SORT BY:

LIST ORDER
THREAD
AUTHOR
SUBJECT


SEARCH

IPS HOME


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

    RE: iscsi: rev 05 2.5.1 requires ACA support



    
    
    Charles,
    
    After an error if you have commands in flight you want them all dropped
    until you specifically reset the ACA and restart the queue (prevent things
    to be executed out of order).
    
    Julo
    
    "Charles Binford" <Charles.Binford@BlueSpruceNet.com> on 15/03/2001
    20:35:26
    
    Please respond to "Charles Binford" <Charles.Binford@BlueSpruceNet.com>
    
    To:   "IPS Reflector" <ips@ece.cmu.edu>
    cc:
    Subject:  RE: iscsi: rev 05 2.5.1 requires ACA support
    
    
    
    
    Ralph, you missed a reference in iSCSI where ACA is required.  In 7.2 (p
    86)
    it says:
    
      "ACA helps preserving ordered command execution
       in presence of errors. As iSCSI can have many commands
       in-flight between initiator and target iSCSI mandates
       support for ACA."
    
    I agree with Ralph that iSCSI should not be mandating ACA support.
    However,
    I don't really understand what the problem/scenario is that ACA is supposed
    to be solving.  Can someone please explain that to me?  Before I start
    throwing more darts at ACA I'd like to better understand the perceived
    value.
    
    
    Charles Binford
    Blue Spruce Networks
    office: (316) 315-0382 / cell: (316) 210-6404
    e-fax: (509) 756-4425
    
    
    
    
    -----Original Message-----
    From: owner-ips@ece.cmu.edu [mailto:owner-ips@ece.cmu.edu]On Behalf Of
    Ralph Weber
    Sent: Monday, March 12, 2001 10:17 PM
    To: IPS Reflector
    Subject: iscsi: rev 05 2.5.1 requires ACA support
    
    
    The following two 2.5.1 statements require the use
    of ACA and there is no way for an initiator that
    doesn't implement ACA to stop them.
    
       For the <Clear Task Set>, ...
       The target MUST then enter the ACA state for any
       initiator for which it had pending tasks.
    
       In addition, for the <Target Warm Reset> the target
       enters the ACA state on all sessions and all LUs on
       which an AE was sent.
    
    Have you all signed the operating system vendors up for
    rewrites of their class drivers to support ACA?  If not
    anticipate some serious deadlocks when iSCSI is first
    attempted with a non ACA class driver, since the driver
    will never clear these ACA conditions and no progress
    will be made after that.
    
    Alternatively, a way can be found to give the initiator
    control over the "MUST" statements.
    
    Thanks.
    
    Ralph...
    
    
    
    
    
    
    
    
    


Home

Last updated: Tue Sep 04 01:05:19 2001
6315 messages in chronological order