"Reallocated Sectors Count"

Topic

New Reply

My desktop has 2 spinning platter disk drives.  The System is on the primary and I use the other one as a Data drive.  Periodically I run CrystalDiskInfo in order to view the S.M.A.R.T. summary.

Recently, CrystalDiskInfo has been warning me that my “Reallocated Sectors Count” on the Data Drive is too high.  I did some research and found some opinions that catastrophic failure could be imminent.  I backup important data quarterly.

However, the count of reallocated sectors didn’t seem very high to me (see attached image 1) so I ruminated about whether to soon but a new drive or just keep an eye on the count once a week.  The count has not changed.

And then my subconscious smacked me in the head and said “Why not see what Speccy has to say”.  Well, Speccy says the “real” count is 121 and the drive’s Health Status is “Good” (see attached image 2).  Instant Conundrum!  BTW, Image 1 gives the drive’s specs.

So, Is Instant Karma gonna get me or is it reasonable for me to continue with my cautious approach?  I’m hoping that some hardware gals/guys can weigh-in with their expertise.

 

Reply | Quote

Replies

‘Current/Threshold’ is 198/140 in both reports.

I use Hard Disk Sentinel, there’s a demo version available, to get a ‘human-friendly’ overview of the health of my drives: https://www.hdsentinel.com/

If the next bad sector is the one containing the main drive info, you have a new doorstop.

1 user thanked author for this post.

CyGuy

Reply | Quote

You have a value that is above the CDM threshold, so it’s advising you that all is not well. Assuming that the value does not change further you will be OK, but I would keep an eye on it by setting the threshold to one greater than the current value and leaving CDM running in the tray – I do.

cheers, Paul

1 user thanked author for this post.

CyGuy

Reply | Quote

I see this one has had a good long life but, this is the time to get a new drive! However you choose to get data off the drive, cloning to a new one or capturing an complete image. (Hopefully your preferred utility can be set to ignore errors!)

You do not know exactly how much more time you have to wait!

1 user thanked author for this post.

CyGuy

Reply | Quote

It is definitely time to get a replacement drive since that drive is failing.

1 user thanked author for this post.

CyGuy

Reply | Quote

Drives do wear out. The SMART status is good for letting you know that it’s starting to show signs of failure.

Make sure you have a good backup, and if you’re thinking about making one bear in mind that heavy disk use to try to save the data off a drive can result in complete failure, so don’t overwrite a dated backup. Once, years ago, as an exercise I thought I’d see what would happen if I tried to do a CHKDSK /R on a drive that was starting to chalk up errors. It died completely under the load that put on it.

This could be a good time to replace your failing HDD with a 1 TB SSD that you’ve really wanted and kick your I/O into high gear. Better yet, get 2 SSDs and RAID 0 them together for out of this world I/O performance.

I run two RAID 0 arrays of 4 SSDs each myself; trust me, you’ll never want to use another spinning HDD once you’ve experienced computing with solid state storage.

Good luck.

-Noel

4 users thanked author for this post.

Cascadian, jburk07, CyGuy, BobbyB

Reply | Quote

+1 for backup.

There are several free utilities that will backup to an external hard disk – the cheapest insurance you can get.
Aoemi Backupper, EaseUS ToDo, Paragon Backup, Macrium Reflect.

cheers, Paul

1 user thanked author for this post.

CyGuy

Reply | Quote

while using any freeware backup tool to copy data from the old HD to a new one, get a SATA/IDE to USB 3.0 adapter kit. it makes things easier when swapping hard drives & SSDs.

1 user thanked author for this post.

CyGuy

Reply | Quote

SMART values give you a glimpse into the drive’s internal health, but they don’t tell you for sure that it’s going to fail or that it’s fine.

The reallocated sector count is a warning to look into things further, but it doesn’t in and of itself mean that the drive is going bad (which is why Speccy reports it as good).  If the reallocated sector count stays constant, and the other pertinent SMART numbers stay good, the drive may still perform well for years to come.

If the uncorrectable sectors, reallocated sectors, or pending sectors is/are increasing as time passes, that’s a pretty good sign that the drive is failing.  I am a little troubled by what I see in your images; your first image from CrystalDiskInfo shows that at 26594 hours of uptime, you had 115 reallocated sectors.  The second image, from Speccy, shows that at 26882 hours, you had 121.  If it’s increasing, that’s a bad sign.

If there is any important un-backed-up data on that drive, I’d back it up immediately.  I would consider the drive suspect, but I wouldn’t automatically go replace it at once unless you want to for other reasons too, like wanting an SSD or a bigger rust spinner, or if the possibility of downtime would be really bad rather than just annoying.  I’d watch it like a hawk and keep good backups.

Why would I not take the drive out of service at the first sign of possible problems?  Well, I consider ALL drives to be suspect for imminent failure to some degree, so having one that is starting to show some signs is not too far removed from what I thought of it before.  I’ve had progressive failures that start with a few soft (correctable) read errors and progress to total failure of the drive, but I’ve also had failures where the drive showing nothing at all sinister in SMART just up and died while I was using it, or it would not start up at boot time or wake up from sleep mode.  It’s best not to trust any of them too much.

I do have one drive that has been in service a long time (about as long as yours in uptime; mine is at 28415 hours) that has had 11 reallocated sectors for a long while.  It’s doing duty in my backup server, where it has been just as reliable as my newer disks.  I do consider its age and history when I use it for storing backups, but it’s still in service.  Because I consider all disks suspect, I have redundant backups on separate drives on the backup server, and then even more redundant backups on external USB hard drives that get unplugged when they’re not backing up or restoring.

One thing you can do (after backing up any important data on the drive) is to set CrystalDiskMark to view the current sector count as the highest possible acceptable number for reallocated sectors.  From the Function menu, select Advanced Features, then Health status setting.  Move the slider for reallocated sectors to the right until the number shown all the way on the right is one higher than your current reallocated sector count.  That way, it will return to saying it is okay now, but if the count gets any higher, it will alert you again.

Keep in mind that not every reallocated sector was once an unrecoverable error… if the drive can’t read any given cluster on the first try, it will keep trying to read it.  Desktop and laptop drives will generally try heroically to read the data, while drives sold as NAS drives will give up very quickly and report the sector as failed.  This is because the RAID arrays that these drives are often used in may mistake the drive’s heroic recovery efforts (during which time it is not responding) for complete drive failure, at which time the array will mark the entire drive as failed.  If there is data redundancy, often it is best to accept the error and read that sector’s data from another drive.

Most often, through the drive’s heroic recovery efforts, a sector that generates errors on the first attempt can eventually be read, but any such sector that takes more effort to read than it should is marked as potentially bad.  This is represented by the pending sector count… those are the sectors that have been marked as potentially bad, but the drive is not sure yet.  Some soft errors can be corrected once that sector is rewritten, and in that case, the drive will return those sectors to normal service.  If the sector continues to require retries even after being rewritten, it gets marked as bad and added to the reallocated clusters list.  The pending sectors will drop back down when it does this.

I would be suspect of your drive given that increase from 115 to 121.  I’d keep watching that, and plan for replacing the drive unless that count holds fast at 121 (and even then you may wish to for your own peace of mind).  Everyone has different tolerances for risk and downtime.

Dell XPS 13/9310, i5-1135G7/16GB, KDE Neon 6.2
XPG Xenia 15, i7-9750H/32GB & GTX1660ti, Kubuntu 24.04
Acer Swift Go 14, i5-1335U/16GB, Kubuntu 24.04 (and Win 11)

5 users thanked author for this post.

jburk07, SueW, CyGuy, MrBrian, Elly

Reply | Quote

Thank you for taking the time to provide such a detailed and helpful post.

1 user thanked author for this post.

Ascaris

Reply | Quote

S.M.A.R.T. basics

What SMART Stats Tell Us About Hard Drives

5 users thanked author for this post.

jburk07, ch100, SueW, CyGuy, Elly

Reply | Quote

@mrbrian Excellent find, in particular the second URL.
All that anyone needs to know about the subject.

1 user thanked author for this post.

CyGuy

Reply | Quote

I have concerns though that the data in the second link is skewed if the company removes operational drives based upon S.M.A.R.T. data.

1 user thanked author for this post.

CyGuy

Reply | Quote

I also regularly use CrystalDiskInfo. Perhaps 12 months ago, for the first time it gave caution on (and only on) Reallocated Sector Count. My raw value is 9, and the other normalized values are the same as CyGuy’s. In my case, the raw value has remained at 9, and the disk is still working fine.

2 users thanked author for this post.

CyGuy, jburk07

Reply | Quote

Under the “believe it or not” category. I’ve actually temporarily revived a dead hard drive by using the old “put it in the freezer” for a day. It doesn’t always work but it has worked for me more than once. Wrap the drive in plastic, freeze it and then next day fire it up and grab your stuff quick (if it works).

Red Ruffnsore

2 users thanked author for this post.

CyGuy, SueW

Reply | Quote

Well speccy is always saying that everything is good.. in my experiences i have situations that drives fail without any warning shown in smarti so if your drive is getting old don’t store important data on it or at least do some backups in your situation that drive can only be used as a secondary for non important data like movies or something like that.

1 user thanked author for this post.

CyGuy

Reply | Quote