This is in relation to Linux on my main laptop thread. It became a distro question, and now it has graduated to a hard drive question.

Original thread:
https://www.linuxquestions.org/quest...op-4175667927/

So just discovered my hard drive had a bad sector count of 200 sectors. I have installed the SMART tools (smartctl, gsmartcontrol) and I have attached a screenshot and logs. It still managed to pass the extended 3 hour SMART test.

I think in the coming months I will likely get a new laptop but I was just wondering what some of these numbers/results mean and what kind of future my drive may have (if any). I'm going to probably end up keeping the laptop as a test laptop until she finally gives up.

Attached Thumbnails

 

Attached Files

TOSHIBA_MQ01ABD075_33MMT1S8T_2020-01-24.txt (27.9 KB, 22 views)

I'd get the OEM diag suite if possible to run it. They usually have the numbers that tell good or bad. It is common for a drive to show errors and I think smart will give some reference number. Laptops get some rough treatment.

12K hours doesn't really mean the thing has been moving data, just that it was powered on.

1 members found this post helpful.

Laptops can get surface damage on spinning disks if not handled carefully, your dead sectors could be from rough handling, or age, but whichever it is, I've been running some old HDDs for years with some bad sectors present.

If you're worried that it may be failing, I'd suggest keeping an eye on how many new bad sectors appear, & how quickly.

1 members found this post helpful.

I would not panic at a few bad sectors. they may have been there since the drive was new. a certain number are normal and allowed.

That said, I would be more aware of your backup integrity and backup schedule, and make sure my critical data was backed up or duplicated in another environment less subject to loss in the unlikely case that the drive fails catastrophically and unexpectedly.

2 members found this post helpful.

there use to be, back, way back in the stone age of hdds that did not get to close 1GB, you could mark the bad sectors to not use them then get on with life.

2 members found this post helpful.

My desktop's drive had over 100 bad sectors and still worked — until it suddenly stopped. In future I'd take bad sectors as as the the writing on the wall.

2 members found this post helpful.

A bad sector count of 200 suggests that the drive might be on its way out. A laptop can easily develop a small number of bad sectors and not be of concern, but "200" is really not a small number in that context. The SMART system built into the drive will not predict failure until the drive has almost exhausted its supply of spare sectors (thousands), but that is far, far beyond the point at which the drive should have been removed from service.

If you do continue to use the drive, you should use a monitoring daemon like smartd to report any changes in status. If you see additional bad sectors develop, the drive should be replaced immediately.

1 members found this post helpful.

The number of bad sectors isn't the thing to worry about, it's their escalation rate. i.e. how fast the count increases. All hard disks have a bad block file on the disk, they are made to look perfect to the operating system by having the bad block manufacturing imperfections vectored out to a different are of the disk. (There are probably a couple of spare cylinders allocated for this purpose.) If the bad block count is increasing quickly then the disk is probably on the way out. If it's stable you're probably OK, but keep monitoring now and again. Laptops are more prone to rough handling.

Going by your disk geometry file, your 750Gb disk has actually got a capacity of 750,156,374,016 bytes with a physical sector size of 4096 bytes. This, with my poor grip of mathematics, seems to equate to 183,143,646 sectors of which you have a mere 200 marked as bad. That's 0.0001% of the total sectors faulty according to my calculator which is struggling to work with such large numbers.

I reckon your disk is probably OK for the reasons above but check it every now and again just in case. In the bad old days I used to reformat dodgy partitions periodically on 500Mb drives to re-vector bad blocks, even 200 sectors on a disk as small as that didn't warrant immediate replacement.

My

Play Bonny!

2 members found this post helpful.

I had wondered about finding some sort of OEM diagnostics. I dug out my old Ultimate Boot ( https://www.ultimatebootcd.com/ )hardware diagnostic just to check, but under its hard drive diagnostics it didn't have anything under Toshiba manufactured hard drives...mainly the other ones: Seagate, WD, etc. The other utilities MAY have worked, but I hadn't tried them yet.


I kinda thought maybe it did get some bad sectors from rough handling. It doesn't really get too abused as maybe some laptops, but I'm sure it's been bumped a time or two.

I had already backed up the files from that drive anyways. I'm going to be moving them to my other laptop and then some will probably go on my file server. I posted another thread on the topic, but I have been definitely been on more of a kick to fix our backup routine/schedule. Here in the next few months I'll purchasing some larger terabyte drives to keep some copies of our data on.

Good suggestion, I may look into that.


Thanks for the replies everyone. Right now, the laptop is not really going to be used for anything mission critical especially since I will get a replacement in the coming months. And I'm definitely going to be keeping an eye on it for increasing bad sectors.

SMART doesn't monitor bad sectors only. There are many other things concerning hard drive's health. I don't think hard drives die from lack of spare sectors. More serious problems can be with heads, drive electronics, interface electronics... For instance, voltage shock from a defective power supply unit can damage drive's electronics (as well as motherboard, CPU and other parts too). Drive's temperature is something to monitor, too.
Nevertheless, an increased number of bad sectors could make us think that there is something that "helps" them go bad

And yes, backup is the only way to keep our data safe.

1 members found this post helpful.

To put numbers to it all, I like to refer to the Google study that examined the collected details of over 100,000 hard drives used in their data centers.

They concluded based on this study:

(a reallocation count can be read as a bad sector that needed reallocation).

And of course, I know, a laptop hard drive has a different use, a 2,5" hard drive is not a 3,5" hard drive, the study is about hard drives from the early 2000's - however the paper makes for interesting reading and might shed some light on what you can expect of a hard drive with bad sectors. You can find it here online (PDF): https://static.googleusercontent.com...k_failures.pdf

It is up to you if you want to accept data loss on that drive. For example, I have a drive with some bad sectors running in a backup RAID array. It is the backup array, it is in a RAID 1 array so I don't worry about the drive: sudden and complete drive failure will not result in data loss and even the loss of the complete array will not directly result in data loss. So I don't mind and sleep well at night.

However in your case, things might be different. You can choose:

1. replace the drive now, on a time of your choosing with all data preserved
2. replace the drive later when it has failed, with possible data loss, and at a time you might not have the time to do it
3. do not replace the drive and buy a new laptop when the drive fails, knowing all files are safe because you made sure no files are to be found on the laptop only

In your case, I'd opt for 3 because 12,000 hours of lifetime activity means 500 days of constant use. And because you have to sleep and do other things, the laptop may be an old laptop. And because, finally, you always have to make sure no data is on a single point of failure.

Choose wisely

Thanks for the article Hermani...I'll check it out. It'd be interesting to see what their experiences were.

As for the drive/laptop, currently the laptop is just being used as a simple Linux Mint workstation that I sit next to my file server. I use it just for looking up things on the net if I have questions/issues with my server. No major data is stored on it now. If it dies tomorrow, life will go on with no major data loss. I have moved all my major data over to a Samsung laptop with similar specs. I'll just use it until I get a new laptop.

1 members found this post helpful.

Other hard drive diags may provide some help and maybe 100% working but I always try to get the OEM diags on their web site. Sometimes might have to email for support. The OEM diags in my opinion tend to be best as they have inside track on device.

Generally if drive starts developing bad sectors, data on it no longer can be considered safe. But possibly you can still use it to hold some non important data you're not afraid to loose
If your drive however is getting new bad sectors then you have a problem immediately copying all data off of the disk and replacing it. You can format the drive
If you have part of a file in a bad sector then that file will be corrupted and likely unusable.