Sign in to follow this  
TheJack

Strange problem with my Maxtor

Recommended Posts

Since two 60GXP 'Deathstars' died on me, I bought myself the new 40 gig D740X Maxtor drive (6L040J2), a couple of months ago.

It worked fine for a while, but then something strange happened.

I was watching a DivX movie (from HDD, not cd-rom), but suddenly the movie stopped and closed. I tried again and it closed again at the same place (in the movie). OK, something is wrong with the movie (not uncommon). So I went to my friend, who has the same copy of the exact movie, and his copy worked fine (on his computer).

I ran Powermax utility (re-certification-something) and to my horror it found mistakes on the drive! It (the utility) also told me, that it can try to correct mistakes with I-don't-know-what (I was in shock, so I wasn't paying attention). I had to type YES with capital letters, and in a few seconds it was done.

All the files were OK, except that movie, which had anomalies at the place the movie stopped before. But it didn't close anymore.

I ran Powermax again the same day (re-certification), and everything was OK. I even run the burn-in test for a couple of hours.

That was a week ago. I ran re-certification test almost every day since then (paranoia ;)), and it found no mistakes. Everything seems to be OK now.

My question is: What was wrong with my drive at the time? Can it happen again?

The drive is cooled, it runs at 30-35 degrees C.

P.S.: Sorry about bad English. It's not my native tangue.

Share this post


Link to post
Share on other sites

Did the diagnostic program erase the drive's contents? If so, it was probably a bad sector and has now been put out of use. In this case this is most probably the last you'll see of it.

Share this post


Link to post
Share on other sites

It sounds like PowerMax erased just the bad/corrupted sectors on your drive. Which would happen to mean erasing parts of the movie that were stored on defective parts of the disk.

What happened is that some sectors on your disk, for whatever reason, were somehow corrupted to the point that the drive couldn't read data from them. Erasing that data allows the drive to check whether those sectors just contain corrupt data (say, if the drive was affected by one of those OS shutdown bugs) or if that part of the media itself is bad and needs to be avoided. (This remapping would have happened while PowerMax was erasing these sectors.)

Speaking of OS shutdown bugs... I think you forgot to tell us which operating system you're running. (The middle of a movie file sounds like a strange place for it to happen, if it's the OS's fault and not the drive's fault, but I don't think we can rule out the possibility at this point.)

The easiest/cheapest way to fix the file is to obtain another copy from another source (whether it's a CD you already have, a CD burned by your friend, another download, ...).

Share this post


Link to post
Share on other sites

First of all, thank you for your answers!

Do you mean, there are bad sectors on the disk? I checked the HDD with ScanDisk (I just typed 'chkdsk'' in the Command Prompt) and it didn't found any bad sectors.

Erasing that data allows the drive to check whether those sectors just contain corrupt data (say, if the drive was affected by one of those OS shutdown bugs) or if that part of the media itself is bad and needs to be avoided. (This remapping would have happened while PowerMax was erasing these sectors.) 

This is really the main thing I want to know. Was it just a ''software thing'', or is there a physical error (=bad sectors) on the drive? How can I check my HDD for bad sectors?

This would be easy, if I would be using just Windows 98. But it's a little more complicated than that. I have two OS's on the drive: Win98 (10 GB) and WinXP (30 GB), which is on NTFS. I partitioned my drive with Partition Magic 7 and for booting I am using System Commander. I was watching movie in WinXP.

I told already, that I checked my HDD (just the WinXP partition - there was the error) with 'chkdsk' in Command Prompt and it showed no bad sectors. Is that reliable? How can I know for sure, that there are no bad sectors on the drive?

I have already obtained another copy. That's not really the problem. The problem is, whether my brand new HDD contains bad sectors and/or has any other mistakes.

Share this post


Link to post
Share on other sites
Do you mean, there are bad sectors on the disk? I checked the HDD with ScanDisk (I just typed 'chkdsk'' in the Command Prompt) and it didn't found any bad sectors.
PowerMax remapped them for you; spare sectors have most likely been pulled in to replace the bad ones, so chkdsk won't see any bad ones anymore.
This is really the main thing I want to know. Was it just a ''software thing'', or is there a physical error (=bad sectors) on the drive? How can I check my HDD for bad sectors?

This would be easy, if I would be using just Windows 98. But it's a little more complicated than that. I have two OS's on the drive: Win98 (10 GB) and WinXP (30 GB), which is on NTFS. I partitioned my drive with Partition Magic 7 and for booting I am using System Commander. I was watching movie in WinXP.

I told already, that I checked my HDD (just the WinXP partition - there was the error) with 'chkdsk' in Command Prompt and it showed no bad sectors. Is that reliable? How can I know for sure, that there are no bad sectors on the drive?

I doubt it's software at this point. I think the sectors really went bad.

chkdsk showing no bad sectors is reliable, in the sense that it's confirming that the drive is hiding all the bad sectors from the computer. (That's what the drive's supposed to do. No drive is perfect, but the drive should present an illusion of perfection.)

I have already obtained another copy. That's not really the problem. The problem is, whether my brand new HDD contains bad sectors and/or has any other mistakes.

I'll respond to this part of the post at a later point in time (maybe next week).

Share this post


Link to post
Share on other sites

Thank you again, BarryKNathan! This is the second time you've helped me (the first time being the problem with my PlexWriter).

Even though this answer isn't exactly what I wanted to hear... Well, there's nothing I can do.

But still...How can I ever fully trust my HDD, if it can just destroy my data for no apparent reason? How can I be sure, that this won't happen again next week, next month...? :(

Anyway, I hope you'll find time for your response you've mentioned at the end of your post.

Share this post


Link to post
Share on other sites
But still...How can I ever fully trust my HDD, if it can just destroy my data for no apparent reason? How can I be sure, that this won't happen again next week, next month...? :( 

Never trust a mechanical device to store your data. Backup, backup, backup...

Share this post


Link to post
Share on other sites

Pradeep:I know that, of course...

But on the other hand, except for my current computer and the ''fancy'' 7200rpm drives, I have never had ANY problems with hard drives. Never ever. I've never made backups. I did't have to. The 120 MB drive in my 386 still works. The drive in my 486 also. Not to mention Cyrix with 3.2 GB Seagate HDD. Still works at my friend's. My Celeron comp with 6.4 GB HDD still works in my father's office. No problems with any of these drives. Not a single bad sector in all those years.

And now... this is the third drive in my current (Athlon) computer, that is making problems. It's fast, of course. The IBMs were, too. It's quiet (so were IBMs). For example, you can hear Seagate 3.2 GB drive across the appartment. So, there's a progress. But something's missing here.... I want to be able (as I was before) to trust my HDD with my data. To upload the data on the disk, and still find it there the next day. Intact. Is that too much to ask from the HDD manufacturers? Obviously.

I apologize for being so melodramatic. But I just hate it, when manufacturers of storage products (I see that with burners also) put speed before reliability.

Share this post


Link to post
Share on other sites
I apologize for being so melodramatic. But I just hate it, when manufacturers of storage products (I see that with burners also) put speed before reliability.

Samsung puts reliability before speed. And most 5400rpm IDE drives(except for Seagate U series) outlast 7200rpm IDE drives. SCSI drives will last as long as, if not longer than 5400rpm IDE drives.

Samsung for IDE reliability and...

Seagate for SCSI reliability :D .

Share this post


Link to post
Share on other sites
I apologize for being so melodramatic. But I just hate it, when manufacturers of storage products (I see that with burners also) put speed before reliability.

Samsung puts reliability before speed. And most 5400rpm IDE drives(except for Seagate U series) outlast 7200rpm IDE drives.

I agree so far.
SCSI drives will last as long as, if not longer than 5400rpm IDE drives.

In the last year and a half I've obtained (either bought or received via RMA) five SCSI hard drives and six IDE hard drives. In that time four of the SCSI drives and two of the IDE drives died. What were you saying about SCSI reliability again?

(This isn't an isolated experience for me. I've dealt with several 7200RPM Seagate SCSI drives, none of which have lived as long as my 7200RPM Seagate IDE drives. This brings us back to the "hard drive manufacturers have good times and bad times, good models and bad models" issue, which in my experience is enough of a factor to render the "SCSI is more reliable than IDE" idea useless when you actually need to buy drives.)

Samsung for IDE reliability and...

Seagate for SCSI reliability :D .

I was about to say that I agreed, but then I remembered that Quaxtor's SCSI drives have been more reliable for me, over the years, than Seagate's SCSI drives. These days I don't think you can go wrong with either Seagate or Maxtor for SCSI drives, though.

I'll heartily second the Samsung recommendation as well. (If I had to choose between a recent Seagate SCSI drive and a similarly recent Samsung IDE for reliability, I'd go with the Samsung IDE, although I admit it's a pretty close contest.)

Share this post


Link to post
Share on other sites
The problem is, whether my brand new HDD contains bad sectors and/or has any other mistakes.

The thing about modern hard drives is they all have bad sectors, they're just remapped at the factory. I used to look at the information for my SCSI drives with one of Adaptec's tools and without exception every one had some defects. If this was my drive I'd probably check it once a month for a few months and then forget about it. My guess is that it was an errant defect caused during shipping and only showed itself when data was written over it. This is why I almost always run a surface scan on any new hard drives I get with the utility for that drive.

Share this post


Link to post
Share on other sites
The problem is, whether my brand new HDD contains bad sectors and/or has any other mistakes.

The thing about modern hard drives is they all have bad sectors, they're just remapped at the factory. I used to look at the information for my SCSI drives with one of Adaptec's tools and without exception every one had some defects. If this was my drive I'd probably check it once a month for a few months and then forget about it. My guess is that it was an errant defect caused during shipping and only showed itself when data was written over it. This is why I almost always run a surface scan on any new hard drives I get with the utility for that drive.

now by bad sectors, you mean that the platter got hit by the read/write and made a hole which you can't recover from such area of the drive or that it's a little mistake made by the Drive.

Share this post


Link to post
Share on other sites
now by bad sectors, you mean that the platter got hit by the read/write and made a hole which you can't recover from such area of the drive or that it's a little mistake made by the Drive.
What I mean is no hard drive is perfect. When we're dealing with the capacities we have today (or even 5 years ago) it's almost impossible to produce a hard drive platter with absolutly zero defects. There's always going to be little microscopic areas on the platter that didn't get enough magnetic coating or some other defect that needs to be mapped around. As far as bad sectors developing later, any number of things can cause that. Maybe the crate the drive was shipped in was dropped a little too hard or it passed through a stray magnetic field that screwed up some of the low level formatting or the drive was bumped during operation causing the heads to nick the platter. Sometimes the drive can fix these problems itself but usually you have to run a utility to do it (like PowerMAX).

Check out these pages from the reference section, they explain it pretty well.

http://www.storagereview.com/guide2000/ref...geom/error.html

http://www.storagereview.com/guide2000/ref...rorMapping.html

Share this post


Link to post
Share on other sites
Sometimes the drive can fix these problems itself but usually you have to run a utility to do it (like PowerMAX).

Actually, most of the time the drive can do it without a special utility. Usually the drive will notice that it had to re-read a sector too many times or the like, and it will automatically remap it. It can't automatically remap a fully unreadable sector, unless the OS decides to write over it anyway, because remapping that kind of sector would destroy any chance that some utilities (see below) have of recovering the data.

As I see it, drive utilities for "fixing" this kind of problem fall into three main categories:

(a) Re-read the data and try to do better error correction than the drive alone is capable of. Spinrite http://www.grc.com/ is one such program.

(B) Re-read the data over and over until the drive manages to read the data or until there seems to be no hope. I suppose Scandisk's surface scan falls into this category.

© Make the drive erase and test any bad sectors, with no serious attempt at data recovery. This is what IBM DFT and Maxtor PowerMax do.

If you want to be confident in your drive, one thing to try is "scrubbing" the disk with various data patterns that (in theory) should be hard for the drive to read back. The concept is that if the sector is borderline defective, it might be able to store normal data for a while, but the sector scrubbing patterns should cause an immediate failure and trigger the sector to be remapped before it goes bad with normal data. The most comprehensive program I've seen for this is SpinRite, but newer versions of the "badblocks" program for Linux also have rudimentary scrubbing capabilities. (I think Larry McVoy (the guy who created the BitKeeper soruce code management system) also has a program (or set of programs) for doing this under Unix/Linux, if you don't mind having to come up with your own scrubbing patterns, but I'm not sure off the top of my head where this program (or set of programs) can be obtained.)

I actually have a drive utility of my own, but I haven't made a public release yet. I'm tempted to explain how my utility works, but I think I'll just let it speak for itself when I release it later this month. (TheJack, could you tell me what kind of IDE controller or chipset your Athlon system has, so that I can make sure that my utility supports it?)

Share this post


Link to post
Share on other sites
And now... this is the third drive in my current (Athlon) computer, that is making problems.

If I understand you correctly, you've had three hard drives in this system, and you've had problems with all three? A few suggestions:

Make sure the drive is being properly cooled, and try replacing the power supply in your system (it might be defective or otherwise incapable of properly powering both an Athlon and a 7200RPM hard drive at the same time).

Assuming your power supply and cooling are good, you may want to look at a two-disk IDE RAID-1 setup -- if one disk loses data, the other should be able to provide it still.

Share this post


Link to post
Share on other sites

Well, I certainly learned a lot today. Thank you all very much!

I do not believe, that there is a problem with my system. The drive is properly cooled (with http://www.sspring.com.tw/shdc-a.htm), and otherwise the system works fine. Besides that, the hard disks were making different problems. IBMs (that weren't cooled) 60GXP after a few months both started to make loud noise while spinning (at first I thought, that something's got caught in the proc fan :)) and I just took them back (I had them one after another, of course). There is a different problem with Maxtor (see topic).

The price for another hard disk and a raid controller is a little steep and I can't afford that at a present time unfortunately.

I have ECS K7S5A board, which has SiS735 chipset on it. The IDE controller is SiS530, at least PowerMax says so. Where will your program be released (which site)?

After reading your posts, it seems to me, that this thing is not as bad as I thought it was (I'm a little paranoid, as I said before ;)). So, I'm just gonna do, what Steel suggests: I'll check the drive regularly for a few months, and if everything goes well, I'll just forget about it.

Share this post


Link to post
Share on other sites
Where will your program be released (which site)?

I'll almost certainly post the URL in a new thread in this forum when I release it.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this