provide them by Friday, 11/14, 5pm (EDT), so that we can forward the minutes to the IETF in a timely manner.

Thanks,
  Markus


------------------------------------------------------------------------

Open Pluggable Edge Services (OPES)

58th IETF Meeting

11 November 2003

Eric Burger, scribe

Jabber log is at

<http://www.xmpp.org/ietf-logs/opes(_at_)ietf(_dot_)xmpp(_dot_)org/2003-11-11.html>

=====

Agenda Bashing: Markus Hofmann

=====

Document Status - See slides & ID Tracker

Goal: finish documents in  "ID Exists" status by end of November

=====

OPES Treatment of IAB Considerations

draft-ietf-opes-iab-03 - Abbie Barbir

See slides

On Considerations 3.1 & 3.2: chose to make notifications optional; focus on 
tracing.

Clearly, OPES cannot do end-to-end encryption.  OPES can do hop-by-hop 
encryption.

Sally Floyd: Likes document; addresses IAB issues

Markus: expect WGLC soon after IETF

=====

OPES process and end points communications

Draft-ietf-opes-end-comm-05 - Abbie Barbir

See slides

Tracing

-------

Open issue: What is OPES Tracing? Current wording has problem with "inclusion". 
 Does that mean that trace can be adapted (pruned, modified, added-to)?

Should we allow trace adaptation?

Trace information designed for system administrators, not users.

Bypass

------

IAB Considerations said to bypass non-OPES content.  But, what is OPES content? 
 Solution: punt -- definition of OPES content is outside scope of OPES.

Sally Floyd: Comment: IAB document isn't legal language. Try to figure out what the intent is. The 
IAB document does not mean "user has to have a way to bypass firewalls, boundaries, etc." 
 It is saying, "If a user asks for content from web server, and the content gets mangled by an 
OPES server, then user needs to be able to figure out what went wrong."

MUST/SHOULD/MAY questions: see slides

w.r.t. OPES bypass - Sally Floyd: Source of IAB concern about non-OPES is data 
integrity, e.g., how does one detect malicious transformation of data.

=====

OPC: OPES Callout Protocol

draft-ietf-opes-ocp-core-03 - Abbie Barbir

see slides

Should authentication be in the core or binding (HTTP)?

Should authentication be in this draft, or a separate draft?

Same issues for Encryption.

Should we get IETF review before going to WGLC?  Marshall: "NO"

Application message identifiers are irrelevant for HTTP, but might not be 
robust enough for SMTP.  Should we dump them and figure out the right thing for 
SMTP later, if there is a later?

Markus: Please post comments on AM-ID's to the list!

=====

HTTP Adaptation with OPES

Draft-ietf-opes-http-01 (from Martin and Alex) presented by Markus

See slides

Targeting WGLC for end of November

Need HTTP expert to review HTTP-specific issues, and in particular on 
HTTP-specific security considerations

Open issues: How to avoid blocking; skip uninteresting request bodies in 
'response mode'; and handle HTTP-specific things like 100 Continue and 206 
Partial Content?

Please check for XXX's in document and send text to mail list!

=====

P: Message Processing Language

Presented by Markus

See slides

Open issues:

What information is available to interpreter? Complete message? Headers only? 
Punt?

Is doing HTTP module in scope for work group?   If so, in what document?

Should work group define interfaces between P interpreters and module suppliers 
or callout services?  If so, how do services return results, e.g., from OCP?

Ted Hardie: Working group should specify what information is available to the 
interpreter.  Not a value judgement of where, just we should set expectations 
for community.

Ted Hardie: When do we expect the current charter items to be done?

Markus: Hopefully December.

=====

OPES Future

Markus

Note: Re-chartering is ONLY for adding new work items, not for staying alive to 
do work from old charter.  Keep new work in OPES framework, not something new.

Really need to know who will be committed to working on new work items.

PLEASE POST THOUGHTS AND SUGGESTIONS TO MAIL LIST.  PLEASE VOLUNTEER TO DO 
WORK!!!

OPES for SMTP and IMAP: does it conflict with lemonade?

Eric Burger: Not at all.  Definitely fits into OPES, just need to coordinate 
groups.

If you have interest in a topic, want to be an editor or contributor, please 
post to list.

Marshall: Importance of finishing documents -- "credibility is the coin of the 
realm."


<Prev in Thread] Current Thread [Next in Thread>