Assembly failure
Brian Candler wrote:
------------------------------------
One final point. I would like to be able to monitor for suspect or failed drives. Is my best bet to look at /proc/mdstat output and identify drives which have been kicked out of the array? Or to monitor SMART variables (in that case though I need to decide which ones are the most important to monitor, and what thresholds to set)?
-------------------------------------------
For years I have used smartd without issues and it will log and email anomalies as they occur.
It is also advisable to regularly "scrub" all md devices, to flush out faulty sectors:
echo check > /sys/block/mdX/md/sync_action
See Documentation/md.txt for details.
Regards,
Richard