SBC-4 wording on SANITIZE command behavior

Gerry Houlder gerry.houlder at seagate.com
Tue Jan 6 11:58:33 PST 2015


Formatted message: <a href="http://www.t10.org/cgi-bin/ac.pl?t=r&f=r1501062_f.htm">HTML-formatted message</a>

I agree that that sentence in intended to cover the case I am concerned
about, but the fact that that error case is not mentioned in 5.24.1 leaves
doubt in some readers as to how this should be handled.
I am preparing a proposal to add words to 5.24.1 to resolve the confusion.
It will be on the agenda for next week.
On Tue, Jan 6, 2015 at 1:32 PM, John Geldman (jgeldman) <jgeldman at micron.com
> wrote:
>  Does the restriction description in the model clause (pdf page 64 in
> sbc4r04) cover this? (see 4.11.1)
>
> Note the last sentence in the third paragraph (more included for
> background)
>
>
>
> An application client may request that a sanitize operation be performed
> in the restricted completion mode or the unrestricted completion mode (see
> 4.11.3) using the AUSE bit in the SANITIZE command (see 5.24).
>
> In the unrestricted completion mode, a SANITIZE command with the EXIT
> FAILURE MODE service action exits a failed sanitize operation.
>
> In the restricted completion mode, the only method to exit a failed
> sanitize operation is for a SANITIZE command to request another sanitize
> operation and for that operation to complete without error. If a sanitize
> operation in the restricted completion mode completes with an error, and a
> subsequent SANITIZE command requests the unrestricted completion mode
> (i.e., the AUSE bit set to one), then the device server shall terminate
> that SANITIZE command as described in 5.24.1.
>
>
>
> Happy New Year,
>
> John
>
>
>
>
>
>
>
> John Geldman
>
> Director, Industry Standards
>
> Micron Semiconductor
>
> 540 Alder Drive
>
> Milpitas, CA 95035
>
> Phone:  1-408-822-0348
>
> Mobile: 1-510-449-3597
>
> email:jgeldman at micron.com
>
>
>
>
>
> *From:* owner-t10 at t10.org [mailto:owner-t10 at t10.org] *On Behalf Of *Gerry
> Houlder
> *Sent:* Tuesday, January 06, 2015 7:29 AM
> *To:* T10 Reflector
> *Subject:* SBC-4 wording on SANITIZE command behavior
>
>
>
> This is text in the SANITIZE command describing response to AUSE bit=0.
>
>
>
> If:
>
> a) the AUSE bit is set to zero in the SANITIZE command that requested a
> sanitize operation;
>
> b) the specified sanitize operation completes with an error; and
>
> c) a subsequent SANITIZE command with the EXIT FAILURE MODE service action
> is received,
>
>
>
> then the device sever shall terminate that subsequent SANITIZE command
> with CHECK CONDITION status with the sense key set to ILLEGAL REQUEST and
> the additional sense code set to INVALID FIELD IN CDB.
>
>
>
> I recall discussion of a sanitize failure with AUSE=0 that a subsequent
> SANITIZE with an option other than EXIT FAILURE MODE also had to have
> AUSE=0 in order to be accepted by the device server. i thought we intended
> that a subsequent SANITIZE command with AUSE=1 would also result in illegal
> request but i can't find any words to that effect. Did we intentionally
> drop that requirement or did we forget to add that requirement? Or is it
> somewhere else and i just didn't find it?
>



More information about the T10 mailing list