HDD Issue

[Phileeny]

So I had a HDD Ironwolf 8TB in my system for a long time, over time I filled it with various files, scanned photographs and negatives, media rips, you get the idea.

I decided to move all the files to my server then I started having issues, writing files to the drive was fine but reading, some were fine some not.

I did a disc check in windows 11 but that seems to hang, then I found a program called DiskGenius and set it running.

Wow, it says its been running for 155 Hours but that's not true, the timer only moves then it's checked a cylinder, and that could take an hour, I turn it off after a week to let the drive calm down, then resume a day later.

It's been checking the drive for over three months, is it worth it? no, well sort of, most files I have a backup or can redo, it would just be a pain, about 10% no backup but no problem if I loose them, at 66% done I am now curious if the program can remove the bad areas and if I can get any files back.

[porina]

Based on the screenshot about 16% of the scanned disk so far is busted. That's too much to "repair". I would not trust any files on that disk now as they could be corrupted. Even if it is a pain to restore from elsewhere, that would be safer. I don't know how these disks are arranged internally, but if we assume it is 3 platter, for 6 sides, then one of the heads might have gone causing one side to be unusable.

 

Would be interested to see a CrystalDiskInfo screenshot for that disk if possible.

[Levent]

I used to do a lot of things to fix my hdds but in the end, out of like 30 disks I spent countless days trying to fix only ONE actually got "fixed".

 

Best you can do is to do a random writes to entire disk, then attempt to read again to see if anything changed. If not, throw the disk away. If it did, you may get a little more life out of it.

 

As far as recovery, I am doubtful that you will have any useful data recovered out of that.

[Phileeny]

yes I'm aware the drive is fubar, I could copy/move some files off the drive at first but some not, that's why I started this, now it's going to be fun (for me) to see how much of the 8TB I can get.

oh CrystalDiskInfo will have a look at that, it would be interesting

[Phileeny]

CrystalDiskInfo

The drive has been on an external USB hub with a Noctua fan strapped to it for months

[jaslion]

24 minutes ago, Phileeny said:

CrystalDiskInfo

The drive has been on an external USB hub with a Noctua fan strapped to it for months

Oh thats done for say goodbye to the data

[porina]

That is... interesting. The drive has been on for over 7 years?

 

Reallocated sector count: lots. It makes me wonder, just how many spare sectors do they have on these drives? I don't know if the raw value is a literal count, but it seems that way on other failing disks I've had. That would be nearly 400k decimal sectors. Don't know if they're counted as 512b or 4k sectors, but that would be equivalent to 1.6GB worst case. The value of 60000 is suspect given it is so round, so I do wonder if it hit some internal cap. Likewise for the FFFF Reported Uncorrectable Errors, yet there is 0 on Uncorrectable Sector Count.

 

As I said earlier, I'd treat any data on that drive as suspect now. If you can recover anything from other sources, I'd do that. Anything you do pull of this drive, do have a look at it to see if it's been corrupted.

 

If you really want to tinker with it more, normally I'd do a full write followed by a full read. But I suspect in this case all the spare sectors have already been consumed so there isn't really any further repair capacity.

[kokosnh]

I think he literally hit the smart cap on some values, or some value did overflow. The surface scan confirmed that the hdd is actually in really bad shape ( died like a year ago, he just didn't know it )  so it's not a smart bug, that hdd is really that bad.
he even hit the threshold on the read error rate, and that is rare on seagate.

[Pusbucket]

5 hours ago, Phileeny said:

 

anybody else notice something odd. The groups of bad cylinders are all 7-8 cylinders and not contiguous.

Never saw anything like that before. Could indicate a flaw in the the manufacture of the platters or PCB.

When my 23 year old drive began to fail, all the bad cylinders were at one end of the drive with a few random single cylinders marked bad and it continued to spread to adjacent cylinders, exactly what you would expect with the magnetic failure of a hard drive.

This looks like something else because of being consistent size and spread across the entire drive.

Weird.

[porina]

3 hours ago, Pusbucket said:

anybody else notice something odd. The groups of bad cylinders are all 7-8 cylinders and not contiguous.

Never saw anything like that before. Could indicate a flaw in the the manufacture of the platters or PCB.

I mentioned earlier, the indicated errors are roughly 1/6 of the disk. If we assume 3 platters, for 6 sides, then if one head has gone bad, that might account for it. I don't know what internal structure that disk might have.

 

Also "cylinders" is weird usage in that software. In ancient times, it referred to a physical track on the disk you had to enter into bios manually. I'm not sure it has a meaning in the modern era. I think the software is referring to sectors/LBA.

[kokosnh]

 

1 hour ago, porina said:

I mentioned earlier, the indicated errors are roughly 1/6 of the disk. If we assume 3 platters, for 6 sides, then if one head has gone bad, that might account for it. I don't know what internal structure that disk might have.

It’s a 6/12 Disk/platter model, so like 2 heads, or one Disk, strange yes, but not really something we can do anything about.

[Ryker Robb]

On 1/11/2026 at 1:35 AM, Phileeny said:

So I had a HDD Ironwolf 8TB in my system for a long time, over time I filled it with various files, scanned photographs and negatives, media rips, you get the idea.

I decided to move all the files to my server then I started having issues, writing files to the drive was fine but reading, some were fine some not.

I did a disc check in windows 11 but that seems to hang, then I found a program called DiskGenius and set it running.

Wow, it says its been running for 155 Hours but that's not true, the timer only moves then it's checked a cylinder, and that could take an hour, I turn it off after a week to let the drive calm down, then resume a day later.

It's been checking the drive for over three months, is it worth it? no, well sort of, most files I have a backup or can redo, it would just be a pain, about 10% no backup but no problem if I loose them, at 66% done I am now curious if the program can remove the bad areas and if I can get any files back.

Based on the bad sector patterns, this drive probably has a bad head, you'll need to take this to a professional data recovery service to have a head stack swap to recover all the data.

On 1/11/2026 at 2:40 AM, porina said:

I don't know how these disks are arranged internally, but if we assume it is 3 platter, for 6 sides, then one of the heads might have gone causing one side to be unusable.

This is a 6 platter 12 head drive actually, but you're right.

21 hours ago, Pusbucket said:

anybody else notice something odd. The groups of bad cylinders are all 7-8 cylinders and not contiguous.

Never saw anything like that before. Could indicate a flaw in the the manufacture of the platters or PCB.

When my 23 year old drive began to fail, all the bad cylinders were at one end of the drive with a few random single cylinders marked bad and it continued to spread to adjacent cylinders, exactly what you would expect with the magnetic failure of a hard drive.

This looks like something else because of being consistent size and spread across the entire drive.

Weird.

This grouping of bad sectors is common with drives with a specific head or platter surface that is bad, while the rest of the drive is still working.