Chkdsk on a recovered drive not fixing an error

[Pavisakiwi]

I had a hard disk start failing on a Windows 10 (1903) machine so I cloned it before it died completely by using ddrescue from a Live Linux rescue ISO to a brand new hard disk and I got a 99.95% recovery after 17 hours of operation. Windows booted fine on the new hard disk and so far, nothing stands out as missing or having suffered from any losses.

On researching ddrescue, I found it has a habit of copying everything, warts and all, including bad block records to the new drive and last time I did this, I only had to run chkdsk C: /b to have the bad blocks record cleared and the disk space used by C: fully scanned to recheck the surface for bad blocks. On reboot, the Windows log file clearly showed that no bad blocks existed on the partition post scan.

However, the issue I'm seeing moving forward is that the log also showed under the stage 1 phase a swag of entries about a USA Check value being incorrect with no indication that they were rectified, and the same list is persistently there every time I manually run a chkdsk /r

The machine works just peachy and the Windows logs show nothing out of the ordinary.

Here's a sampling of the chkdsk output.
I get 32 entries at Stage 1:Examining basic file system structure ... and the same list repeated again at Stage 2: Examining file name linkage ...

Checking file system on C:
The type of the file system is NTFS.


A disk check has been scheduled.
Windows will now check the disk.                         

Stage 1: Examining basic file system structure ...
The USA check value, 0x0, at block 0x1 is incorrect.
The expected value is 0x5.
The USA check value, 0x6361, at block 0x1 is incorrect.
The expected value is 0x5.
The USA check value, 0x70, at block 0x1 is incorrect.
The expected value is 0x3.
The USA check value, 0x0, at block 0x1 is incorrect.
The expected value is 0xb.
The USA check value, 0x0, at block 0x1 is incorrect.
The expected value is 0x3.
The USA check value, 0x0, at block 0x1 is incorrect.
The expected value is 0xe.

Can anybody tell me what these are, if they can be fixed, or should I just ignore them?

Thank you in advance

 

[Digerati]

I only had to run chkdsk C: /b to have the bad blocks record cleared and the disk space used by C: fully scanned to recheck the surface for bad blocks.

First, a minor point just to clear things up. /b (which implies /r) does not clear the bad blocks record, unless the block is found to be good. If you run chkdsk /? you see it "Re-evaluates bad clusters on the volume".

As for this problem, Google sure isn't helping. There have been a couple other users in the past who saw these errors, but there is no common solution. BTW, they also did not experience other problems - that is, the drive appeared to work fine otherwise.

That said, unfixed errors on a drive would bother me too.

You might try running SeaTools to see if that helps.

If not, try running from an elevated command prompt, sfc /scannow.

 

[britechguy]

Another potential option is re-cloning the drive using a utility other than ddrescue, particularly if you have access to another Windows machine.

I haven't used a Windows-based cloning utility in years that either attempts to fix, or ignores if it cannot fix, bad blocks/sectors during a clone. Those error messages may be a side effect of having used ddrescue.

If you do happen to go this route, please report back on what you find after having done so.

I also doubt that the System File Checker (SFC) will have any effect, as the error class that I see in that log is not something that SFC would have any effect on.

 

[Pavisakiwi]

Thanks to both. After cloning the damaged drive, I used Acronis True Image to make a second copy of the clone onto another good, but well used drive as a secondary backup in case something went screwy along the way, That drive has not been touched so I might drop it in and see how it pans out.

Acronis, or at least the version I have does not work well with damaged drives, hence my choice to use ddrescue.

I downloaded seatools bootable from the link above, what a mental system they have, the downloaded zip is named SyncExplorer.zip, but its contents are that of Seatools !

Both drives were 500GB in size, but the measured usage is small enough I could re-clone to a smaller drive. I have a spare 256GB Samsung Evo around, so I'll try cloning to that using Acronis and see if the shift away from sectors to pages will have any effect, but I'm out of time for the day, got to go to work. Will report back my findings

 

[britechguy]

Check to see if Samsung actually has a "clone to our SSD" utility first. There are a number of SSD manufacturers who do, and using the purpose dedicated utility is best practice whenever possible.

 

[Pavisakiwi]

This drive didn't come with any additional tools, don't know why, it just didn't and it is not eligible for any either. As for using a manufacturers utility, that's good advice for noobs, but I don't buy that "clone to our SSD" utility first because drive manufacturers usually just bundle a light version from a vendor that sells fully licensed products and usually, the fully licensed products have more features and abilities than the lite versions, so my preference is to use the real thing, not the stripped down versions provided through hardware manufacturers. I have a few fully licensed cloning tools loaded onto my Ezi2boot USB flash drive and my preference is the Parted Magic ISO which can clone almost anything, in fact, I believe every tech needs to have this in their toolbox.

Anyhow, cloning to the Samsung SSD seemed to do the trick. Post cloning, I cleared all logs and ran the chkdsk utility, watched it do its stuff over a reboot so I can confirm it did the check and after rebooting, upon checking the logs, there was nothing, which is typical of when chkdsk has nothing to do. I'm gonna play with it a bit more to see if any further breakages happen, then clone it back to a spinning platter disk and recheck.

I'll report back again when I'm done because if the move from a CHS type storage media to a SSD which uses pages with emulation ends up being the ultimate solution to this nuisance for techies, them we'll have all learnt something new and be all the wiser for it. Can't say when but I'll aim to do it within the next few days or so.