Re: [tsvwg] draft-ietf-tsvwg-rfc6040update-shim:SuggestedFragmentation/Reassemblytext

[Bob Briscoe <[e-mail]>]

David,

I'll reply to your response on the shim text here, and start a new 
thread for the encap draft, to keep this from growing too long...

On 22/03/2021 22:00, Black, David wrote:
> First, the good news - the reassembly text in Section 5 of rfc6040update-shim
> draft looks good:
>
>     Section 5.3 of [RFC3168] defines the process that a tunnel egress
>     follows to reassemble sets of outer fragments
>     [I-D.ietf-intarea-tunnels] into packets.
>
>     During reassembly of outer fragments [I-D.ietf-intarea-tunnels], if
>     the ECN fields of the outer headers being reassembled into a single
>     packet consist of a mixture of Not-ECT and other ECN codepoints, the
>     packet MUST be discarded.
>
>     If there is mix of ECT(0) and ECT(1) fragments, then the reassembled
>     packet MUST be set to either ECT(0) or ECT(1).  In this case,
>     reassembly SHOULD take into account that the RFC series has so far
>     ensured that ECT(0) and ECT(1) can either be considered equivalent,
>     or they can provide 2 levels of congestion severity, where the
>     ranking of severity from highest to lowest is CE, ECT(1), ECT(0)
>     [RFC6040].
>
> I would slightly rephrase the first paragraph as follows:
>
>     Section 5.3 of [RFC3168] specifies ECN requirements for tunnel egress
>     reassembly of sets of outer fragments [I-D.ietf-intarea-tunnels] into IP packets
>     when at least one of the fragments is CE-marked.  If none of the fragments
>     in a set to be reassembled is CE-marked, then the following two additional
>     requirements apply:

[BB] I'm afraid that's incorrect. This requirements from RFC3168 covers 
cases other than those with at least one CE fragment:

    "Reassembly of
    a fragmented packet MUST NOT change the ECN codepoint when all of the
    fragments carry the same codepoint."


To achieve what you want, I believe i just need to replace the first 
para with the following, then bullet paras 2 & 3:

    Section 5.3 of [RFC3168] specifies ECN requirements for reassembly of
    sets of outer fragments [I-D.ietf-intarea-tunnels] into packets.  The
    following two additional requirements apply at a tunnel egress:


Also, we agreed that we would not update RFC3168 here. This section 
starts by saying it updates RFC6040, so I've kept it to just that (my 
draft-13 text was wrong in this respect too). I've moved the words 
"...at a tunnel egress" so they confine the scope of this update to a 
tunnel egress, rather than incorrectly redefining the scope of RFC3168 
(which is about reassembly anywhere, not just "...at a tunnel egress:")

>
> and then itemize the second and third paragraphs (e.g., number them 1. and 2.).
>
> The rationale for this rephrasing is that Section 5.3 of RFC 3168 does not specify
> a process, but rather imposes requirements on what is allowed.  I also believe
> that this text the bulk of what's needed to address the reassembly comment
> that's been holding up these drafts.
>
> ---------------------------------
See new thread for the rest about ecn-encap...


Bob

-- 
________________________________________________________________
Bob Briscoe                               http://bobbriscoe.net/