On Mon, Aug 15, 2011 at 11:19:22AM -0700, John R Pierce wrote:
So I'm curious how SAS JBOD arrays and linux MDraid as implemented in CentOS6, and SES (SCSI/SAS Enclosure Services) backplane controllers 'get along' and how much configuration is needed to get the warning lights to work properly.
scenario: whitebox server with a SAS backplane or two, daisy chained on a SAS HBA (like an LSI Logic 2008), and disks organized as several raid5/6 using linux mdraid. Does this system have to be configured to ensure that the activity and failure lights match the drive bays, or does the SAS/SES spec all sort that out 'plug and play' ?
does mdraid properly set those warning lights automatically on a failed raid element?
Afaik the LEDs won't work out-of-the-box.
There's no daemon currently in Linux responsible for monitoring disk status and lighting up SES LEDs accordingly..
Feel free to write such a daemon! :) (or even better, prove me wrong)
-- Pasi