Re: Software RAID (was [OT] 19"/2U Cases)

martin f krafft wrote:
> Ever since I've instituted smartd on all my machines (about 3 years
> ago), I've only ever *re*actively replaced a RAID drive once. All
> the other times, I *pro*actively replaced it after smartd spat out
> errors.
>

This week, I've been working on recovering a professor's hard drive for him. He claims that, for a week, he was getting messages at boot time claiming "something about the drive operating outside normal parameters..." which is how Dell machines report SMART problems. He'd just hit F1 to get past the warning... for about a week, that is. Then, his machine stopped booting... and I got called in. So, score another point for SMART.

I should point out something particular about Martin's tale: he was running smartd. If you're just using the SMART built in the BIOS, then you're only going to get warned of disk issues at boot time. If you're running a machine that gets rebooted once every three months and let's say that SMART starts issuing warnings about a week before the drive is kaput, then you've got a 1-in-12 chance of catching a SMART problem at boot time.

So, if you're not running some SMART monitoring in the OS, then... don't bother with SMART at all, I'd say...

• Joe

-- 
To UNSUBSCRIBE, email to debian-isp-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org


application/x-pkcs7-signature attachment: S/MIME Cryptographic Signature