SMART for RBC

[Devon Worrell]

* From the T10 (formerly SCSI) Reflector (t10 at symbios.com), posted by:
* Devon Worrell <worrell at dt.wdc.com>
*
Hi,

I was curious as to why the technique that was specified in the
SMART application guide is not being used. There the Fault/Failure
Reporting Page (1Ch) that allows a reporting of the failure on a basis
specified in the page. If the page is not used then the Default
would be MRIE of 4 (report recoverable error) and a Interval
Timer of 6000 (every 10 minutes).

I don't think that it is necessary to report on ONLY the TEST UNIT
READY command. Any normal command can be terminated with the recovered
error and Sense Code 0x5D. It is true that a Host could then poll
using the TEST UNIT READY to sense the error, but the error could be
lost this way to "Other" OS and Application commands. The error
is only reported once in the specified Interval. Thus the driver
would need to hold any occurrences of 01/5D for later retrieval by
the appropriate application or driver software.

Currently the DVD devices don't go to the detail that is being
discussed. Those devices only report 01/5D/00 Device Threshold
exceeded or 01/5D/01 Media Threshold Exceeded.

If some small changes to the Sense Code Qualifier and the change to
use the Page or at least the page defaults, is made then it will be
compatible with the application Guide and MtFuji...

I don't quite understand the reason for two nibbles. Are they independent
or is there some encoding? Are the Upper Nibble values used as the
hard failure codes and the Lower Nibble the May Fail codes? If they
are independent then shouldn't there be a value in each Nibble that
indicates "No Error"?

If the codes are used, then I would propose the following Sense Code
Qualifier change:


Sense Key  = 0x1
Sense Code = 0x5D

Sense Qualifier  =

	Value		Meaning
	-----		--------------------------------------------
upper nibble:
  	0h 		General Drive or Media failure
	1h 		Hardware impending failure
	2h		Controller impending failure
	3h		Data Channel impending failure
	4h		Servo impending failure
	5h		Spindle impending failure
	6h		Firmware impending failure
	7h		Channel parametrics indicate impending failure
	8h		Controller detected impending failure.
	9h - Dh: 	Reserved
	Eh:		No impending failure to report
lower nibble:
	0h 		General Drive Error threshold exceeding limits.
	1h		General Media Error threshold exceeding limits.
	2h		Data Error Rate exceeding limits.
	3h		Seek Error Rate exceeding limits.
	4h		LBA reassignment exceeding limits.
	5h		Access Times exceeding limits.
	6h		Start Unit Times exceeding limits.
	7h		Throughput performance exceeding limits
	8h		Seek time performance exceeding limits
	9h		Spin-up retry count exceeding limits
	Ah		Drive calibration retry count exceeding limits
	Bh-Fh		Reserved.	

I am proposing that the Page 1Ch "Functionality" if not the actual page
be adopted. If the page is present then the Host can sense the time, change
the time and turn OFF SMART, if not required. If the page is not
present then the drive makes use of the defaults as specified above.

Thanks,


Devon,

    _/_/_/_/    _/_/_/_/    _/      _/ | Devon B. Worrell <worrell at dt.wdc.com>
   _/      _/  _/      _/  _/      _/  | Senior Staff Engineer
  _/      _/  _/_/_/_/    _/  _/  _/   | Western Digital Corporation
 _/      _/  _/       _/ _/_/_/_/_/    | 8105 Irvine Center Dr, Irvine CA 92618
_/_/_/_/    _/_/_/_/_/  _/      _/     | Host Interface & Technology Staging
                                       | Ph: (714) 932-7042 Fax: (714) 932-8806

"Every truth passes through three stages before it is recognized.
 In the first it is ridiculed, in the second it is opposed, in the
 third it is regarded as self-evident!"

"Not speaking in an official capacity for my employer"

PGP Key Fingerprint = 4D37 E5B5 45FD 66DB 9DDB  1785 F060 6662 E5BA D357



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