Random HD errors, even after a chkdsk? What the heck?

[MDesigner]

I swear, I have a thread posted for every single component of this f***king computer.. I'm seriously about to just take the whole rig and toss it out a 10 story building.

So now the hard drives are giving me a hard time. I JUST did a "chkdsk /f" on the D drive. It fixed stuff. I ran it again.. it was clean. I rebooted the machine.. now I ran chkdsk /f again on the D drive, and now it says:

Correcting errors in the Master File Table (MFT) mirror.
Correcting errors in the Volume Bitmap.
Windows found problems with the file system.
Run CHKDSK with the /F (fix) option to correct these.

What the HELL?!?!!? I just ran a chkdsk which corrected any problems! Same with the C drive, I JUST did a chkdsk /F, rebooted..it cleaned stuff up.. and now I boot into windows & run a "chkdsk" on C... it says there are probs w/ the file system, and it needs to correct it.

These are two SATA drives. The motherboard is an Abit AN8 32x SLI. In the BIOS, the SATA controller is on (of course) and the SATA mode is set to "IDE". Any bad settings here? I'll note something odd, that Seagate's "SeaTools" diagnostics showed the two drives TWICE. One in a section called "Other Drives" and one in a section I think called "BIOS Drives." If I ran a disk check on the 160GB in one of those sections, it would lock up the machine at a certain point. But if I ran the check on the 160GB in the other section, it tested perfectly. What the hell?

What the HELL is going on? I suspect it's the motherboard.. and I really get the feeling I should just take this Abit piece of crap and smash it, and go back to Asus. These components (the RAM, the HDs, etc) all came from a previous machine (Soltek motherboard) and they all worked flawlessly on that machine. Suddenly, I transplant the parts to an Abit m/b with a dual core Athlon64, and things are all haywire.

HELP!!!

 

[Nocturnal]

Hard drive may be going bad. Try running the diagnostics software for that particular brand HDD.

 

[Suspicious-Teach8788]

Get a SMART reader like Speedfan or ACtivesmart, or HD Inspector and tell me if any of your readings look whacked. If so, uh oh.

Plus SeaTools is giving you problems... Abit is fine. If anything, I would blame the HD. 7200.9s are still too new to show signs of weakness, but if oyu look at 7200.8 which is almost the same drive, those were DAMN UNRELIABLE (StorageReview database). I've had a 7200.8 fail on me as well.

 

[Spikey217]

It's rather funny that you post this. I've been going through the SAME EXACT PROBLEM. It's rather frustrating for myself as well.

In my situation, I've narrowed it down to the mainboard. Memtests runs fine. I've tried a fresh install of Windows XP Pro on 4 different hard drives (2 of them brand new) and they all have the same disk error. Ran WD drive tools, no problems, ran Seagate tools, no problems. Ran a DVDInfoPro CRC check on my Windows cdrom, no problems. Already set up a RMA.

The funny part is that my mainboard is an Asus, and I was thinking about going back to Abit.

Asus P5VD1-X w/P D805

I know my post is no help, but good luck with your problem.

 

[MDesigner]

I can't connect to my home machine right now to verify, but I'm pretty sure I have 7200.9 drives. I'll check when I get home.

Spikey: hmm.. that motherboard has an Intel chipset. I figured maybe we both had nForce4 in common. Guess not.

I'll see if I can get one of those SMART readers tonight and run some diagnostics. I'll post the results here ASAP. Between Active Smart & HD Inspector, which would you recommend?

 

[SparkyJJO]

I have Everest that I use to check SMART readings and video card temps (CCC only shows GPU temp, Everest shows GPU, VRM, and GPU ambient).

Check your data cable too - those can be the cause of numerous problems. My one comp would sometimes tell me that there was this system file missing (NTDLR I think) and it wouldn't boot. BIOS knew the drive was there though. I just would bump the cable and it would work fine and once it was booted up it was fine. But reboot and it could act up again. Changed the cable, all was well.

 

[MDesigner]

One quick thing.. I tried out the Seagate SeaTools, and I think it said SMART was disabled, or maybe not supported..not sure. It was greyed out. But my drives are both pretty new.. Seagate SATA drives, NCQ, etc. So I can't imagine the drives not having SMART. Do I have to enable it somehow?

 

[MDesigner]

My boot drive is a 160GB, ST3160827AS (7200.7) and my secondary drive is a 300GB, ST3300831AS (7200.8). I'm not really sure what the .7 and .8 mean though. So 7200.8 is bad? Bad, as in, worth selling this thing on eBay and getting a new one?? Or just use SMART monitoring 24/7 to keep an eye on it?

 

[MDesigner]

OK, Active SMART says my HDs are doing just fine. These are the full stats on each drive:

160GB primary drive: http://www.samhulick.com/misc/160gb.txt
300GB 2ndary drive: http://www.samhulick.com/misc/300gb.txt

Do those numbers look OK?

 

[MDesigner]

Grr, my computer keeps locking up..event log reports error ID 51:

An error was detected on device \Device\Harddisk0\D during a paging operation.

I think it's time to back up, and reformat, immediately. This ship is going down. I'm almost afraid to back up & reformat though.. who's to say the backup files won't be corrupt too? I feel like just buying a new 160GB HD and installing XP on that, and then putting in the possibly bad 160GB as a secondary drive and copying the files that way..

Damn sometimes I hate computers.. I really do.

 

[Spikey217]

At first I thought that my problem was with my hard drive too. On my very first installation of Windows, I wouldn't have thought to run chkdsk to see if there were hard disk errors. I mean, it's a fresh install, what can go wrong?

It was until the random symptoms of lockups, CRC errors, steam updates to games already fully updated (eg: Half-Life Episode One started updating itself, and upon checking the update log, there were no updates to the initial release, therefore I realized that some of the files on the hard drive must be missing or damaged. Now that doesn't happen unless something was wrong with your hard disk). So after the symptoms, I ran chkdsk in the Windows repair console. I thought it was fixed but once I booted into Windows and ran chkdsk again, the errors were there again.

I jumped ship too early and blamed the hard drive. I didn't have another empty hard drive to install Windows on so I did an advance RMA on my raptor, something which I regret now since my relacement is no only recertified but more importantly carries a crappy three month warranty. When I received my replacement, I promptly installed Windows on it and upon the first boot into Windows, I ran chkdsk only to find the same errors as before.

So I google searched and found a mysterious thread from the VIAArena forums claiming instablility/incompatability issues with Western Digital hard drives and the specific VIA chipset on my mobo. Well, it didn't sound like it was true, but at that point, I was inclined to believe anything to get the problem fixed. So I purchased a Seagate 160GB 7200.9 SATA drive. Brand new retail box. I put it into my machine. Nothing should go wrong now right? Boy was I wrong. I install Windows on my brand new hard drive and upon bootup I run chkdsk. WHAM, same errors as with my Raptor.

So what does it boil down to? Bad RAM? Bad CPU? Bad mobo?

I ran Memtest and there were no errors.

So I thought it was a bad SATA chip on my mobo. So I found an old IDE Maxtor and did the usual. Reformat, install windows, run chkdsk. WHAM, same errors.

I'm at a dead end now. So I've decided to RMA my mobo and see if that is the culprit because I have run out of ideas. If my replacement mobo comes back with the same errors, I'll just use your idea... toss everything off the roof of my house.

Good luck to you getting your problem fixed.

 

[MDesigner]

This PC you had the issues with wasn't an Athlon dual core CPU by any chance, was it?

 

[Spikey217]

Asus P5VD1-X w/ Pentium D 805

Dual core, but Intel. With VIAPT880 Ultra chip.

I'd suggest getting another hard drive to install Windows and running chkdsk then. If it fails, you know for sure that you can't blame your hard drive.

Different chipsets but similar problem. The only item we have in common is the Seagate HD and the EVGA 7900GT card.

 

[MDesigner]

Well, we do have dual core in common. I already ordered a new Seagate HD (7200.9) and will have it probably Friday. My plan is to install a fresh copy of XP on that drive, and see how that works out. If I still get lock-ups... then I'm afraid I'm completely out of ideas. This is depressing. I've built tons of PCs in my life and never had an issue I wasn't able to resolve.

 

[rasczak]

if you still have lockups, you either have

A: a heating issue

or

B: power issue

or

C: driver issue.

the driver issue should be taken care of after you reformat. So give us some stats on your processor and GPU heat.

 

[MDesigner]

A: No heating probs. CPU is at 33C, "SYS" is at 35C, "PWM" is at 42C.. not sure what PWM is. CPU voltage is normal.. at 1.37V currently, fluctuates between 1.36V and 1.38V. Hard drives are also running fine, in the mid 40s (Cel). GeForce 7900GT peaks at around 66C when under heavy load (courtesy of rthdribl), and the CPU temp jumps up to around 40C. At idle, GeForce temp hovers a little below 50C.

B: Don't know how to troubleshoot that one. I have an Enermax Liberty 500W. It powers the computer, that's about all I can verify. If you tell me where to buy a tool to test my PSU, and how to work it, I'll definitely buy one without hesitation, as troubleshooting power issues has always been an impossibility for me.

C: It's POSSIBLE, but I highly doubt it. We'll see after the reformat. The computer works perfectly until I start copying or moving mass amounts of data around.

 

[Auric]

Use the manufacturer's utility or chkdsk to check for bad sectors. /f switch is just for the fixing the file system so if there are bad sectors which have not been found and marked unusable there will be repeated file system errors.

run: chkdsk /f /r

 

[MDesigner]

What the heck?? OK, this is interesting. I have two hard drives. I took an 83MB movie file, and copied it from C: to D:. Then I ran a CRC32 and an MD5 on both files.. the CRC & MD5 didn't match!! It basically got corrupted on its way over to the second hard drive. I delete the copy on D:, copy it from C: to D: again, and it has a totally different CRC this time.

Another test, copying files WITHIN the C: drive (not leaving the partition or drive). I made a duplicate of a file. The CRC matched fine. I made a copy of that copy.. and THAT duplicate's CRC didn't match the original two.

For all I know, all my files could be corrupt. This is bad, very bad. I get the feeling I should just shut down the computer and not even touch it till I can pull out these hard drives and put them in a SATA->USB enclosure and plug the drive into a RELIABLE computer. At the very least, I don't think I should be re-saving or copying any files..in other words, I shouldn't be opening files (for example) in Word, and working on them, and saving them.

What the **HELL** is going on here?!?!? ARGH!!

PS: I can copy small files around all day long, they are perfectly fine. It seems that only larger files (at least 100MB or so) are the ones that don't copy over properly.

 

[Nocturnal]

Have you run Memtest on your memory yet? Let it do a few passes and let us know if you received any errors.

 

[MDesigner]

A week or so ago I ran it for 8 hrs.. no errors. I'll run it again tonight.

Another interesting fact: I was wrong about copying small files all day & no prob. The smaller it is, the more times I need to re-copy it before I get a corruption. A 14MB file took me about 6 tries to get a bad copy. But a 300MB file mis-copies immediately on the first try. The odd thing is, that 14MB file was a DivX video clip.. and I watched the corrupted copy. No glitches of any kind. Odd.

This is frustrating, but at least I'm getting closer to a solution.

 

[MDesigner]

Whoa, major discovery. I copied a 360MB file onto my 1GB USB flash drive. Then I duplicated the file right on the USB drive without taking it to either of the hard drives. The result: the dupe file had a different CRC than the original. Since the USB drive is not an IDE device nor a SATA device, this rules out the nForce IDE/SATA drivers. It may not rule out the nForce4 chipset though.. but this sure seems like a RAM issue now. I'll do that RAM test and report in tomorrow.

 

[Stas]

Sounds like the cache memory on that hard drive is glitching.

 

[Auric]

Yeah, seems like system memory is the likely culprit. I would make sure it is plugged in the preferred slots securely sans dust and the timings and voltage etc. are correct then run Memtest86+ and Windows Memory Diagnostic.

 

[MDesigner]

I ran memtest86 overnight.. at 8 hrs running time and 13 passes, there were no errors. I'm at work and I left it running.. so by the time I get home, should be around 26 passes or so. Then I'll see how it looks.

What if memtest86 reports no errors? Then what's the next step? Prime95 in torture test mode? Blend mode? On Friday I'll have a new 160GB HD, so I can install WinXP on it, and copy a 300+ MB file to it (after reinstalling XP) and seeing if the same prob still occurs.

One step at a time.....

PS: The one mystery thing here is that the USB flash drive corrupted a file on duplication too. This rules out any hard drive specific problems, I think. But if the RAM tests OK, then what else would cause this? Could the CPU be glitching?

 

[SparkyJJO]

Could be CPU. I had one get hit by a surge from a lightning strike - mobo was a goner but the CPU seemed to work OK at the moment. This was back on my old Pentium 1 133MHz system. Strange thing was, I couldn't for the life of me get windows installed. I tried a different HD, different RAM, data cables, CD drive... nothing. I was stuck using DOS. Frustrated, I decided to play some Descent, and while playing the game it glitched up on me and instead of displaying graphics it put all the hexadecimal coding on the screen, then froze up. CPU was bad. I replaced it with a 100MHz chip, which I overclocked later to 133MHz

Another thing it might be is the mobo. If it is screwed up nothing will work right.