SORT BY:

LIST ORDER
THREAD
AUTHOR
SUBJECT


SEARCH

IPS HOME


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

    RE: iSCSI Autosense Consensus, Connection next steps



    > -----Original Message-----
    > From: owner-ips@ece.cmu.edu [mailto:owner-ips@ece.cmu.edu]On Behalf Of
    > Black_David@emc.com
    > Sent: Sunday, September 03, 2000 8:20 AM
    > To: julian_satran@il.ibm.com; ips@ece.cmu.edu
    > Cc: Black_David@emc.com
    > Subject: iSCSI Autosense Consensus, Connection next steps
    >
    >
    > > It looks like we have consensus - but the chairmen have the call.
    >
    > Indeed we do, and I apologize for the delay, due to an inability to
    > get connected.
    >
    > First, on Autosense, I believe that rough consensus exists for
    > iSCSI to require Autosense.  I have seen only one objection;
    > If anyone other than Doug Otis disagrees with this, please
    > say so on the list.
    
    I am in favor of removing the A bit to control Autosense.  I do not disagree
    with respect to that change.  The only reason for having an Autosense option
    is to support legacy applications.  As things like tape drives, loaders,
    etc, use untagged CA environments without suffering limitations even in a
    packet mode, to preserve the application (as you would not want to bit
    fiddle the CDBs within the client side drivers) allowing lack of automatic
    sense *data* at the adapter may provide a solution.  It would not violate
    SAM-2.  Although you certainly can enforce interface compliance, you have
    less ability to enforce application compliance sitting  above a SCSI API
    that does not make decisions enabling ACA. Microsoft assumes the SCSI
    interface is SCSI-II as example.
    
    As network based SCSI is likely to introduce multiple initiators into the
    picture normally occupied by a single adapter, CA still provides untagged
    protection for this sloth equipment.  The legacy application would not have
    knowledge of this limitation and so the option was on the wrong side of the
    interface.  There is no need for any option if a back door is desired and no
    sense data does not cause an error within the client side driver. (You can
    always insist such is an error, but if it increases customer satisfaction it
    may be hard to enforce.)  I suspect there are many archive applications that
    will be excluded from a pure Autosense environment as it relates to a
    network version of SCSI.  Archive equipment tends to have a longer life
    cycle than disk drives and as such remains behind the curve with respect to
    conformance to the latest versions optimized for disk drives that have a
    life cycle in months.
    
    Doug
    
    
    
    
    
    


Home

Last updated: Tue Sep 04 01:07:35 2001
6315 messages in chronological order