SORT BY:

LIST ORDER
THREAD
AUTHOR
SUBJECT


SEARCH

IPS HOME


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

    Re: SCSI URL scheme



    
    Right after I sent the attached, I thought of another reason that folks
    might me having a problem here, and that is, they might think that the
    Report Luns reports all the LUNS that are within the Storage Controller.
    As I understand it, this is not the case, but it only reports the LUs that
    are authorized to be seen by the questioning Host/Initiator.
    
    .
    .
    .
    John L. Hufferd
    
    
    
    John Hufferd/San Jose/IBM@IBMUS@ece.cmu.edu on 10/03/2000 12:19:19 PM
    
    Sent by:  owner-ips@ece.cmu.edu
    
    
    To:   ips@ece.cmu.edu
    cc:
    Subject:  Re: SCSI URL scheme
    
    
    
    OK, now I am missing something here.
    
    The note below seems to be talking about a Host trying to find a LU and
    Raghavendra suggest that it is a big effort to do the normal Report Luns
    and Inquiry.
    
    First off, this is suppose to be done (normally) once, but perhaps at
    various (logical) Plug & Play  points.  If we are talking about the Plug &
    Play points in time, then there could be some additional conversation.  But
    independent of that, which so far I think we have been, the Host determines
    ALL his addressable LUs when he comes up, and today that means "REPORT
    LUNS", followed by approprate INQUIRYs.  This is not a big deal at bring
    up.
    
    I have not followed the turn in the road that would have a Host that should
    know what is connected to it (and its WWN), needing to find the WWN at some
    other time.
    
    Now it is possible that we are somehow talking about something that is akin
    to the NAS approach to "Mount" (Or Drive Map).  In this mode I expect that
    the Client Host has been given some Name that needs to get resolved to an
    LU, or Set of LUs.  But even in that case, if the name get resolved so that
    it targets a given IP Storage Device (Storage Controller), it also should
    perform a Login, and then it can still perform the normal REPORT LUNs and
    an approprate Inquiry.   In most cases for Desktop systems this should be 1
    LU, probably the LU0.  But in any event I do not see the problem.  The
    Biggest problem the Host might find is if the information found in the VPD
    reported by the Storage Controller differs from what is expected. (They
    call this an administrative error and needs to get fixed by people.)
    
    In any event I do not see the value add for the Name to go all the way down
    to the LU.  In fact if does not we can use the normal network processes to
    connect a Client Host to the approprate Target Storage Controller.
    
    Third party commands, however, seem to be a different set of discussion
    items, so perhaps it would make since to focus on that area where it is not
    clear what the approach is, rather then work areas that have a technique
    that (IMHO) work OK.
    
    And it might be valuable to discuss the issue of Plug & Play, and how that
    is affected.  (By Plug and Play I mean the process of adding a new LU or
    taking an LU away while every thing else continues.)
    
    .
    .
    .
    John L. Hufferd
    
    
    
    Raghavendra Rao <jpr@divyaroot.India.Sun.COM>@ece.cmu.edu on 10/03/2000
    10:55:41 AM
    
    Please respond to Raghavendra Rao <jpr@divyaroot.India.Sun.COM>
    
    Sent by:  owner-ips@ece.cmu.edu
    
    
    To:   ips@ece.cmu.edu
    cc:
    Subject:  Re: SCSI URL scheme [WAS: Re: iSCSI: 2.2.6. Naming & mapping]
    
    
    
    
    > out-of-band).  I don't see the point of the initiator sending its "view"
    to
    > the target at any point.  If the initiator sends an incorrect view,
    you've
    > just created an error scenario which needs additional definition.
    
    Well, to extend what Julian said further, the initiator could ask the
    target
    for a LUN value corresponding to a WWN as part of the authentication
    process,
    instead of presenting its view - whether right or wrong. But something like
    this needs to be done as there is no easier/faster way to translate a LU
    WWN
    to a LUN value.
    
    This may also help in building an authentication method for a LU access
    where
    no such thing exists today ?
    
    
    > case, the target will send its "view" for that initiator in REPORT LUNS
    > SCSI command after login (which is actually the layer that cares about
    the
    > LUN Map at all anyway).
    >
    
    This will be a problem if the storage controller has a lot of LUNs. Since
    REPORT LUNS only returns a list of LUNs, the initiator still has to send an
    INQUIRY (page 83) request to each LU to to find a matching LU WWN (or until
    a matching WWN is found).
    
    -JP
    
    
    
    
    
    


Home

Last updated: Tue Sep 04 01:06:52 2001
6315 messages in chronological order