[SGVLUG] monitoring hard drives

[Claude Felizardo]

So, after reading the Google paper on disk drive failures, I was
wondering what people are doing to monitor their hard drives?  Does
anyone use smartd or run smartctl in a cron or what?

Slashdot and referenced paper:
    http://hardware.slashdot.org/article.pl?sid=07/02/18/0420247
    http://216.239.37.132/papers/disk_failures.pdf

To monitor the RAID-5 on my file server at home I have a cron that
compares the output of /proc/mdstat against an expected result and
emails me if there's a diff.  It warned me of a sync problem after a
power failure a few years ago.  Otherwise the only other thing I have
is SystemGraph which uses RRDTool in a cron to capture and graph some
measurements such as disk temp and disk usage as well as cpu and
memory usage but you have to remember to check it.

    http://www.decagon.de/sw/systemgraph/

check out the screenshots

btw, does smartctl support SATA drives yet?

claude