Info on S.M.A.R.T hard drive monitoring

[JimInKS]

Has anyone had success using SMART hard drive monitoring to predict hard drive failures?

I have several Maxtor hard drives in service that seem to have a high failure rate and was wondering if running a SMART monitor program would be of any benefit?

I found a free program called HDD Health from

http://www.panterasoft.com/

that seems to work pretty well, but, I have not been able to find much information on what each parameter being monitored really means.

 

[SCFROMDC]

I use it in some instances, but if a smart test fails, no matter what it may mean, get the data off right then.
Then, if possible, run two drives in a mirror configuration....saves a lot of hassle...and if your having high HD failure rates, first, if they are more than three years old and are in use alot, well, that comes with the territory, but if they are newer, you should be on the phone with MAXTOR.
And in all seriousness, sometimes they just die, abruptly, monitored or not, so a mirror config, or some raid config would be a logical move...
S.M.A.R.T. is a great thing, but I have had warnings after a overnight, unscheduled, mysterious reboot, only to find that the warnings come too late.

http://www.deadcenter.us

I hear and I forget. I see and I remember. I do and I understand.
Confucius

 

[softwarehero]

The best product I saw that handles S.M.A.R.T was from DEK Software International. They have a program called DEKSI SmartCheck and I would highly recommend downloading it.

http://www.deksoftware.com/new/dsc.exe

It will most likely be the solution you seek. They have info at

http://www.deksoftware.com/dsc/index.html

 

[bcastner]

softwarehero, that looks like a nice product.

My issue is similar to SCFROMDC, where the drive was SMART monitored, and rolled-over dead one day without any apparent rhyme or reason. (The drive controller board failed, likely heat related).

I was using another good product, Active SMART:

http://www.ariolic.com/activesmart/

The other comment I had was that these software products do not explain the paramaters. Many are completely non-intuitive. And they are decidedly non-standard between drive manufacturers. An average value read of 100 when the threshold is 0, can be good for one value, and disastrous for another.

And the reporting can be an onerous chore. The DEK software example above shows a 70-page report for 6 drives!
(I know you can do exception handling, but still).

JiminKS asked if any of this help. I hate to say this, as the idea is a great one, but it has not helped me in predicting failures in practice.

 

[JimInKS]

Thanks for the responses.
This basically confirms my own limited experience.
And, since I can't find any drive manufacturer that really promotes SMART, I would guess that it is of very little benefit.

 

[jurgen36]

Smart can be a dubious help. It also slows the drives down some small amount. The best way is to use raid 0+1 arrays, this way you have reasonable high speed and reliability. Or you can go overbord and use Raid 10 with 9 or more drives. Regards

Jurgen