Q: Need advice on hard drive/optical drive testing software

SMART technology is about as good as a thermometer is for detecting illness. WIth a thermometer, if you have a fever, it indicates you're sick, but how sick? Do you have a simple cold of flu, or do you have Ebola? If you have no fever, does it mean you're not sick at all, when in fact you may have cancer?

I reference the following link for more info on SMART:

http://en.wikipedia.org/wiki/S.M.A.R.T.

In particular, the following quote is interesting:

"Conversely, little correlation was found for increased temperature and no correlation for usage level. However, the research showed that a large proportion (56%) of the failed drives failed without recording any count in the "four strong S.M.A.R.T. warnings" identified as scan errors, reallocation count, offline reallocation and probational count. Further, 36% of drives failed without recording any S.M.A.R.T. error at all, except the temperature, meaning that S.M.A.R.T. data alone was of limited usefulness in anticipating failures"

Just exactly how good is a technology that's vendor dependent and may or may not indicate if a real problem exists? I've seen people toss perfectly good drives because SMART technology was reporting trivial errors that they didn't understand, and other drives just drop completely dead with no SMART indicators showing anything. I've also seen some drives last for years after SMART tests predicted an "about to fail any minute" status.

Apple doesn't seem to give SMART reporting much importance, and I suspect it's for good reason.

I would take SMART technology as something to be aware of but not necessarily base my life on. It's not abnormal for a drive to drop a block a few times during its life, and it's actually normal for SSDs to do so.

Comparing SMART to a thermometer is a good analogy. One of the problems it has is that once it marks a drive as failing the settings are there to stay. For example, if a head crash occurred and it wasn't severe enough to wipe out all the spare sectors, the drive can be zeroed out and put back into use. It might work fine for the rest of a normal life, or it may continue to get worse and eventually die.

I have a Seagate drive in an old system that crashed probably about five years ago. All the SMART reports have reported that it's about to die, but none of the settings have changed in years. I don't use it for anything critical, and with drives being as cheap as they are now it's probably safer to replace the drive. The system is an old "junker" I use in the lab and it just keeps running. It has an incredible number of power on hours and I"m amazed it hasn't died from something else. When the system or drive die, I'll just scrap the whole thing. It's an old iBook and I use it to send e-mail out of the lab, and do graphs etc. In this case I'm glad I didn't replace it because it would have been a waste of money, and not scrapping the system saved us money.

Those descriptions and that link on the Google report do not sound promising!

You can get a kernel extension for SMART for external drives, provided the housings themselves support it. I think it's called SAT-SMART Driver, or something like that (google it!). The only problem is it's freeware, I don't now how often it's updated, and I think any time the OS does an update you might need to re-install it or get a new one.

Just thought I'd throw that out there for anyone interested.

ThomasB2010 wrote:

 

I've noticed there isn't a whole lot of mention about SMART in this thread. Is it really that unreliable?

It's not very reliable. People run out and buy these tools like DriveDX and SMART Utility, which are actually pay-for front ends to smartctl (which is free) which then detects *something* so they run down the halls screaming "My drive's about to fail, my drive's about to fail!" without knowing what they're actually seeing or doing and then replace the drive. Probably not really the fault of  smartctl, which is the free binary in smartmontools doing the actual work for stuff like Drive DX or SMART Utility, but rather the users themselves not understanding what they're being told.

Thank you all for clarifying.

Treat SMART as an indicator and nothing more, and understand what it's reporting to you. If it reports some change, understand what it is and if needed take it to someone that can explain it to you. It's not total garbage, it just ins't that  great/reliable.

ZV137 wrote:

 

I reference the following link for more info on SMART:

http://en.wikipedia.org/wiki/S.M.A.R.T.

 

In particular, the following quote is interesting:

 

"Conversely, little correlation was found for increased temperature and no correlation for usage level. However, the research showed that a large proportion (56%) of the failed drives failed without recording any count in the "four strong S.M.A.R.T. warnings" identified as scan errors, reallocation count, offline reallocation and probational count. Further, 36% of drives failed without recording any S.M.A.R.T. error at all, except the temperature, meaning that S.M.A.R.T. data alone was of limited usefulness in anticipating failures"

 

Just exactly how good is a technology that's vendor dependent and may or may not indicate if a real problem exists?

An interesting study might be how long drives continue to run after some of these "SMART" tools report that the world is about to end. I know of no such studies.

If you do a Google on that topic you'll get responses all over the place from, as you put it, "run down the halls screaming 'My drive's about to fail, my drive's about to fail!'" to "keep it but keep an eye on it" which makes much better sense to me.

I had an old drive that some of those "smart" utilities told me was going to crash and burn. It lasted about 3 years more and finally got pulled when I needed more space. I suppose it still works unless a drive sitting on a shelf can rot.

I have an old Maxtor external drive that I got in 2008. At one point it was used for a boot drive on a Mac Mini. It still seems to work great. I have to wonder if I did a SMART status check on it what it would say? Throw it out immediately, the world is about to end?

Sector relocations on a hard drive aren't that uncommon or unusual, and they're downright normal on SSDs unless rampantly growing. SMART testing software just scares people, IMHO.

The only ways you could test that old Maxtor's SMART status would be to remove the drive from the housing and make a direct connect to another unit and get the readings or use the SAT Smart driver package and hope the drive supports it. To use the external SMART drivers the drive itself has to be SAT supporting enclosure. My big WD backup drive shows up with it but some others don't so it's a crap shoot.

Those drivers seem to work but they need to be re-installed often after an update. Also, you have to disable kext checking in Yosemite. I have no idea how or if they'll work on El Capitan, or how they could be installed, if possible.