SORT BY:

LIST ORDER
THREAD
AUTHOR
SUBJECT


SEARCH

IPS HOME


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

    iSCSI: MaxCmdSN rule


    • To: <ips@ece.cmu.edu>
    • Subject: iSCSI: MaxCmdSN rule
    • From: "Buck Landry" <blandry@crossroads.com>
    • Date: Thu, 7 Feb 2002 14:40:16 -0600
    • content-class: urn:content-classes:message
    • Content-Type: multipart/alternative;boundary="----_=_NextPart_001_01C1B017.A69C7610"
    • Sender: owner-ips@ece.cmu.edu
    • Thread-Index: AcGv2E8+xCZeXc/8SOaWwEDSJyqAawAOywEw
    • Thread-Topic: iSCSI: case of none

    Hi all, I'm struggling over how to interpret this rule in section 1.2.2 of draft v10:
     
    >>> The target MUST NOT transmit a MaxCmdSN that is less than
            the last ExpCmdSN.
    <<<
     
    This confuses me because it says to me (in some cases) that a target must be able to handle an unlimited number of commands.  For example, let's say the target has just about reached the limit of the commands it can handle:
     
    T=>(Nop-In) (MaxCmdSN = 1, ExpCmdSN = 1)
     
    I=>(Scsi Read Cmd) (CmdSN = 1)
     
    T=>(Data-in) (MaxCmdSN = 1, ExpCmdSN = 2 /* acknowledging scsi read cmd */)
     
    Now the target may have many more data-ins to transmit.  But according to the rule, since the last ExpCmdSN the target transmitted was 2, the target is now forced to send a MaxCmdSN of 2:
     
    T=>(Data-in) (MaxCmdSN = 2, ExpCmdSN = 2)
     
    But this allows the initiator to send another command, even though the target can't handle it!
     
    Obviously something is wrong with my reasoning here (perhaps my interpretation of 'last ExpCmdSN', or when I"m supposed to acknowledge the scsi cmd); can anybody clarify?
     
    --buck
     
     
     
     


Home

Last updated: Thu Feb 07 18:18:00 2002
8703 messages in chronological order