Drive behavior issue when START STOP unit forces drive to idle

[matthew.jacob at sun.com]

* From the T10 Reflector (t10 at t10.org), posted by:
* matthew.jacob at sun.com
*
> Possible solution: change the START STOP UNIT command behavior so that
> sending the command causes the drive to transition immediately to the
> requested power condition but don't disable the timers. Then if the drive
> goes to active power condition to process a media access command, it will
> still go to idle  power condition based on the timers. This would at least
> limit the unnecessary active time of the drive to the values set in the
> timers.
What is the behaviour of the timers in the presence of a START command 
(e.g., from another initiator)? Does this disable the timers? Does a 
random media access command or START UNIT from some random initiator 
also confuse the initiator that thought it had put the drive into a 
power-down-after-a-while mode? Do you want to make START/STOP or the 
times keyed on per-initiator (i.e., have an implicit RESERVATION with 
power mode)?
I think the problem here is that this command, like so many that date 
back to 1984 and previously, are not really suitable for multi-initator 
environments.
IMO, such as it is, these commands that are not well suited for a 
distributed environment really can't be fixed and bashed into such an 
environment, and need to be protected/utilized under other mechanisms 
that do undestand such environments- and those called out specifically. 
That way you can say that the power state has limitations defined by not 
handling multi-initiator cases. This will force people to make sure that 
if one system uses idle mode timers, that all systems must use (and set) 
idle mode timers in a multi-initiator configuration.
-matt
*
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at t10.org