At Fri, 12 Aug 2011 22:35:12 +0100 (BST) CentOS mailing list centos@centos.org wrote:
On Fri, 12 Aug 2011, Robert Heller wrote:
*snip*
The disk's *firmware* updated itself. So long as the *disk* is powered up and spinning, its *firmware* is 'running' (or runs when the disk is accessed or something like that). Modern disks are a long, long way from the simple MFM drives of the 1970s (which presented little more than a buffered interface to the drive mechanism and read/write heads to the host controller -- eg little more than a simple floppy disk drive) -- modern disks have actual embedded micro-processors on them doing various stuff, including monitoring and logging things like sector errors, drive temp., and so on.
So does accessing some of the smart data cause any disk i/o at all, or is this all done from directly firmware?
Depends. Some of the SMART information is stored on special sectors of the platter (sectors that are not part of the sectors available for normal use). Some of the information is fixed burned into the (E*)ROM(s) (eg serial numbers) on the disk's logic board and some information is 'live' (eg current disk temperature). There IS I/O between the host system and the disk's logic board.
Thanks Robert. Reason I ask is because I'm using GKrellM to monitor my HDD temperatures. So I just wondered if querying the HDD temps to often would create more disk i/o ?
I believe the HDD temps are a 'live' sensor reading. The firmware might log record/excessive high temps to a reserved sector as part of the disk's long term health status, but it is unlikely to be constantly logging routine temperatures.
Kind Regards,
Keith Roberts
Websites: http://www.karsites.net http://www.php-debuggers.net http://www.raised-from-the-dead.org.uk
All email addresses are challenge-response protected with TMDA [http://tmda.net]
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos