SORT BY:

LIST ORDER
THREAD
AUTHOR
SUBJECT


SEARCH

IPS HOME


    [Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

    Re: iSCSI: draft vs. 14 typos, suggestion, questions




    Thanks - comments in text - Julo


    Tomáš Bartušek <tomasb@s3group.cz>

    07/03/2002 02:38 PM
    Please respond to Tomáš Bartušek

           
            To:        Julian Satran/Haifa/IBM@IBMIL
            cc:        ips@ece.cmu.edu, ivan.pavelka@s3group.com
            Subject:        iSCSI: draft vs. 14 typos, suggestion, questions

           


       Hello Julian!

    We'd like to inform you about some TYPO errors, some
    SUGGESTIONS to clarify something / explain something
    better and also to ask few QUESTIONS (and sorry for so long
    email).

    *** Section 4.3 : TYPO & SUGGESTION
    In draft vs. 14 possibility to negotiate some parameters
    (e.g. TargetAddress) more than once.

    Vs. 14 says:
       Neither the initiator nor the target should attempt to declare or
       negotiate a parameter more than once during login except for
       responses to specific keys that explicitly allow repeated key decla-
       rations (e.g. TargetAddress). If detected by the target this MUST
       result in a Login reject (initiator error). The initiator MUST drop
       the connection
    TYPO: Missing fullstop at the end of a paragraph.

    Vs. 13 said:
       Neither the initiator nor the target should attempt to negotiate a
       parameter more than once during login. If detected by the target this
       MUST result in a Login reject (initiator error). The initiator MUST
       drop the connection.

    SUGGESTION: However the change resulted in badly-readable paragraph,
    because the reader can be confused and think, that sentence "If
    detected ... MUST result in Login reject" is applicable to parameters,
    for which "repeated key declaration IS allowed". I would suggest to
    write "attempt to re-negotiated other parameter" instead of "this" in
    that sentence.
    +++


    How about the text:

    Neither the initiator nor the target should attempt to declare or negotiate a parameter more than once during login except for responses to specific keys that explicitly allow repeated key decla-rations (e.g. TargetAddress). If an attempt to re-negotiate/re-declare parameters  not speciffically allowed is detected by the tar-get the target MUST respond with Login reject (initiator error); if detected by the initiator the initiator MUST drop the connection.
     

    +++
    *** Section 6.5  : TYPO & SUGGESTION
    TYPO: In paragraph explaining initator behaviour
    bad ALIGN (possibilities a) and b) not aligned verticaly!
    SUGGESTION: We would recommend to move explanation of
    target's SNACK processing into the chapter about SNACK. Here
    it could be/is confusing - We would say, that target's reaction
    to recieved SNACK is not dependent on whether initiator sends
    this SNACK in response to corrupted-digest data PDU or from
    (any) other reason. Or is it? */
    +++I do not see the missalignment and I do not understand the comment

    The behaviour on SNACK is specific to this case and it outlines the things target has to do depending on the target state
    +++
    *** Section 6.11 : TYPO
    (probably) TYPO: What does "initiators detect protocol errors ..." mean
    in the first paragraph of section 6.11? Do initiators communicate to each
    other?? If not, I would suggest rather "initiator detects ...".
    +++ will fix text +++
    *** Section 6.11 : QUESTION
    Draft says:
      When the session timeout (the connection state timeout for the last
      failed connection) happens on the target, it takes the following
      actions:

        - Resets or closes the TCP connections (closes the session).
        - Aborts all Tasks in the task set for the corresponding initi-
          ator.

    What does the "corresponding initiator" mean? We think (:-)), that
    there is only one initiator for the session. The only possible
    explanation we see for that paragraph is, that the target should
    abort also other tasks of the same in __other__ sessions, but
    why?
    +++ your interpretation is correct - the statement means the initiator that "owned" the session.

    Are you suggesting other wording?
    ++++
    *** Section 6.12.1 : TYPO
    In sentence "Sequence reception timeout is generally a
      large enough value to allow the data sequence transfer to be com-
      plete."
    should be "completed" instead, (at least IMHO).
    +++ will foix +++
    *** Section 6.13 : TYPO
    "Digest failure recovery is comprises ... " is TYPO IMHO. I would say
    "Digest failure recovery comprises ..." only.
    +++ will fix text +++
    Best regards
    &
    Thanks for replay in advance

    Tomas Bartusek & Ivan Pavelka

    --------------------------------------

    Tomas Bartusek

    SW Engineer
    Tomas.Bartusek@s3group.com

    Tel:+420-2-33099184

    Silicon & Software Systems     www.s3group.com       Fax:+420-2-33099151
    Philips building, Safrankova 1, 155 00 Prague 5, Czech Republic






Home

Last updated: Mon Jul 15 17:18:52 2002
11326 messages in chronological order