2TB External HDD became 100% full after failed Chkdsk

[Kusuri]

Hello,

This is a very weird problem I've never seen before and no one seems to know what caused it or what's it about...
Some days ago I was running a diagnostic chkdsk /r on a USB 3.0 2TB external HDD NTFS that had about 30% used space or so (ran it because a program had interrupted the moving of files and I just wanted to make sure they were all ok). The drive is only 6 months old.

The chkdsk ran fine and I saw it get to Stage 5 of 5 (I think) which is the longest one that is looking at millions of free clusters. So I went away for a while.

When I came back all there was to greet me instead of chkdsk log results was this continually being spammed:
"Error in writing the output log"

I closed the command window and checked My Computer only to see that now instead of having 70% space free the disk had 0 bytes free of 2TB.

I went to check all files inside and they all seem intact. I checked the properties and they all only add up to around 600GB.
Also no hidden, system or recycle bin files that could explain the full space were found either.

2 programs were able to give some (although vague) information:

WinDirStat just says that, among all the normal files taking space, the disk also has <Unknown> taking up 1.4TB worth of space.

Defraggler shows no free blocks whatsoever (as expected). There are only the blue blocks from my expected files followed by lots of red/fragmented blocks each said to contain 1 file called $BadClus:&#8206;$Bad, Size: 1.4TB, Type Unknown, Location G:&#8206;\.

SMART Disk information says Health is GOOD and there are no values that report any kind of error, CRC error or the such.

Additionally:
Besides the chkdsk resulting in these free space woes every time I connect the drive the System log reports "The default transaction resource manager on volume G: encountered a non-retriable error and could not start. The data contains the error code.".

I am also unable to use the USB Safe-removal like I used to because it always says the disk is being used even if it isn't (I have to shut down PC to be 100% safe now).

I cannot reformat as I have no other drive with enough space to clone it, at least not yet... so I was looking for other options.

What I tried so far:

Ran chkdsk /b and I saw that between Stage 1 and 3 it reported finding and deleting all the "bad sectors" that had replaced all the free space on the first chkdsk, but then it goes back to "Error in writing the output log" at the very end of chkdsk. Drive is unchanged and still with 0 bytes free.

Ran Western Digital Data Lifeguard Diagnostics Extended Test. Results: PASS.

Does anyone have any idea of what happened here and if there's anything else I could do instead of formatting?
Thank you in advance.

 

[VirtualLarry]

I wouldn't mess with it any more than necessary, until you can obtain another drive to copy the contents that you can still access off of the drive.

 

[Burpo]

WD external drives encrypt the data, but standard utilities should work (including chkdsk). Something went awry. Western Digital's Data Lifeguard Diagnostics works on Windows systems with external USB, eSATA drives. It does a physical surface scan (ignoring format and file system) and will rewrite $badclus based on the scan's result. It even works with other manufacturers drives.

 

[Kusuri]

WD external drives encrypt the data, but standard utilities should work (including chkdsk). Something went awry. Western Digital's Data Lifeguard Diagnostics works on Windows systems with external USB, eSATA drives. It does a physical surface scan (ignoring format and file system) and will rewrite $badclus based on the scan's result. It even works with other manufacturers drives.

It is actually a Toshiba STOR.E, but I have WD Lifeguard Diagnostics from my other WD external.

Unfortunately the results have left me stumped. Both "Quick" and "Extended" tests passed with a green "PASS"...

The disk, however, is still the same: 0 bytes free of 2TB.

The only solution is to format?

 

[imagoon]

Pretty sure you need to run chkdsk /B to cause it to reevaluate bad clusters. If they pass they should get cleared.

 

[Kusuri]

Pretty sure you need to run chkdsk /B to cause it to reevaluate bad clusters. If they pass they should get cleared.

I did try that and it did seem to report clearing up bad clusters between stages 1 and 3.

The problem is that when it's about to finish stage 5 of 5 it just starts spamming "Error in writing the output log" again.
When I came back to Windows nothing had changed, still 0 free space.

 

[Kusuri]

Well it appears this is without a doubt a mysterious issue with chkdsk and HDDs. Everywhere I searched couldn't really find a solution and most seemed stumped.

Diagnostic tools from WD or HD Tune just pass and say everything is fine while chkdsk just screws up at the end and fills up all the free space to the last byte with unknown, invisible, corrupt data for no reason.

I'll have to try to get hold of an other 1TB HDD or so to be able to backup and format this problematic one and just remember to never run chkdsk on it again. Better use different diagnostic tools... Just hope it does indeed fix it...

 

[imagoon]

More than likely chkdsk uncovered an issue that it couldn't repair correctly that was already there rather than a chkdsk bug.. It happens most often on USB drives because most people fail to eject them properly resulting in silent corruption.

 

[Kusuri]

I've always taken care to safe remove any USB drives I have connected, this one included. It is weird however that all the data that is in the drive is intact and well.

Just all the free space that got filled up with garbage, invisible data. Never saw such thing happen before and there's not much to be found online...

Also don't know what it really means when it says "Error writing output log".
Just did chkdsk /l and it reports that the current size of the NTFS log size is 65536 KB and from what I gathered that is normal.

Hopefully should get a hold of a spare HDD to back up the data and reformat this one tomorrow.

 

[imagoon]

I've always taken care to safe remove any USB drives I have connected, this one included. It is weird however that all the data that is in the drive is intact and well.

Just all the free space that got filled up with garbage, invisible data. Never saw such thing happen before and there's not much to be found online...

Also don't know what it really means when it says "Error writing output log".
Just did chkdsk /l and it reports that the current size of the NTFS log size is 65536 KB and from what I gathered that is normal.

Hopefully should get a hold of a spare HDD to back up the data and reformat this one tomorrow.

File system errors were preventing writes to the log. If I had to guess, something is wrong with the MFT. Even if you files appear to be there, they maybe corrupt and fail to copy.

 

[Burpo]

Did you try deleting or moving some of the files off the drive & then run chkdsk? If there's no space to output the log, chkdsk /b can't finish.

 

[Kusuri]

File system errors were preventing writes to the log. If I had to guess, something is wrong with the MFT. Even if you files appear to be there, they maybe corrupt and fail to copy.

I should have mentioned I was indeed able to copy files off it. I copied the most important ones (about 40 GB worth) successfully. Still will copy off all others before trying anything else.

Did you try deleting or moving some of the files off the drive & then run chkdsk? If there's no space to output the log, chkdsk /b can't finish.

That is a good point and something I have yet to try. If something went wrong with the first desktop chkdsk due to some program interference causing the 100% full space then I suppose even the 2nd boot chkdsk will also always fail due to no space available. I will try that as soon as I do a full backup.

 

[Kusuri]

So I moved my good data off the HDD and decided to run chkdsk just once more with 300GB free space just to see if it was a problem caused by having none at all but it was back to "Error writing the output log" and as expected filled it up all to 100% again.

I did a quick reformat and have all my free space again. The event log error messages are also gone and I can safely remove (actually found this latter issue was probably caused by booting with the drive connected).
Still I'd like to know why chkdsk fails like so on this drive. It does take a very long time and it's both my first USB 3.0 and first 2TB drive.

Taking another look at SMART I've noticed the Power-off Retract Count number coincides with the number of times I've ran chkdsk so far, which is 3.

Also I've noticed that during stage 4 or 5 the HDD activity LED seems to turn off? Also can't feel or hear any activity, even though chkdsk shows to seemingly be checking free clusters normally. Wonder if that's normal behaviour...

Do you recommend me anything else I could do now that I have all my free space back?