Dud NF4 Controller?

[FLD]

Deal all,

 

few days ago, all of a sudden, all the folders in my PATA HDD, which I used to store my data, disappeared when I browsed through the HDD using Window Explorer. But I found that in Window Explorer, although it was all blank, the there were XX GBs used, which means all the folders were still there but just could not be seen by Windows.

 

I left the PC on over night just to find that, in the next morning, my PC went into DOS screen as if it had rebooted during the time but could not get into Windows as the boot HDD was now not bootable. The boot HDD was a SATA II HDD.

 

I have had this kind of data corruption once or twice before.

 

I don't remember when I had the first few data corruption the PC was OCed but on this last occasion the details are as below:

 

250X10 the ram sticks were running at 250 2-2-2-5 @ 3.5V and blown by 2 96cpm 120mm fans. The CPU was running at 1.375 + 113% and cooled by TT Tsunami + 106cpm 120mm fan. The temp on a hot day was never more than 37c full load.

 

Before this last corruption, it had been OCed for weeks 24/7. It had also been tested using many of OC stability software. The mobo is a DFI NF4 SLI-DR, which, I think, has the SATA and PATA ports locked. The SATA II HDD was connected to the NF4 SATA Port 1, while the PATA was connected to a paralel port.

 

Some one pointed out that it could be a dud controller which could have caused the data corruption. If so, how would I know that the mobo is the culprit.

 

Although I get very good support here and replacement will be quite easy, some good convincing word may be needed to verify with the shop I got it from.

 

Thanks very much.

 

Best regard,

FLD.

[red930]

It could be one or more of many different things.

 

Start with the cables. Then use the drive manufacturers tools to verify the drives. If those check out OK, move the drives to different ports or controllers and test again.

[FLD]

I managed to connect the PATA drive (not the SATA drive, which is still connected to the first PC as a second drive and has data recovery software running on it) in the 2nd PC.

 

When the 2nd PC, which whose specs are shown in my sig, was turned on the HDD check (I am not sure I am calling it correctly) process was activated immediately and checking on the file system integrity or something. I think the Windows on this PC is trying to fix the HDD.

 

This process is still running on the second PC as of now.

 

Unlike when it was connected to the 1st PC, this process was not activated and it would get into Windows, where this HDD would show no files and folder in it but the sapce of XXX GBs was being used.

 

Would this imply something? I'll keep you guys updated when it's done.

[FLD]

Update 2:

 

After the Windows checkdisk completed, the PC went into Windows and I could see all the folders, which I am not sure if there are all but am quite happy enough to see them there.

 

I am going to copy all the folders on this HDD to another HDD and reformat this HDD. Is it OK to do so?

 

Please note that when this HDD was connected to the old PC I could not see any folders. In Windows Explorer, it was all blank while saying the space of XXX GBs was used.

 

What would this all mean?

 

Thanks a lot.

[FLD]

Following to my Update 2, I have found some new folders while some are not, which I am unsure what to do with them. Can you help me please?

 

What are folders, like #ROOT, found.000, MSOCache, RECYCLER, and System Volume Information

 

In #ROOT, it seems like there are duplicate folders as those in the main folder. Should I copy this #ROOT folder to a new HDD, too?

 

In found.000, there are hundreds of folders named dir0000.chk to dirXXXX.chk, which cointain files and folders, and files like file0000.chk to fileXXX.chk

 

What are these folder and files? What should I do with them?

 

What shopuld I do with the res 3, MSOCache, RECYCLER, and System Volume Information folders?

 

Thanks a great deal in advance.

[red930]

Well, I can't help you with the "found" folders. You'll have to look through them to see what you want to salvage.

 

But, the dir0000.chk to dirXXXX.chk, which cointain files and folders, and files like file0000.chk to fileXXX.chk are from the scandisk operation.

 

It looks like the File Allocation Table got hosed. The FAT is where the drive keeps a record of whats in each cluster.

 

When scandisk found data in clusters that weren't in the FAT, it saved them with the names you see.

 

If you don't have a backup of the data you're looking at a lot of data loss and hours of trying to put things back together again.

 

Good luck!

[FLD]

It looks like the File Allocation Table got hosed. The FAT is where the drive keeps a record of whats in each cluster.

 

When scandisk found data in clusters that weren't in the FAT, it saved them with the names you see.

 

If you don't have a backup of the data you're looking at a lot of data loss and hours of trying to put things back together again.

 

Good luck!

 

Hi ExRoadie. Thanks a lot for your helpful input. So, the FAT is damaged, you reckon?

 

Also, do you tink that the controller could have caused this corruption? I have had data corruption probably 2 -3 times already. But this time was the worst because it happened to both of the HDDs.

[red930]

While possible, it's very unlikely that the controller had any part in it. More likely the cable or the drive. In all my years I've only come across one drive controller that trashed drives. Cables and drives fail all the time.

Hi ExRoadie. Thanks a lot for your helpful input. So, the FAT is damaged, you reckon?

 

Also, do you tink that the controller could have caused this corruption? I have had data corruption probably 2 -3 times already. But this time was the worst because it happened to both of the HDDs.

[FLD]

While possible, it's very unlikely that the controller had any part in it. More likely the cable or the drive. In all my years I've only come across one drive controller that trashed drives. Cables and drives fail all the time.

 

 

The 2 HDDs tested OK using WD software in their comprehensive test.

 

How do I know it was the cable that caused the problem? And if so how could they do so?

 

I was using the yellow SATA and the rounded cables provided in the box. These two things are not sophisticated enough that, if faulty, they should have shown their defectiveness when used. Unlike in electronics, sometimes the fault can be intermittent while in some other times it can be very obvious if it's completely dead.

 

Thanks.

[red930]

It's possible that the connection at one end or the other was at fault or maybe a cable is bad but working now.

 

You can retest the cables by running the drive test software and move the cables around to see if any errors are generated.

 

If the drives and cables test OK with the WD software I would go ahead and say they are good.

The 2 HDDs tested OK using WD software in their comprehensive test.

 

How do I know it was the cable that caused the problem? And if so how could they do so?

 

I was using the yellow SATA and the rounded cables provided in the box. These two things are not sophisticated enough that, if faulty, they should have shown their defectiveness when used. Unlike in electronics, sometimes the fault can be intermittent while in some other times it can be very obvious if it's completely dead.

 

Thanks.

[FLD]

It's possible that the connection at one end or the other was at fault or maybe a cable is bad but working now.

 

You can retest the cables by running the drive test software and move the cables around to see if any errors are generated.

 

If the drives and cables test OK with the WD software I would go ahead and say they are good.

 

 

After the corruption, I tested the 2 HDDs using WD software's comprehensive test in DOS mode without moving anything. They both tested OK.

 

I have had this mobo for like 3 - 4 months and this was probably the 3rd time I had data corruption.

 

The previous 2 were on the SATA II HDD, which I was using as the boot disk. But this time it happened on the PATA HDD and then the SATA II HDD. After each corruption, I always tested the HDDs using WD software and there was no faults on them.

 

So, from what you suggested, I should be able to say that the HDDs are OK and were not the cause of data corruption.

 

Now, what should I do next?

 

EDIT:

 

I remember once I posted my OC results someone pointed out that the temps at full load at stock and OCed speed were too high especially when @ OCed speed. Do you think this could have caused the data corruption?

 

The the term high is subjective. Normally, what should be the temps on the NF4 chipset at fullload and no-load at stock speed and when OCed the maximum temp that it should never ewxceed?

[red930]

Temps are sometimes the cause for failure but honestly I've put my hand on the case of rigs that were burning hot and never had a failure.

 

Since you've had failures on both the SATA and PATA controller ports I'm gonna say it's highly unlikely that the controllers are at fault. If you truly feel they are to blame you can request and RMA based upon your results and see what happens.

 

Since there are so many variables involved I would definitely try a completely different configuration. You might try putting your rig together outside the case with some airflow across the components.

 

On a side note... I use Spinrite from www.grc.com to test and validate my drives, cables and controllers. I never install a drive in a production rig until it passes three full tests. Additionally, I run full tests on my production drives at least once every six months. You'd be surprised how much error correction is hidden from the user on modern drives.

 

On temps... I try to keep a max delta of 30F between room temp and the highest reported component temp. Normally the high performance video card is the hottest part in my rigs. But, I use cases with excellent airflow and move fans around to enhance the airflow. I will even install a "stirrer" fan just to move the air inside the case around to eliminate any stagnant air.

 

I hope this gives you some information to digest.

After the corruption, I tested the 2 HDDs using WD software's comprehensive test in DOS mode without moving anything. They both tested OK.

 

I have had this mobo for like 3 - 4 months and this was probably the 3rd time I had data corruption.

 

The previous 2 were on the SATA II HDD, which I was using as the boot disk. But this time it happened on the PATA HDD and then the SATA II HDD. After each corruption, I always tested the HDDs using WD software and there was no faults on them.

 

So, from what you suggested, I should be able to say that the HDDs are OK and were not the cause of data corruption.

 

Now, what should I do next?

 

EDIT:

 

I remember once I posted my OC results someone pointed out that the temps at full load at stock and OCed speed were too high especially when @ OCed speed. Do you think this could have caused the data corruption?

 

The the term high is subjective. Normally, what should be the temps on the NF4 chipset at fullload and no-load at stock speed and when OCed the maximum temp that it should never ewxceed?