[AB^2=BCxAC]

I have a 1ghz Ti Book with a 60gig drive and I've been habitually backing up to a Lacie firewire drive.

1. Is there a program that can run at startup and check every few hours the S.M.A.R.T. status of your drives and alert you if it changes?

Right now, I just check Disk Utility periodically, but it'd be nice to get some warning if my laptop's drive might be ready to walk.

2. How come Disk Utility can't tell me the S.M.A.R.T. status of my Lacie Firewire drive? There isn't even a field there to say verified or not. Does Disk Utility overlook firewire drives, or is it my particular drive that doesn't have the function? Or is it going to fail soon?

[tooki]

1. DiskWarrior is one program that can periodically check the status and alert you if something goes wrong.

2. Only ATA and SCSI disks can report SMART status, USB and FireWire storage standards do not include that feature.

tooki

[bleee]

Originally posted by AB^2=BCxAC:
I have a 1ghz Ti Book with a 60gig drive and I've been habitually backing up to a Lacie firewire drive.

1. Is there a program that can run at startup and check every few hours the S.M.A.R.T. status of your drives and alert you if it changes?

Right now, I just check Disk Utility periodically, but it'd be nice to get some warning if my laptop's drive might be ready to walk.

2. How come Disk Utility can't tell me the S.M.A.R.T. status of my Lacie Firewire drive? There isn't even a field there to say verified or not. Does Disk Utility overlook firewire drives, or is it my particular drive that doesn't have the function? Or is it going to fail soon?

I've found smart to be inaccurate and unreliable, I had a 20GB IBM hard drive fail on me back in December, it didn't totally die at first it started with clicking sound basically the arm was having mechanical failure. My machine would boot sometimes I would be able to do 20 minutes to an hours worth of work before the arm would seize up and the machine would lockup. I ran the Apple Diagnostics CD and Disk warrior, Disk Warrior said SMART reported the drive to be functioning normally. When it died I took it into an Apple Authorized dealer they replaced the hard drive and my iBook has been fine ever since.

[McFarmer]

I second that. In January I had two HDs die in my PowerBook. First the original 80GB 5400rpm Toshiba drive and then the 100GB Seagate replacement drive. The second one obviously a dud.

In both cases the drive died within 10 days starting to make loud noises at first, then clicking sounds and would finally get massive read/write errors and freeze the PowerBook.
After a reboot I could use it again for another 5-20 minutes, but eventually the HD would lock up again. And so on.

In all these cases I was no longer able to do a full backup (but I had previous ones, so not much was lost) as the HD would never function long enough. Yet in all this SMART _still_ listed as 'verified'. Even after so many read/write failures and with the HD in a state too wrecked to allow making a full backup.

Eventually, once the HD refused to spin up and the PB no longer booted from it, the SMART status would finally report 'failed' - when I booted off an external HD to check it. Big help! If you did not know by then (and believe me you will because of the lockups and noises it makes), it will be too late to take any actions - like making a backup.

I had this happen twice in the very same pattern.

Morale of the story:

SMART is useless for the purpose you want. It will NOT warn you in time of an imminent HD failure. So don't bother with that. All it will do is 'warn' you after you already experienced massive problems. It will switch to 'failed' only when the HD is about to die within the hour. Too late at that point for anything.

In a nutshell: the SMART status is only a final verification that the HD is indeed dead. But not an early warning system.

[tooki]

Well, it can warn you early. Think of it like this: if SMART says your drive is going to die, you can be pretty darned certain it will.

But just because SMART hasn't detected a fault is not proof that the drive is fine.

This is like many things: you can't predict random events, like random drive failure. But if you find an indicator of a problem, you can make a much more valid prediction.

(For example: If your car's oil temperature light comes on, you know that something is wrong, and that the condition will result in engine damage. But if no indicator lights are on, that is no proof that your engine's timing belt won't suddenly break and result in crushed valves tomorrow.)

Note also that different drives have different levels of SMART support. Many variables can be monitored, but not all drives monitor all of them.

tooki