Re: [tcpm] Summary of updates in draft-ietf-tcpm-accurate-ecn-17

[Bob Briscoe <[e-mail]>]

Yoshi,

On 20/03/2022 09:50, Yoshifumi Nishida wrote:
> Hi Bob,
>
>   * 'that acknowledges new data' -> 'if any byte counter has
>     incremented since the last ACK' {*includes new markings on
>     retransmssions*}
>
> Does this mean when you retransmit a packet, the acc option in the 
> packet should be updated to contain the latest info?
> If we do this, wouldn't it be hard for the receivers to tell which 
> packet contains the most recent info when there's reordering?

[BB] Yes and No (I don't think so, but perhaps I misunderstand your 
question).

For first question: see end of §3.2, which has been there since the 
start of AccECN:

    Whenever a host feeds back the value of any counter, it MUST report
    the most recent value, no matter whether it is in a pure ACK, an ACK
    with new payload data or a retransmission.  Therefore the feedback
    carried on a retransmitted packet is unlikely to be the same as the
    feedback on the original packet.

For second question, when you say "the receiver"
* if you mean the Data Receiver (i.e. the sender of the feedback), it 
doesn't need to know. It just increments its counters in whatever order 
markings arrive, whether on retransmissions or not, and feeds back the 
latest counter.
* if you mean the Data Sender, see A.1 or A.2.1 where the example 
algorithms at the Data Sender check whether a packet is superseded 
(based on ackno, sack & timestamps if available) before using the feedback.

Does that answer your questions?


Bob

> --
> Yoshi
>
> On Mon, Mar 7, 2022 at 3:26 PM Bob Briscoe <[e-mail]> wrote:
>
>     tcpm-ers,
>
>     The AccECN draft has just been updated to roll up all the changes
>     over the last month since draft-16 (3 Feb 22).
>     https://datatracker.ietf.org/doc/html/draft-ietf-tcpm-accurate-ecn
>     All the changes have already been discussed on this list, but they
>     are enumerated below for convenience. The last normative change
>     below about ACK filtering has been discussed before, but the
>     precise wording hasn't been posted to the list until now.
>
>     As always, the full diff can be accessed from the datatracker (via
>     the above link).
>
>     Normative (Changes intended to make a technical difference are *in
>     bold*)
>
>       * §3.1.1
>         <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-17.html#section-3.1.1>:
>         Negotiation during the TCP handshake
>         Reordered the requirements for feeding back the incoming
>         IP-ECN field on the SYN-ACK to make more sense {no technical
>         change, just clearer}
>       * §3.2
>         <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-17.html#section-3.2>:
>         AccECN Feedback
>         made "MUST increment byte counters" conditional on support for
>         sending of the AccECN TCP Option
>       * §3.2.2.3
>         <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-17.html#section-3.2.2.3>:
>         Testing for Mangling of the IP/ECN Field
>         If continuous CE-marking is detected, Data Sende:
>           o SHOULD NOT send ECN-capable packets and *then SHOULD NOT
>             respond to any ECN feedback* {added second part of this
>             requirement}
>           o *Throughout, it MUST remain in the AccECN feedback mode*
>             {additional requirement}
>           o and it MUST continue to feed back any ECN markings on
>             arriving packets (in its role as Data Receiver) {just
>             reworded a bit}
>       * §3.2.2.5.1
>         <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-17.html#section-3.2.2.5.1>:
>         Data Receiver Safety Procedures
>         Increment-Triggered ACKs:
>           o Clarified that the condition about 'new data' means 'newly
>             delivered data' {no difference in meaning, just clearer}
>           o Changed to 'In either case, *'n' MUST be no greater than
>             7*.' (was 6)
>       * §3.2.3.3
>         <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-17.html#section-3.2.3.3>:
>         Usage of the AccECN TCP Option
>         Condition for 'SHOULD include an AccECN TCP Option on every
>         scheduled ACK':
>           o 'that acknowledges new data' -> 'if any byte counter has
>             incremented since the last ACK' {*includes new markings on
>             retransmssions*}
>       * §3.3.3
>         <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-17.html#section-3.3.3>:
>         Requirements for TCP ACK Filtering
>           o 'SHOULD preserve the correct operation of AccECN feedback'
>             {*previously it described how this SHOULD be done, now it
>             just gives the high level requirement*}
>
>
>     Technical
>
>       * §3.2.1
>         <https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-17.html#section-3.2.1>:
>         Initialization of Feedback Counters
>         Made initialization of ECT(1) feedback counter r.e1b the same
>         as r.e0b, now that we have the two different TCP Option orders.
>
>
>     Editorial
>
>       * Made other parts of the spec consistent with the above changes
>       * Called out the updates to other RFCs in the abstract
>       * Increased depth of table of contents to 4 (was 3)
>       * Added titles to all tables (xml2rfc now adds a 'Table xx'
>         caption to the HTML rendering whether or not there is a
>         caption title, and even if suppress-title=true i enabled).
>       * Other minor edits
>
>
>
>     Bob
>
>     On 07/03/2022 23:17, [e-mail] wrote:
>>     A New Internet-Draft is available from the on-line Internet-Drafts directories.
>>     This draft is a work item of the TCP Maintenance and Minor Extensions WG of the IETF.
>>
>>              Title           : More Accurate ECN Feedback in TCP
>>              Authors         : Bob Briscoe
>>                                Mirja Kühlewind
>>                                Richard Scheffenegger
>>     	Filename        : draft-ietf-tcpm-accurate-ecn-17.txt
>>     	Pages           : 60
>>     	Date            : 2022-03-07
>>
>>     Abstract:
>>         Explicit Congestion Notification (ECN) is a mechanism where network
>>         nodes can mark IP packets instead of dropping them to indicate
>>         incipient congestion to the end-points.  Receivers with an ECN-
>>         capable transport protocol feed back this information to the sender.
>>         ECN was originally specified for TCP in such a way that only one
>>         feedback signal can be transmitted per Round-Trip Time (RTT).  Recent
>>         new TCP mechanisms like Congestion Exposure (ConEx), Data Center TCP
>>         (DCTCP) or Low Latency Low Loss Scalable Throughput (L4S) need more
>>         accurate ECN feedback information whenever more than one marking is
>>         received in one RTT.  This document updates the original ECN
>>         specification to specify a scheme to provide more than one feedback
>>         signal per RTT in the TCP header.  Given TCP header space is scarce,
>>         it allocates a reserved header bit previously assigned to the ECN-
>>         Nonce.  It also overloads the two existing ECN flags in the TCP
>>         header.  The resulting extra space is exploited to feed back the IP-
>>         ECN field received during the 3-way handshake as well.  Supplementary
>>         feedback information can optionally be provided in a new TCP option,
>>         which is never used on the TCP SYN.  The document also specifies the
>>         treatment of this updated TCP wire protocol by middleboxes, updating
>>         BCP 69 with respect to ACK filtering.
>>
>>
>>     The IETF datatracker status page for this draft is:
>>     https://datatracker.ietf.org/doc/draft-ietf-tcpm-accurate-ecn/
>>
>>     There is also an HTML version available at:
>>     https://www.ietf.org/archive/id/draft-ietf-tcpm-accurate-ecn-17.html
>>
>>     A diff from the previous version is available at:
>>     https://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-accurate-ecn-17
>>
>>
>>     Internet-Drafts are also available by rsync at rsync.ietf.org::internet-drafts
>>
>>
>>     _______________________________________________
>>     tcpm mailing list
>>     [e-mail]
>>     https://www.ietf.org/mailman/listinfo/tcpm
>
>     -- 
>     ________________________________________________________________
>     Bob Briscoehttp://bobbriscoe.net/
>
>     _______________________________________________
>     tcpm mailing list
>     [e-mail]
>     https://www.ietf.org/mailman/listinfo/tcpm
>

-- 
________________________________________________________________
Bob Briscoehttp://bobbriscoe.net/