This message is from the T13 list server.


At 7:25 AM -0600 9/26/03, Hale Landis wrote:
This message is from the T13 list server.


Lets see, the SATA problem list goes something like this:


* cables can be a problem.
* connectors can be a problem.
* clock signals can cause errors.
* any kind of noise causes errors.

This is different than anything else, how?


* these problems cause the device to reset in the
  middle of commands.

That isn't what they are supposed to do, but that indeed may be happening. This is something that needs to be correctly handled on the drive side of the equation.


* there is little or no way to detect these resets.

Again, this is something that is likely within the black-box at the device end of the cable.


* command parameter data appears to get corrupted
  causing ABRT errors or worse data compare errors.

Can't help you there.




But I have a suggestion for you SATA testers out there - put two
SATA drives on the same system and power supply and then do lots
of Read/Write commands to both drives at the *same* time.  That
means one drive on the controller's primay side and one drive on
the same controller's secondary side.  See if that doesn't cause
errors on both drives.


Been there, done that. Works fine for me.





--

---------------------
I make stuff go.
---------------------
Larry Barras
ATA CPU Software
Apple Computer Inc.

Reply via email to