Oddly disk errors

[Peewee]

Heyho.

Bought a 250Gig SATA drive for buttons on Amazon. Connected it up and installed 7. Things started going haywire so I checked it with Seatools and others. It FAILED.

I used KillDisk to write zeroes, and then formatted it and tested it again. PASSED.

I installed XP. It went wonky again. Tested it. FAILED.

Did the above again and installed 7. It went bad again. FAILED.

All the tests used were useless in terms of information; I just got "FAILED" or "PASSED". The only really helpful information came from one of the progs which offered, "An error occured while attempting to repair bad sectors" - but that's not really much, is

it?

So, the drive clearly has bad sectors which are evidently irrepairable. It's a start. The drive has bad sectors which only come to light when the drive is actually used. The FORMAT process, I believe, doesn't perform any READ tests, and so I understand why it will return no errors. But chkdsk also always reports "0 bytes in Bad Sectors"

Many years ago when hard disk technology was still relatively young, a small percentage of Bad Sectors was deemed acceptable provided they were due to manufacturing defects in the platter surface and were not due to damage caused by platter/head collision. At that time many programs (mostly DOS) would simply mark those bad sectors in the File Allocation Table (FAT) as bad, and then the system simply never again bothered to use those sectors, and just got on with it, usually (always, in truth) without subsequent problems.

Thing is, the current batch of Hard Disk "Utilities" are, frankly, pretty useless. Not only do they give no useful information and not a single one of them - Seagate, Maxtor, Fujitsu - will even format a drive (DUH!), but none offers a way to mark bad sectors as bad in the FAT (NTFS), and so I go around in a circle. All this problem might need is a program that marks the bad sectors as being bad in the FAT, and maybe the drive will be useable - unless it is damaged, as stated earlier.

So, does anybody know of a decent hard disk utility that will just note the bad sectors, mark them as such in the FAT, and let me use this disk? It was a doddle 20 years ago, how come it's now impossible?

I must add that I have NEVER had a single IDE drive die on me in all the years I've been using them; I just dissed them as I moved on up in size. But 50% of all SATAs I have owned have died young deaths (2 out of 4). Newer, younger, clearly doesn't equate with

better.

Also, don't suggest returning the drive - it was bought for me by someone who I now wish to never have to see ever again - I would sooner throw the drive in the bin.

Thanks.

[jaclaz]

peewee

you are seemingly missing the main point

If the manufacturer utility says that the drive has failed, it has failed.

Now you can try to "revive" it by using a few programs, but unless they are trivial (fixable) errors, that disk drive is deemed ultimately to the dustbin.

The effect of the format command is DIFFERENT in XP and in Vista (and later) OS:

You may want to try it with a "third party utility" such as Victoria for Windows, and/or invest a few bucks in HDD regenerator, but that's in practice everything you can make.

The way DOS marked bad sectors as bad is still effective, but generally speaking never has a chance to "kick in" as the drive firmware will prolly re-map the defective sector in a way that is transparent to the OS and filesystem utilities before they can detect it.

If there is one (or a few) biggish chunk(s) of bad sectors, simply partition the drive in such a way that the bunch is outside the partitioned/accessible space, but there is NO guarantee whatsoever that the "bad zone" won't "expand" over time.

Always think that a 250 Gb drive has more or less 25 times the capacity of a (largish) DOS era drive in the same physical space, this basically calls for a 25x precision in *everything* , no actual surprise that it has more chances of developing a defect.

jaclaz

[Peewee]

Jaclaz, many thanks for your reply.

Been offline for a while, hence the tardy response.

Some of this reply concerns human interaction, as well as technical stuff.

You suggest that I am "seemingly missing the main point". Such a statement is semantically wrong.

I cannot miss any point if I made the point in the first place. I am not missing even "A" point, never mind "THE" point or "THE MAIN" point. I made the main point, and so the main point is mine. And so I cannot miss the main point. Logic.

But, to explain - again... The actual MAIN point is not what the test programs return, instead that, in the past, bad sectors were simply marked bad in the FAT, and the drive always went on to succeed - something I can no longer find a way to do, hence the post. The main point.

Further, if the manufacturer's utility says that the drive has "failed", but subsequently offers "passed" for the same drive, then the utility is obviously flawed, as I said, is evidently useless, and, importantly, the drive's "failed" status is then questionable; as indeed is its "passed" status also. Meaning the drive may not be beyond recovery.

Fail, pass, fail, pass. Which is it? Surely one should expect a disk utility from a disk manufacturer to be consistent, and actually know what is what? They don't know. And therefore should we not question everything that the program produces\returns\does? I absolutely do. All those I tried continually and sequentially offered conflicting summations, and so they were\are useless. Could, might, possibly, not sure. Useless.

You offer that there do exist trivial, fixable errors on drives. Well, that's where I want to go. Clearly, I didn't make myself clear, clearly.

My use of Format was purely an example - for illustration, to suggest why Format does not; perhaps cannot, suggest a drive is a write-off. There has to be both read and write processes, and many of each, before a judgement can be made. Norton Utilities was particularly useful in this. Sadly, that no longer exists, and Norton's name is no longer held in high esteem. Perhaps they could revive their past status by producing a new set of Utilities? It is plain that someone ought to fill this evident gap.

You offered,

"The way DOS marked bad sectors as bad is still effective, but generally speaking never has a chance to "kick in" as the drive firmware will prolly re-map the defective sector in a way that is transparent to the OS and filesystem utilities before they can detect it." Perfect. I was asking if there was a program which could do this. You agree it is possible. I'm glad you agree that something (the firmware) is stopping good practice. All I want is for good practice to be restored, and find a program which offers such. It must be asked, though, why the firmware would remap, and ignore important information.

As for partitioning outside of bad sectors, and thereby excluding bad sectors, my point about lack of useful information, surely, explains why this cannot be done. If any of the programs gave me detailed information concerning geography of bad sectors, I might have a chance to use this approach, but they don't, and so I can't. How can I create a partition that excludes bad sectors when I do not know which sectors are involved? I can't, obviously. Poor programming.

There is no guarantee that ANY drive won't just fail now, tomorrow, in ten years time. However, as I stated, in the past, when some percentage was deemed acceptable, and provided it was due to manufacturing processes supplying surface defects, and not caused by physical damage, everything worked fine until I dissed them to get a bigger drive.

No, I'm talking about 10, 20, 40, 80, 160, 250MB drives, way, way smaller than 250GB. Regardless, size always follows technology - not the other way around. Improvements in technology enable larger disks. Everything progresses; except rap, boy bands, TV, and jean's with a long inside leg. Manufacturing methods (must) have improved, obviously. So we are here talking about technology, microtechnology, and engineering, not percentages. Since I had those tiny drives, we've experienced two and a half inch floppies, CD, Minidisc, DVD, and BlueRay - a veritable ton of surface technology advancement. I expect better because I know that better exists now. By any calculation a percentage failure now, would, at the very very worst, be an equall percentage failure as in the past; though if that were the case, given improvements in technology, I would shout, "Shame!" No, sorry, technology has moved on, and that suggestion is specious.

But thank you for your reply. I will look up the programs you mention.

[jaclaz]

Jaclaz, many thanks for your reply.

Been offline for a while, hence the tardy response.

Some of this reply concerns human interaction, as well as technical stuff.

You suggest that I am "seemingly missing the main point". Such a statement is semantically wrong.

I cannot miss any point if I made the point in the first place. I am not missing even "A" point, never mind "THE" point or "THE MAIN" point. I made the main point, and so the main point is mine. And so I cannot miss the main point. Logic.

I meant the main point about proper ways to deal with the issue you were having. And rest assured, you missed it.

You are evidently trying to extend your past experience to modern technology.

It simply won't work, because whilst some things actually remained the same, and thus for these past experience is very useful, for some other things, that have changed, past experience is not only unuseful, but - to a certain degree - counterproductive, in the sense that it may make you believe things that were correct, but that are not correct anymore.

In other words, though it is perfectly possible that you are right when you say that some specific manufacturer tools suck (and a few actually do suck big ) you cannot state - given the limited experience that you seem to have with modern hard disks and with modern programs for them - that "since this used to be done like this then it must be done in the same way", as said some things have changed very little, and some have greatly changed.

Think about cars, 10, 15 or 20 years ago, with a medium sized toolbox you could fix *any* car, nowadays on many cars you need a specific interface, a specific connector, a speciifc program (besides a computer) even to change a light bulb.

As for partitioning outside of bad sectors, and thereby excluding bad sectors, my point about lack of useful information, surely, explains why this cannot be done. If any of the programs gave me detailed information concerning geography of bad sectors, I might have a chance to use this approach, but they don't, and so I can't. How can I create a partition that excludes bad sectors when I do not know which sectors are involved? I can't, obviously. Poor programming.

Or maybe more simply out of the paradigm of the specific programs you used, among others the suggested Victoria will give you this kind of information.

There is no guarantee that ANY drive won't just fail now, tomorrow, in ten years time. However, as I stated, in the past, when some percentage was deemed acceptable, and provided it was due to manufacturing processes supplying surface defects, and not caused by physical damage, everything worked fine until I dissed them to get a bigger drive.

No, I'm talking about 10, 20, 40, 80, 160, 250MB drives, way, way smaller than 250GB. Regardless, size always follows technology - not the other way around. Improvements in technology enable larger disks. Everything progresses; except rap, boy bands, TV, and jean's with a long inside leg. Manufacturing methods (must) have improved, obviously. So we are here talking about technology, microtechnology, and engineering, not percentages. Since I had those tiny drives, we've experienced two and a half inch floppies, CD, Minidisc, DVD, and BlueRay - a veritable ton of surface technology advancement. I expect better because I know that better exists now. By any calculation a percentage failure now, would, at the very very worst, be an equall percentage failure as in the past; though if that were the case, given improvements in technology, I would shout, "Shame!" No, sorry, technology has moved on, and that suggestion is specious.

Better (in the sense of faster/larger) may not be the same as better (in the sense of more reliable/failproof).

Modern hard disks are way more complex than older ones and - as said - there are some factors that may make them more likely to fail.

Just as an example, perpendicular recording:

http://en.wikipedia.org/wiki/Perpendicular_recording

creates a number of engineering problems in manufacturing, precision and ultimately in reliability/life of the disk.

I also expect that in a few years HAMR:

http://en.wikipedia.org/wiki/Heat-assisted_magnetic_recording

will pose more and new challenges to the engineers.

But thank you for your reply. I will look up the programs you mention.

You are welcome.

jaclaz