Safe Speed Forums

Anybody know how SMART HDD monitoring actually works??

What does it measure and how does it intrpret the results?

D

_________________
"The road to a police state is paved with public safety legislation"

I found this link on Google if it helps:-

http://www.passmark.com/products/diskcheckup.htm

_________________
Useless laws weaken necessary laws.

Do you mean S.M.A.R.T (which you can turn on or off in BIOS) or SMART (which would be software)

Not all disks are S.M.A.R.T enabled, though most new ones are. It's usually disabled by default (never come across a motherboard with it set enabled by default) and just monitors hdd performance over time. I've never turned it on for clients I've built machines for as it delivers a very slight hit in performance and isn't always accurate anyway.

I've never known what it keeps an eye on, always assumed bad sectors, rpm, load times etc. So had a quick look around now. Apparently it varies per manufacturer, but the common ones are

* Head Flying Height: A downward trend in flying height will often presage a head crash.
* Number of Remapped Sectors: If the drive is remapping many sectors due to internally-detected errors, this can mean the drive is starting to go.
* ECC Use and Error Counts: The number of errors encountered by the drive, even if corrected internally, often signal problems developing with the drive. The trend is in some cases more important than the actual count.
* Spin-Up Time: Changes in spin-up time can reflect problems with the spindle motor.
* Temperature: Increases in drive temperature often signal spindle motor problems.
* Data Throughput: Reduction in the transfer rate of the drive can signal various internal problems.


--

Some people report it as working (reporting a problem resulting in them backing up, only for drive to then fail as predicted a few days later) - others report it as incorrect in it's reporting of bad drives which then never break.

Thanks both!

Dixies link is a neat piece of software that told me exactly why I was suddenly getting warnings at the boot screen

Sector remap limit exceeded IYSWIM Its a new disk too!

There was some sort of crash the other night whilst my machine was unattended. When I got up it had restarted booting from a slave disk that had an old operating system on it (resourcefull, if you can say that for a machine!) and when I restarted it I got this SMART warning at boot up telling me that disk failure was iminent.

though I wasnt able to boot from the suspect disk, I was able to acccess it and repair it using a disk utility and was then able to transfer the data to a new disk which now works fine (What a buisiness! there has got to be an easier way of doing this!).

I then reformated the original disk to see if this would "reset" the SMART monitoring (IYSWIM) However even after reformating the old disk I am still getting the SMART warning at the boot screen so I guess the fault alert is written into some sort of perminent memory on the disk drive circuit board.

Oh well, I shall have to see if I can find the paperwork and get my money back but since it is PC-World I wil not hold my breath!

D

(PS I find myself, at this point, thinking about 2001 and psychotic computers and their unreliable fault prediction systems )

_________________
"The road to a police state is paved with public safety legislation"

Dusty , if still in warranty - and they get shirty , try them with a"a report to head office and Trading standards " -had something similar happen with a CD drive - girl could not be bothered till I did that --next thing "yes sir -it is in warranty ,pick a new one from over there "

_________________
lets bring sanity back to speed limits.
Drivers are like donkeys -they respond best to a carrot, not a stick .Road safety experts are like Asses - best kept covered up ,or sat on

Reformatting wouldn't solve a mechnical problem which would be why you would get an error from S.M.A.R.T.

You can turn off those warnings in the BIOS (hold Del when you first turn computer on to get to BIOS menu) - however if it's reporting errors, it should be covered by store warranty or manufacturer warranty.

Yes. There is some instrumentation in the disc that measures certain properties as the device works. If a threshold is exceeded, you get the error report. It gives you a chance to save data to another drive.

The errors can occur in two circumstances. First, there could be a transgression over the limit set. Second, the instrumentation itself could be badly calibrated or otherwise non-functional. In the latter case, you can ignore the problem by turning it off in th BIOS, but there is no way to tell the difference, so that is not safe to do.

In short, the disk is knackered. Take it back and get a new one. Choose a different manufacturer - no use getting stung twice by the same wasp.

I've had two drive failures in recent years - the first was one of the infamous IBM Deathstars, which I did replace with an alternative make (I've still got the warranty replacement drive sealed in its anti-static bag) because no-one in their right mind would have put another Deathstar into their system back then. The second, more recently, was a WD Caviar, the warranty replacement for which was still going strong a couple of years later up to the point where the entire PC was retired.

So unless the fault is caused by a problem with a large batch of components, or a fundamental flaw in the drive design - and news of either will spread like wildfire across the net these days (even back in the Deathstar days when the net was still a bit of a novelty for most people), so it's usually not difficult to find out if the problem is more than just a one-off bad drive - then I think you're just as likely to run into problems with a new drive from a different manufacturer as you are in replacing like for like. Of course, if you're like me, you'll use the failure as an excuse to buy something bigger/faster in the meantime whilst you're waiting for the warranty replacement to arrive

_________________
Chris

Statistically yes, but I'm philosophical about these things. The message we send to makers when we avoid their products is "you're not getting my business, chum". It's far better for them to try and hear it and react, and firms like IBM do exactly that.

I recently used this http://www.z-a-recovery.com/ to recover the data from a hard disk that my wife almost killed by trying to hot swap it The drive still showed some signs of life but would not boot. ZAR recovered all the data

If you order at the weekend they have a special discount and with the current exchange rate only costs a tenner.

Fair enough when the problem is of a scale where the company ought to have picked up on it during development, and I've steered clear of IBM/Hitachi drives ever since then as a result, even though I know their reliability is now on a par with everyone else.

On the other hand, if it's just a random failure which could have happened regardless of who's name was on the label, then the message you're sending is more like "you're not getting any more of my business because I now know your products aren't 100% reliable, and so I'd rather deal with one of your rivals because, although they also make products which aren't 100% reliable either, I don't know that yet"...

_________________
Chris