State of play with Shim6 documents

[Geoff Huston <gih@apnic.net>]

Working from the minutes of the SHIM6 WG meeting in November, and looking
at the mailing list traffic since then it appears that we are about as
close to submitting the HBA and Failure Detection shim6 documents to the
IESG as we'll ever be, but that the Protocol specification document
requires a little more discussion on a number of topics (noted below) and
definitely another rev of the draft.

Here's the state, as I understand it, relative to the comments raised in
the Last Call and in the IETF 67the Working Group meeting

  1. HBA (draft-ietf-shim6-hba-02.txt )

    -- IPR issues - we appear to have reached a level of consensus with
       acceptance of this document in the light of the Ericsson and
       Microsoft IPR statements

    -- IANA considerations - no further issues

    -- Security Considerations - no further issues

  2. Shim6 Protocol (draft-ietf-shim6-proto-07.txt)

    -- IPSEC - text added to the draft as noted in the WG meeting

    -- Provide SHIM 6 security based on IPSec SAs - no further action

    -- ULID Security Mechanism - no further issues

    -- DOS attacks on the context tag set - no consensus for further change

    -- Context Forking - no further issues

    -- CGA Key Length - text added to the draft as noted in the WG meeting

    -- Broken Flag - no further issues

  3. Failure Detection (draft-ietf-shim6-failure-detection-07.txt)

    -- Behaviour Section - text added per comments


Since then we've had a review of proto-7 by Iljitsch van Beijnum  (11
December http://ops.ietf.org/lists/shim6/msg01742.html), as well as a
number of other suggestions regarding thos document  The major issues
raised have been were:

  * context tag

    This review revisited the topic of inclusion of the context tag in the
    shim6 header.
    Followup postings to this were at (
    http://ops.ietf.org/lists/shim6/msg01772.html, and
    http://ops.ietf.org/lists/shim6/msg01777.html)

    As far as I can tell there is no apparent WG consensus to alter the
    existing text at this point in time,

  * congestion control

    It also noted issues regarding TCP congestion control.

    A TCPM approach (http://ops.ietf.org/lists/shim6/msg01788.html) is
    described in the document
    http://tools.ietf.org/wg/tcpm/draft-schuetz-tcpm-tcp-rlci-00.txt

    There appears to agreement to reference this draft in the proto
    document in order to address this issue.

  * explicit shim6 error message (the lack thereof)

    The review proposed a shim6 error message
    (http://ops.ietf.org/lists/shim6/msg01769.html)

    One posting in support of this
    (http://ops.ietf.org/lists/shim6/msg01775.html)

    **** Its not clear what the WG want to do on this. Suggestions? ****

  * TCP Checksum Failure

    The review noted that any intermediary system that performed a TCP
    checkum would fail.

    Proposal to add a middlebox consideration section to point out the
    problem and an idea to use some form of probe mechanism to detect paths
    with this problem

    Alternative proposal for the SHIM to alter the TCP and UDP checksums to
    match the selected locators on outgoing and match the ULIDs on
    incoming.

    **** Its not clear what the WG want to do on this. Suggestions? ****


  * Control Message length

    Text proposed (http://ops.ietf.org/lists/shim6/msg01780.html)

    Are we in agreement with this suggestion?

  * Sundry

    The review also listed a number of grammar errors, typos and
    suggestions relating to terminology

    These appear, on the whole, to have been accepted by the document
    editor (http://ops.ietf.org/lists/shim6/msg01774.html)

 *  Changes to the shim message exchange

    Proposed in http://ops.ietf.org/lists/shim6/msg01756.html

    This suggestion appears to have been accepted by the document editor

    Also http://ops.ietf.org/lists/shim6/msg01758.html


  * Receiving Receiving payload without extension headers

    This suggestion appears to have been accepted by the document editor


So we are almost ready with these documents, and as far as I can tell there
is a rough consensus for the HBA and failure detection document, but a
clear need to resolve the outstanding issues with the proto document and
incorporate the changes as noted above.

The three topics that I see from the mailing list as being unresolved at this stage
are:

    * explicit shim6 error message (the lack thereof)

          Its not clear what the WG want to do on this. Suggestions?

    * TCP Checksum Failure

          Its not clear what the WG want to do on this. Suggestions?

    * Control Message length

          Text proposed (http://ops.ietf.org/lists/shim6/msg01780.html)
           Are we in agreement with this suggestion?

thanks,

  Geoff & Kurtis