Well, ig I had to guess, it's
this post was submitted on 17 Feb 2026
10 points (100.0% liked)
datahoarder
10247 readers
1 users here now
Who are we?
We are digital librarians. Among us are represented the various reasons to keep data -- legal requirements, competitive requirements, uncertainty of permanence of cloud services, distaste for transmitting your data externally (e.g. government or corporate espionage), cultural and familial archivists, internet collapse preppers, and people who do it themselves so they're sure it's done right. Everyone has their reasons for curating the data they have decided to keep (either forever or For A Damn Long Time). Along the way we have sought out like-minded individuals to exchange strategies, war stories, and cautionary tales of failures.
We are one. We are legion. And we're trying really hard not to forget.
-- 5-4-3-2-1-bang from this thread
founded 6 years ago
MODERATORS
18 write errors and starting to show read errors? She’s cooked. Get the data off there before it’s too late.
It's not worth risking with anything important. That being said I have a few drives that are in bad shape still being used for a Steam library. I'm comfortable with that use case because if the disks die I can just re-download the data. So unless you have a similar no consequences use case I would retire that drive.
I do the same! Failing drives are nice for storing games. They tend to still last for longer than we expect
Another option is to use some old/bad drives in a raid setup with lots of redundancy. I still wouldn't store anything mission critical on there but it's fine for most things
59 hours of usage with 140 reallocated sectors and a failing read error rate - my guess would be the drive has been dropped or damaged somehow, with such low hours of usage and a failing error rate. This would be an RMA if under warranty. Don't trust it for anything important.
Typically I would expect a mechanical WD drive to last for 5 full years of usage - about 44000 hours of run time or more. I've had several reach that age with zero or very few errors.
It's definitely out of warranty by now. Think it was bought around 2018 from Costco. I don't believe the drive has been dropped but it has been moved several times.
There's a separate Power On Hours up in the information field at the top showing 30020 hours (3.5 years). Maybe CrystalDisk is having trouble interpreting the SMART data.
CrystalDiskInfo reports it’s bad but CrystalDiskMark shows decent read/write speeds
It can be bad and fast at the same time.
This is true. I get insane write speeds when storing data in /dev/null
It is much more important if the numbers are increasing then how high the numbers are. You can have multiple bad sectors or on SSDs Media Errors and the drive will be good for years to come.
I would recommend data hygiene in the first place. Have a working backup! And if you can afford it (can you afford to loose your data) some kind of redundancy like raid zfs or therelike.
I have as of now multiple drives at home and work in operation that have some form of error but have not changed their error values in literally years. Could i have afforded to replace the drives? Sure, but i also could have had a drive as a replacement that fails during the first resilver of the array.
Are you trolling?
Even if though it says it’s writing, if it can’t read you might as well backup to /dev/null.
Had something similar happen to one of my WD drives a few weeks ago. Luckily still in warranty, triggered the RMA, got the new drive a few days ago. So a least they still have stock for RMAs.
there's pending and uncorrectable sectors, stop using it. is this an external drive? considering the read error rate it might be caused by a cable issue, or possibly even the power source.
back in the days when i had no drives i'd zerofill the whole drive and perform a bad sector check, then partition it so that the bad sectors are skipped then use the remaining parts. i don't recommend doing this if you can afford proper replacements though, and it's also very likely to get worse over time.
Yes, it's an external drive. Specifically WD MyBook
Commenter didn't say it, but also the IO board may be bad too. Drive may be fine if you crack it open and treat it as an internal drove. (Note you may be forced to reformat it so the data would be lost, but functionally it might be fine)
It's a fine paper weight.
Current pending sectors....
No good.
If that number stays anything not 0, that means it's toast. Can't rely on it.
Turn it into a clock or Frisbee.
Thats bullshit. I have multiple drives that have pending sector count higher then 1 and they perform perfectly fine after that for years now.
Calling it bullshit is funny.
Just because it might be working for you, the reliability of the drive under load is compromised.
This literally means there's data changes awaiting a new replacement chunk to be allocated after a region deemed unstable. If this number doesn't return to zero I a timely fashion, it means there's data changes unwritten.
If you run a raid, expect issues from any drives with pending sectors.
This is from experience managing statistics on dozens of data centers... Not sure what ShortNotes use case is, but my priority on a storage device is data integrity.
For data integrity you do not rely on single drives but on for example FS that handle that.
As i said in another post, the important thing about smart is not the values itself but if they are start to increase or not.
And even if an read error occurs, the sector gets remapped and you can restore the block/file from backup or the fs will handle it without interference.
Ok, but that doesn't work on hardware raid. Regardless, his drive has already failed remapping and it's dying.
I use ZFS too. That's not the issue.
This is pointless since the op asked about single drive and the numbers show it's failing. Pending means stuck, and uncorrected means it won't get any better.
In this case , they shouldn't trust the disk with anything they care about. That's it.
Yes in this case i would not disagree. The overall stats look no good and if it is a single drive then yes. Get the data off.
I take issue with the general statement of yours:
If that number stays anything not 0, that means it's toast. Can't rely on it.