SSC-4: ERASE(16) command description flaw?
tjmac at tolisgroup.com
Fri Apr 23 12:33:59 PDT 2010
* From the T10 Reflector (t10 at t10.org), posted by:
* Tim Jones <tjmac at tolisgroup.com>
On Apr 23, 2010, at 10:18 AM, Ballard, Curtis C (StorageWorks) wrote:
> I believe the difference is that the logical position following an
ERASE(16) command with a long bit set to one is always undefined. The
standard doesn't have any qualifiers about the command having been
> When the locate operation fails the logical position is now undefined even
when the long bit is set to zero.
> It would also be true to drop that qualifier since the position is
undefined for both values of long following a locate failure.
After reviewing this scenario back to SSC-2, I agree that this is simply the
case in both situations. The "probability" is that a LONG bit of 1 will not
result in this if the current location is incorrect as the ERASE command will
perform a select and position to the specified partition.
My curiosity was one of an ISV so that we make sure that we're handling both
situation properly. It becomes very important in dealing properly with tapes
created in an LTO-5 with the LTFS functionality.
Tim Jones tjmac at tolisgroup.com
TOLIS Group, Inc. http://www.productionbackup.com
480-505-0488 x227 480-505-0492 (fax)
BRU ... Because it's the RESTORE that matters
* 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