CA/ACA question

Ralph Weber ralphoweber at compuserve.com
Fri Feb 7 16:25:19 PST 2003


* From the T10 Reflector (t10 at t10.org), posted by:
* Ralph Weber <ralphoweber at compuserve.com>
*
The sentence means that an application client on one
initiator can detect the generation of a CA condition
for another initiator by:

  1) Setting TST to have all initiators in one task set
     (i.e., 000b),
  2) Setting QErr such that a CA condition condition
     results in all tasks in the task set being aborted
     (i.e., 01b) and,
  3) Setting TAS to have TASK ABORTED status returned
     for tasks aborted by another initiator (i.e., 1b).

A suitably minded SCSI compliance engineer can use these
settings along with an out-of-band communications path
between two initiators to verify that  the establishment
of a CA condition immediately results in tasks for the
other initiator being aborted.

This is an obscure side effect of the settings noted above.

Note that even though CA will will disappear in SAM-3 r05
and SPC-3 r11 based on the January T10 approval of 03-002r3,
the effect described above will remain.

Regards,

.Ralph

Ken Craig wrote:

> 
>* From the T10 Reflector (t10 at t10.org), posted by:
>* "Ken Craig" <kcraig at istor.com>
>*
>I have a question concerning Note f in
>Table 25 of Section 5.9.1.1 of SAM3-rev04.
>
>The last sentence of this note states that
>tasks that are aborted for the application
>client make a CA condition detectable by
>the client.  I do not understand the meaning
>of this statement after reading Section 5.7.2
>on the results of a SCSI initiator port
>aborting its own tasks.
>
>Any help would be greatly appreciated.
>
>Thanks,
>Kenneth Ray Craig, Jr.
>*
>* For T10 Reflector information, send a message with
>* 'info t10' (no quotes) in the message body to majordomo at t10.org
>
>  
>



*
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at t10.org




More information about the T10 mailing list