SORT BY:

LIST ORDER
THREAD
AUTHOR
SUBJECT


SEARCH

IPS HOME


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

    Re: iSCSI: enquiry key




    Bob,

    "Robert D. Russell" <rdr@io.iol.unh.edu> wrote on 20-12-2001 12:48:31:

    > Julian:
    >
    > A few questions:
    >
    > 1. Does the value of MaxOutstandingR2T include the implied initial R2T
    >    when operating in unsolicited mode (as described in 2.2.5)?
    >
    >    I believe the answer is "no", but would suggest a minor rewording
    >    in Appendix D item 29 to remove any doubt:
    >    "Initiator and target negotiate the maximum number of outstanding R2Ts
    >    per task, excluding any implied initial R2T that might be part of that
    >    task."
    >

    OK
    >
    > 2. I believe the upper limit for MaxRecvPDULength should be 2**24-1,
    >    not 2**24 as is currently shown in Appendix D item 24.
    >    The reason is that the Data Segment Length field in the header is only
    >    24 bits, so 2**24-1 is the largest number that can be stored in it
    >    (i.e., a PDU with 2**24 bytes of data could never be sent).
    >    To be consistent, if the limit for MaxRecvPDULength is changed to be
    >    2**24-1, perhaps we should also change the limit for MaxBurstSize and
    >    FirstBurstSize to be the same 2**24-1.
    >

    OK
    >
    > 3. In section 2.2.4 it says: The value "?" with any key has the meaning
    >    of enquiry and should be answered with the current value of
    >    "NotUnderstood."
    >
    > 3a.It is not clear what this means when keys are "one-way".  For example,
    >    assume the initiator sends the target "MaxRecvPDULength=?".  Should the
    >    target send back the maximum PDU length it (the target) expects to
    >    receive from the initiator, or should the target send back the maximum
    >    PDU length it (the target) will ever send to the initiator?
    >    The same question applies to "FMarker=?" and "RFMarkInt=?" and
    >    "SFMarkInt=?".  Regardless of what the answer to this question is,
    >    another question immediately arises -- how does one side ask for the
    >    other value?
    >

    Yes - I will clarify that both values have to sent and state how

    > 3b.It is not clear if the usual restrictions to LO, IO, and FFPO
    >    should apply to these enquiries.  For example, during a login for a
    >    second (third, fourth, ...) connection in a session, can one side
    >    send "MaxBurstSize=?" to the other side to obtain the current value
    >    for the session, even though MaxBurstSize is an LO parameter and can
    >    not be negotiated for the second (third, fourth, ...) connection?
    >    Or in another example, can a text request from the initiator include
    >    "MaxBurstSize=?" even though MaxBurstSize cannot be negotiated or
    >    changed after the Leading Login?
    >

    will clarify
    >    If the answer to these examples is "yes", then further clarification in
    >    the standard is necessary to indicate that the "key=?" can be sent
    >    at any time and is not restricted by the LO, IO, and FFPO labels.
    >
    >    In the "yes" case, we also need to clarify whether it is possible to
    >    send "key=?" enquiries for operational keys during security negotiation,
    >    and vice versa.
    >
    >    If the answer to these examples is "no", then the value of an enquiry
    >    appears to be marginal, and I would suggest that it be eliminated
    >    from the spec (when can it be used to any benefit?).
    >
    >    My personal opinion is that "key=?" seems fairly useless -- if they are
    >    to operate correctly, both sides of a connection really have to know the
    >    values of all the keys at all times, and the rules for negotiation make
    >    it clear how and when these values change.  The only real "enquiry"
    >    situation is when an initiator is attempting to discover targets, and
    >    this is handled completely differently by the use of "SendTargets".
    >    So when is there a use for either side to send "key=?".  Unless there
    >    is a clear example where this is useful, we should eliminate it and thus
    >    remove another special case during parameter processing.
    >

    I am not sure that we have all the use scenarios to afford removing it
    >
    > Bob Russell
    > InterOperability Lab
    > University of New Hampshire
    > rdr@iol.unh.edu
    > 603-862-3774
    >


Home

Last updated: Thu Jan 03 15:17:55 2002
8274 messages in chronological order