• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Can't stop NTFS corruption!

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Silver_Pharaoh

Likes the big ones n00b Member
Joined
Sep 7, 2013
I'm having a serious issue with corruption here, and I can't stop it.
Chkdsk /f always comes back clean when it does the boot scan. (The first time it ran though, it came back with lots of issues which it said it fixed.)

Interestingly enough though, the boot scan will fail with "Autochk cannot lock the disk for direct access due to a recently installed software package" -_- I haven't installed anything i didn't have before I switched to AMD. :sly:
Also, the OC in question is a RAID0 on the built in AMD motherboard RAID using 2 WD green drives. (WD ERZX Sata 3 and WD EARS Sata 2, both 1Tb) I cannot see the WD ERZX drive in HD Sentinel... I was able to see the drive before with Intel RAID on the Intel mobo I was running.

I first noticed this when I was overclocking and when I was setting the FSB to 250. It wasn't quite stable at 250 yet until I upped the CPU/NB volts. It did freeze few times until I upped the volts. I've now set the system back to stock, leaving the Cool 'n' quiet and virtualization + all that other crap off. I also still have core 5 and 6 disabled. (Core 5 is weak and want more Vcore to be rock stable so I disabled it)

Previously I was running 4.7 Ghz. 1.568 Vcore, 1.15 NB volts, 1.300 CPU/NB v, 1.65 RAM volts, NB was @ 2300 Mhz, HT link @ 2700 Mhz, PLL (VDDA) was set to 2.8 Volts.

Temps were warm yes, but I was still prime stable for 8+ hours blend.

Since Chkdsk is always coming back clean, would it be safe to ignore this?
Or is one of the drives dying?
Any way to safely check the drives without destroying the RAID array? (IDE mode then run Hiren's DOS tools to check the SMART data on both drives?)

Some pics:
The latter being brand new (The AHCI warning one) - never seen this until I checked before I typed this post.
ntfs.PNG
ahcix64.PNG


EDIT: If it also matters, the RAM passes memtest without issue. I loosened the RAM as I am unable to run it any lower - not even at XMP settings 9-9-9-24 @ 1600. last time I did that memtest failed. Perhaps I can try again since I'm stock now.
 
Last edited:
S_P

I looked but don't see which O/S you are using. Win XP or Win 7.

WD10EARS 1TB Sata 2 << that drive was first built in 2007 and was said to be a 'transitional' drive moving to 4K formatting and not 512. That drive has some caveats the other one does not have in special attention must be paid to partition alignment. Not so critical for the WD Green WD10EZRX 1TB IntelliPower.

RGone...
 
Will read up on these. Thanks!
I looked but don't see which O/S you are using. Win XP or Win 7.

WD10EARS 1TB Sata 2 << that drive was first built in 2007 and was said to be a 'transitional' drive moving to 4K formatting and not 512. That drive has some caveats the other one does not have in special attention must be paid to partition alignment. Not so critical for the WD Green WD10EZRX 1TB IntelliPower.

RGone...

Win 7 x64.

Both drives are from 2011. The WD10EARS was from the prebuilt I bought back in 2011. the WD10ERZX was bought separately for the purpose of RAID.

The ERZX drive did at one point have one reallocated sector and one offline uncorrectable error according to SMART. But that was almost 2 years ago and has been fine since.

The WD10EARS was also a bit funny when I was overclocking my 3570K. Probably was just a bad overclock though, HD Sentinel reported that the EARS drive failed catastrophically, with huge spin up times and many many errors.
That all disappeared when I re booted and the drive was fine. Last time I checked, the SMART was fine.


EDIT: Also I was mistaken, I can see the WD10EARS in HS Sentinel, it's the WD10ERZX that is missing from HD Sentinel.
 
try a new sata cable?

as for checking the drives in raid i use HDD guardian it lets me see each drive in my array great smart info and all :)
 
Installed the "recommended" AMD SATA driver from HD Sentinel. I now have both drives showing up. But the RAID Array prevents getting all the SMART data from the ERZX, it only shows a general SMART health. Which is good btw. Drive has no SMART errors that I can tell.
try a new sata cable?

as for checking the drives in raid i use HDD guardian it lets me see each drive in my array great smart info and all :)

Will try HDD guardian. As for the SATA... I'm all out of those. I will admit, they are cramped in my case, I didn't make the board stand off the case much so I had to jam the cables in real tight.... Next case I make (soon) will have higher mobo stand offs.
 
In light of RGone's PM, I am going to attempt to destroy the RAID 0 Array and switch to AHCI while leaving my OS intact..

With luck, I'll be okay and thing will go smoothly.


Anyways, the issue still persists, I received those NTFS errors for 14 seconds straight at 12 AM last night...
Either I have bad Sata cables, bad drives or AMD's drivers for RAID are garbage. Time to find out I suppose.
 
if i remember right, amds raid drivers suck, Rgone played old scratch getting it to work.
 
if i remember right, amds raid drivers suck, Rgone played old scratch getting it to work.

I know he had issues getting the drivers to load to install Winderz...

I found an article that points to VSS causing the NTFS issues...
Basically, flush the VSS cache, disable it then re-enable it and voila. I'm going to try that next. If it fails, I'm going to destroy the Array, and re-image one of the disks while leaving the BIOS in RAID mode. If I'm reading correctly, if there is no array, the disks will act as AHCI drives since RAID is based from AHCI.
 
Okay here's something to chew on:
Searching from the start menu by typing in "c" or "m"(without quotation marks) results in those ntfs errors.

Reproducible every time. :confused:
wtf? I have disabled windows search service (which is what the start menu search uses right?). It was the first thing I do on fresh installs...


EDIT: So r,l,c,m all trigger the ntfs errors...
I'm beginning to think this is just a case of Windows being Windows :sly:
 
Last edited:
The ERZX drive did at one point have one reallocated sector and one offline uncorrectable error according to SMART.

Bad drive. You know it already but refuse to accept it. ;)
 
Bad drive. You know it already but refuse to accept it. ;)

That's cause it hasn't seen another one yet.... ;)
And I was super cheap back then lol

Perhaps it's time to retire these greens anyway.....
Maybe save them for a server one day?


I know my brother wants a big fat drive. The WD ERZX might be for him! :muahaha:
WD Blacks, Purples or Reds... I heard the blacks also had that stupid "green" spin down feature like the greens have?
 
Get what data you can from the drive before it's too late. If you are using RAID 0, better grab the data before you loose it all.
 
Combining posts to avoid the dreaded triple post :chair:
Post#1
--------
Even 1 reallocated sector makes it a bad drive.
Gah. you're right... besides, these "green" drives are DOG slow :p
Get what data you can from the drive before it's too late. If you are using RAID 0, better grab the data before you loose it all.

Imaged it yesterday. Only issue at hand is re-imaging after I break the array...
If I can shrink the partition to under 1 TB, I can just image it, then break the array and re-image one disk with it and all should be good. (The array is 2Tb in size right now......)

Trouble is, I can't shrink the partition in Windows because of the NTFS errors. It stops and says "The disk might be corruption run chkdsk first then try again" :mad:

Post#2
--------
Was finally able to shrink the partition to under 1 Tb!

Will image the WD10EARS tomorrow then hack the registry to accept AHCI mode.
If AHCI mode doesn't stop this corruption, then it's something to do with Windows :(

New post #3
--------------
Okay, now I was able to re-image the drive onto the WD10EARS.
Partition is aligned properly, and the offset is divisible by 4096.

All drive report good SMART health, and the ERZX drive has no re-allocated or pending sectors anymore :shrug:
BIOS is in AHCI mode now and I'm still getting the NTFS errors in the event logs :mad:

I don't get it. The OS drive is healthy, chkdsk /f still finds no errors and sfc /scannow finds no errors.
Either Windows is totally borked or it's just being stupid.

I'm going to run a chkdsk /r tonight.
Anything else I can try guys?
 
Not much you can do. I learned my lesson a long time ago about using RAID 0. When you can afford it, I would buy an SSD for your OS, and software, and have a spinner for storage. I'd would use RAID for a NAS, but I would not use RAID 0.
 
Not much you can do. I learned my lesson a long time ago about using RAID 0. When you can afford it, I would buy an SSD for your OS, and software, and have a spinner for storage. I'd would use RAID for a NAS, but I would not use RAID 0.
I wanted to do RAID 5, but I didn't have enough drives....

So when I grab an SSD, would you recommend a fresh install of Windows or re-image it again?
I guess in the mean time, I'll keep an eye on the partition. Run chkdsk once a week or something.

Oh and Chkdsk /r came back clean :-/
No bad sectors..
 
I wanted to do RAID 5, but I didn't have enough drives....

So when I grab an SSD, would you recommend a fresh install of Windows or re-image it again?
I guess in the mean time, I'll keep an eye on the partition. Run chkdsk once a week or something.

Oh and Chkdsk /r came back clean :-/
No bad sectors..

If your system has been crashing I'd do a fresh install. But you will loose the array, soon.
 
Back