Question on Read-Write Error Recovery mode page PER operation

Gerry.Houlder at seagate.com Gerry.Houlder at seagate.com
Mon Nov 10 08:02:44 PST 2008


* From the T10 Reflector (t10 at t10.org), posted by:
* Gerry.Houlder at seagate.com
*
A careful reading of proposal 07-447r1 discloses  what I hope is another
unintended change:
In table 126, cases with PER=0 have the wording "the device server shall
not report recovered read errors". The word "read" was added to that
sentence. Does this now mean that recovered write errors must still be
reported? That is a technical change from what has been done for many years
and would be an undesireable change. I think the word "read" should be
changed to "read or write" because the next sentence adds mention of
"performing a read or write operation".
	     "Knight,							   
	     Frederick" 						   
	     <Frederick.Knight						To 
	     @netapp.com>	       <Bill.Martin at emulex.com>,	   
	     Sent by:		       <t10 at t10.org>			   
	     owner-t10 at t10.org						cc 
	     No Phone Info	       <Narayan.Ayalasomayajula at emulex.com 
	     Available		       >				   
								   Subject 
				       RE: Question on Read-Write Error    
	     11/09/2008 07:04	       Recovery mode page PER operation    
	     PM 							   
Well, SBC-2 and SBC3r11 have the NOT.  07-447r1 is the one that changed it
|from this older text:
A post error (PER) bit set to one specifies that the device server shall
report recovered errors. A PER bit set to
zero specifies that the device server shall not report recovered errors,
and the device server shall perform
error recovery procedures within the limits established by the error
recovery parameters.
    Fred Knight
From: Bill.Martin at Emulex.Com [mailto:Bill.Martin at Emulex.Com]
Sent: Friday, November 07, 2008 6:43 PM
To: t10 at t10.org
Cc: Narayan.Ayalasomayajula at Emulex.Com
Subject: Question on Read-Write Error Recovery mode page PER operation
In the description of PER in the Read-Write Error Recovery mode page, the
following is stated:
A post error (PER) bit set to one specifies that if a recovered read error
occurs during a command performing a read or write operation, then the
device server shall terminate the command with CHECK CONDITION status with
the sense key set to RECOVERED ERROR. A PER bit set to zero specifies that
if a recovered read error occurs during a command performing a read or
write operation, then the device server shall terminate the command with
CHECK CONDITION status, and shall perform error recovery procedures within
the limits established by the error recovery parameters. If the DTE bit is
set to one, then the PER bit shall be set to one.
It seems that the first and second sentences require mostly the same thing
as far as reporting a CHECK CONNDITION.  I think that the second sentence
is missing a “not” in the device server shall NOT terminate the command
with CHECK CONDITION status, …
This interpretation is consistent with the first two rows of table 128
also, which state “The device server shall not report recovered read
errors. The device server shall terminate a command performing a read or
write operation with CHECK CONDITION status before the transfer count is
exhausted only if an unrecovered error is detected.”
Bill Martin
Emulex
Office of Technology
Industry Standards
916 772-3658
916 765-6875 (Cell)
bill.martin at emulex.com
*
* 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