Windows sees RAW, chkdsk sees NTFS

Topic

New Reply

As the title says, I have a partition that shows up as RAW in Disk Management, but chkdsk recognizes it as NTFS and runs repairs on it just fine.
However, it remains RAW in Disk Management even after multiple chkdsk fixes. There are 3 other partitions on the same drive that are fine.
TestDisk also recognizes the NTFS partition just fine and I tried writing a new partition table to the drive with TestDisk, with no change.
I’m looking to re-write the MBR at this point but I figured I’d ask if anyone had any ideas.
This is a secondary drive, this is not a boot problem.
It’s MBR-formatted, not GPT.

I haven’t been able to find anyone else describing an issue like this. This happened to me fairly recently, when I was trying to run some RAM OC for the first time. I have since returned to stock and run plenty of system checks and everything seems to be in order. I’ve run both DISM and SFC.

This is the most similar thing I came across even though it’s clearly different:
https://community.spiceworks.com/topic/2299706-windows-10-drive-converted-to-raw-format

Reply | Quote

Replies

https://www.diskpart.com/articles/healthy-primary-partition-raw-5740.html

Desktop mobo Asus TUF X299 Mark 1, CPU: Intel Core i7-7820X Skylake-X 8-Core 3.6 GHz, RAM: 32GB, GPU: Nvidia GTX 1050 Ti 4GB. Display: Four 27" 1080p screens 2 over 2 quad.

Reply | Quote

If it doesn’t contain any data, delete the partition with Disk Management and recreate it.

cheers, Paul

Reply | Quote

It did contain data, though not important data.
I would of course re-format if need-be but since most of it seemed to to be intact, I wanted to try to recover it.
Some messages I received while running chkdsk with /r or /b options:

Windows replaced bad clusters in file
An unspecified error occurred (75736e6a726e6c2e 500).
Adding bad clusters to the Bad Clusters File.
Correcting errors in the master file table’s (MFT) BITMAP attribute.
Correcting errors in the Volume Bitmap.

I ended up running chkdsk /r on a loop using a batch file for a few days until it consistently ran without finding any errors to fix. I also attempted to backup the partition using Macrium Reflect, and tried checking the partition using GParted. Pretty much everything had failed in some way or another and I was preparing to run GNU ddrescue or RecuperaBit just to see if I could move the contents, format the partition, and then move them back to see if that made things better, before running bootrec to fix the MBR.
In any case, today it finally showed up as NTFS in disk management. I have no idea which step ended up finally getting Windows to recognize it, but as I said earlier, chkdsk saw it just fine, as did all the other applications I ran on it so to me it was never a case so dire that I should have to delete the partition entirely.
I think it’s likely that re-writing the partition table with Testdisk solved it, I maybe just needed to restart the system to see it take effect.
… and you know I say that but as soon as I unmounted the partition it went right back to showing up as RAW. How very interesting. I mount it, and it shows up as NTFS again.
That’s special. Guess I’mma continue with backing it up and trying to fix it fully.

Reply | Quote

Cormy1 wrote:

Windows replaced bad clusters in file

Your disk is about to die and needs to be replaced ASAP.
Make a backup NOW.

Modern hard disks have internal error checking and correcting and never report bad sectors unless the ECC has failed / run out of spare sectors.

Run a SMART utility to see where the errors are occurring.
CrystalDiskInfo
HWinfo

cheers, Paul

Reply | Quote

Neither of those tools would indicate where the errors are occurring, SMART data isn’t that smart, you just get a count of various statistics. chkdsk on the other hand does tell me where it is finding bad clusters, relative the filesystem.
Bad clusters appear when data is failed to be read. There is nothing to correct if you cannot read the data. This is an error that occurs before ECC can take place.
The disk is not about to die, it is dying, as all disks are. It’s just further along the process.
The disk is over 10 years old, I’m well aware it is suffering bit rot.
As I said, there is nothing important in there, I just wanted to know how to fix the partition table, as that is what Windows references in Disk Management to populate the partition data, whereas chkdsk and other applications will actually analyze the data structure to identify the partition type.

Wouldn’t be the first time I’ve run into and fixed a similar kind of corruption, spinning rust doesn’t like being dropped while powered on very much (laptops). I was just looking for more input regarding what I saw as a fairly unique scenario I couldn’t find mentioned anywhere else online.

Reply | Quote

Cormy1 wrote:

Neither of those tools would indicate where the errors are occurring, SMART data isn’t that smart, you just get a count of various statistics.

SpinRite will tell you, but it’s not free, and it’s not cheap, either.

Always create a fresh drive image before making system changes/Windows updates; you may need to start over!

We all have our own reasons for doing the things that we do with our systems; we don't need anyone's approval, and we don't all have to do the same things.

We were all once "Average Users".

Computer Specs

Reply | Quote

ddrescue is free and also tells you, I just hadn’t gotten to the point of running it at the time, I’m also unsure of what the point of finding where the errors are occurring even was in the first place, it’s a strange comment to make. I could cross-reference with TestDisk just to map out which partitions contain the bad sectors and go further to figure out which files are corrupt but the repair steps would still end up being the same, and that is backing up the data regardless, so that it can be moved off of bad sectors.

Reply | Quote

Cormy1 wrote:

I’m also unsure of what the point of finding where the errors are occurring even was in the first place

Not “where on disk”, “where in the disk sub-system”. SMART will show you where it is going wrong.

Stablebit Scanner (not free) claim to be able to show which files are / maybe damaged.

cheers, Paul

Reply | Quote

Cormy1 wrote:

Neither of those tools would indicate where the errors are occurring

True, but given that disk is on its way out at a rate of knots, I would spend a small amount of money on a new one.

cheers, Paul

Reply | Quote

The data of a RAW drive can be saved with Portable Roadkil’s Raw Copy (Windows 8 version runs on Windows 10).

This program copys a disk as a raw image from one drive directly to another. This utility is designed for people who have faulty drive and want to transfer the data directly to another drive without doing a file by file copy. This saves the need for operating system re-installs and allows drives with an unknown file system to be copied (including from console game machines, data recorders, mac etc). The program has a built in data recovery function which will attempt to recover data from bad sectors to ensure all the available data is restored from the drive. This program is designed to run under NT/XP/2000 or later operating systems. It will run under windows 95/98/Me operating systems but only windows logical drives can be copied.

Reply | Quote

It is 2009 vintage and the last forum post is 2016. Have you actually tried this on a W10/11 system?

cheers, Paul

Reply | Quote

I downloaded the Windows 8 portable version and checked on my Windows 10 . It run fine (I don’t have a RAW partition).

Reply | Quote

HDDGURU looks to be a better bet, and it’s freeware.
Or use the clone feature of your backup software.

cheers, Paul

Reply | Quote

Don’t think there’s any reason to look beyond GNU ddrescue but I guess if you don’t want to touch Linux, it’s a cool option.
I have a live boot disk and have used ddrescue before so for me it’s no problem. Just boot, install the package through the manager and run.
Currently have less than 300kb of bad sectors that it couldn’t read after several passes. To me it’s completely insignificant on a 1TB drive that is less than 50% utilized and doesn’t have important data on it anyway.

Anyway the drive did end up showing as NTFS formatted eventually, may have just needed to reboot the system for some changes to take.

Reply | Quote