Re: [multipathtcp] Fwd: New Version Notification for draft-bonaventure-mptcp-timestamp-00.txt

Hi Olivier,

Just had a couple questions to get a better understanding of how an 
implementation might behave.

Is the expectation that an MPTCP-TS option (either the sender-timestamp 
or echo) be included on outgoing packets whenever there is space for it? 
(and only withheld if there is say a larger MPTCP option or we require 
an additional SACK block etc)

At the end of Section 3 there is the following note:
"The MPTCP TS may also be used to verify that a subflow remains active 
by forcing a remote host to reply to an MPTCP segment without sending data."

I assume this is done by sending a MPTCP-TS option with E unset to 
elicit an echo? If this is the case, since replying with a timestamp 
echo is a "SHOULD" behaviour, how much can we rely on this to indicate a 
subflow's status?. Perhaps you could clarify in case I've misunderstood.

cheers,
nigel

On 27/10/14 19:28, Olivier Bonaventure wrote:
> FYI
>
> The main motivation for this draft is to initiate a discussion among
> MPTCP implementers on the usage of the RFC7323 timestamp option by
> Multipath TCP.
>
> Comments are welcome
>
> Olivier
>
>
> -------- Original Message --------
> Subject: New Version Notification for
> draft-bonaventure-mptcp-timestamp-00.txt
> Date: Sun, 26 Oct 2014 11:00:07 -0700
> From: <internet-drafts@ietf.org>
> To: Olivier Bonaventure <Olivier.Bonaventure@uclouvain.be>, "Olivier
> Bonaventure" <olivier.bonaventure@uclouvain.be>
>
>
> A new version of I-D, draft-bonaventure-mptcp-timestamp-00.txt
> has been successfully submitted by Olivier Bonaventure and posted to the
> IETF repository.
>
> Name:        draft-bonaventure-mptcp-timestamp
> Revision:    00
> Title:        Multipath TCP timestamp option
> Document date:    2014-10-29
> Group:        Individual Submission
> Pages:        6
> URL:
> http://www.ietf.org/internet-drafts/draft-bonaventure-mptcp-timestamp-00.txt
>
> Status: https://datatracker.ietf.org/doc/draft-bonaventure-mptcp-timestamp/
> Htmlized: http://tools.ietf.org/html/draft-bonaventure-mptcp-timestamp-00
>
>
> Abstract:
>     The TCP timestamps defined in [RFC1323] were designed to improve
>     round-trip-time estimations and provide protection against wrapped
>     sequence numbers (PAWS).  This draft clarifies the utilisation of
>     timestamps by Multipath TCP and proposes a new timestamp option that
>     better suits the needs of Multipath TCP.
>
>
>
>
>
> Please note that it may take a couple of minutes from the time of
> submission
> until the htmlized version and diff are available at tools.ietf.org.
>
> The IETF Secretariat
>
> .
>
>
>
> _______________________________________________
> multipathtcp mailing list
> multipathtcp@ietf.org
> https://www.ietf.org/mailman/listinfo/multipathtcp