SMART worked, the warnings just didn't get through...

Elledan

[H]ard|DCer of the Month - April 2010
Joined
Oct 18, 2001
Messages
15,913
A colleague of mine had one of his Samsung HDDs fail two days ago, he managed to pull off the last bits of important data, but the drive is completely toast now. Apparently during boot the BIOS reported that the SMART status on the HDD was fine, yet if you went into the BIOS to look up the details it would report that multiple SMART tests had failed, which most decidedly didn't make the status 'fine'. Is this common?

Also, I know that with a program like SpeedFan one can check the SMART values, but is there an utility which pops up a warning if one or more of those SMART values go consistently out of range?

I know that many people consider SMART to be fairly useless, but in this case it were only the constant backups which my colleague makes which saved the day. Knowing about the failed SMART tests would have allowed him to transfer the programs folder and such to a new HDD, saving a few days of restoring things.

Thanks for your thoughts :)
 
http://prefetch.net/articles/diskdrives.smart.html
The smartctl(1m) examples up to this point have provided invaluable status information, but the examples required input from the keyboard. To setup automated alerts when devices fail or a SMART attribute changes, a cron job can be developed to check smartctl(1m) self-test logs, or the smartd(1m) daemon can be configured to analyze devices, and report anomalies when they are detected.

They don't give specific examples, but they say it can report when anomalies are detected. That page is pretty *NIX-y, but smartmontools does run on Windows as well.
 
Back
Top