Re: [ippm] Jim Guichard's Discuss on draft-ietf-ippm-explicit-flow-measurements-03: (with DISCUSS)

[Giuseppe Fioccola <[e-mail]>]

Hi Jim, Tommy,
To avoid misunderstanding, I would suggest to revise section 6 without saying explicitly which particular bits can be used in QUIC or TCP. Figures 16, 17, 18, 19, 20 can simply be omitted. We can only highlight that experiments have been made with QUIC and TCP.

Regards,

Giuseppe

From: ippm <[e-mail]> On Behalf Of James Guichard
Sent: Wednesday, May 3, 2023 5:04 PM
To: Tommy Pauly <[e-mail]>; Martin Duke <[e-mail]>
Cc: Giuseppe Fioccola <giuseppe.fioccola=[e-mail]>; The IESG <[e-mail]>; [e-mail]; [e-mail]; [e-mail]
Subject: Re: [ippm] Jim Guichard's Discuss on draft-ietf-ippm-explicit-flow-measurements-03: (with DISCUSS)

Hi Tommy,

Thanks for your email. Yes, understood, and as long as that is clear I have no issues. As I said in an earlier email my main concern was to make sure that folks with expertise in this area were made aware of the document so that any impact could be assessed. While the document isn’t using the reserved bits it does specifically point to those bits (and TCP header bits) implying that using them is ‘all well and dandy’. So my concern was an implementer doing exactly that and breaking something that I am not qualified to assess. Perhaps simply saying, as suggest by Tim Chown in another email, that the reserved bits could be used (without specifically saying which particular ones) might be a safer option (?).

Thanks!

Jim

From: Tommy Pauly <[e-mail]<mailto:[e-mail]>>
Sent: Wednesday, May 3, 2023 10:37 AM
To: Martin Duke <[e-mail]<mailto:[e-mail]>>; James Guichard <[e-mail]<mailto:[e-mail]>>
Cc: Giuseppe Fioccola <giuseppe.fioccola=[e-mail]<mailto:giuseppe.fioccola=[e-mail]>>; The IESG <[e-mail]<mailto:[e-mail]>>; [e-mail]<mailto:[e-mail]>; [e-mail]<mailto:[e-mail]>; [e-mail]<mailto:[e-mail]>
Subject: Re: [ippm] Jim Guichard's Discuss on draft-ietf-ippm-explicit-flow-measurements-03: (with DISCUSS)

Indeed, this document is not defining any standard, or allocating any code points in protocols like QUIC or TCP. It defines approaches that can be used in the future by these and other protocols.

Jim, to your original DISCUSS, this document *isn’t* using the reserved bits in QUIC, but instead is discussing how they *could* be used. I imagine the right fix to the text is to make that very explicit so that there isn’t room for confusion.

Thanks,
Tommy

On May 3, 2023, at 7:28 AM, Martin Duke <[e-mail]<mailto:[e-mail]>> wrote:

I do not support elevating this document to PS. Whatever normative bits future standards want to use will have to be repeated in that document, which is great because trying to implement this entire draft would be very confusing.

On Wed, May 3, 2023 at 2:07 AM Giuseppe Fioccola <giuseppe.fioccola=[e-mail]<mailto:[e-mail]>> wrote:
Hi Jim,
Thank you for your review.
This is an informational document since we initially followed the approach of RFC 8321 and RFC 8889, which were elevated to standards track when the methods were considered more mature (i.e. RFC 9341 and RFC 9342). The bits used in both QUIC and TCP headers are just examples of experimentation. Indeed, this draft aims to define transport agnostic methodologies that can be theoretically applicable to any transport-layer protocols between client and server. Different techniques are included in this document, while the assumption is that the different standards track documents, which apply these methods to a specific protocol (e.g. QUIC, TCP, CoAP,…), will consolidate the design into fewer bits according to the application scenarios. A first example is https://datatracker.ietf.org/doc/draft-ietf-core-coap-pm/.
I agree with you that a standards track document would avoid any possible objection about the DOWNREF of future documents, but I do not know if there is the consensus of the WG to change the intended status to standards track. Maybe this could be verified with a poll. Moreover, it is also worth highlighting that some of the methods described in this draft are already proposed standards: Q bit is described in RFC 9341 while S bit is introduced as an optional feature in RFC 9000.

Regards,

Giuseppe

-----Original Message-----
From: Jim Guichard via Datatracker <[e-mail]<mailto:[e-mail]>>
Sent: Monday, May 1, 2023 4:20 PM
To: The IESG <[e-mail]<mailto:[e-mail]>>
Cc: [e-mail]<mailto:[e-mail]>; [e-mail]<mailto:[e-mail]>; [e-mail]<mailto:[e-mail]>; [e-mail]<mailto:[e-mail]>; [e-mail]<mailto:[e-mail]>
Subject: Jim Guichard's Discuss on draft-ietf-ippm-explicit-flow-measurements-03: (with DISCUSS)

Jim Guichard has entered the following ballot position for
draft-ietf-ippm-explicit-flow-measurements-03: Discuss

When responding, please keep the subject line intact and reply to all email addresses included in the To and CC lines. (Feel free to cut this introductory paragraph, however.)


Please refer to https://www.ietf.org/about/groups/iesg/statements/handling-ballot-positions/
for more information about how to handle DISCUSS and COMMENT positions.


The document, along with other ballot positions, can be found here:
https://datatracker.ietf.org/doc/draft-ietf-ippm-explicit-flow-measurements/



----------------------------------------------------------------------
DISCUSS:
----------------------------------------------------------------------

I am wondering why this is an informational document when it uses reserved bits
from both QUIC and TCP headers (?). If those reserved bits are used by the
mechanisms described in this document but there is no "official" allocation of
the bits then future documents that wish to use these bits will be limited
and/or clash with an Informational RFC. Adding a DISCUSS as although this is
not a technical area of expertise for me, it seems unusual and I would like to
better understand the document track selection.

I also do not see a transport area directorate review and in fact the document
shepherd highlights that the document could benefit from such a review. Given
that the bits introduced in the document are suggested to be carried in the
QUIC and TCP headers using their reserved bits, then a review by the area
responsible for those transport protocols seems mandatory.






_______________________________________________
ippm mailing list
[e-mail]<mailto:[e-mail]>
https://www.ietf.org/mailman/listinfo/ippm