Seagate 8 TB disk warnings

[Jheroen]

Hello everyone, One of the disks indicates that the number of bad sectors is increasing. This can also be seen in the graph in storage management, but Ironwolf health indicates that everything is fine (normal status). How should you view/assess this deviation?

 

[zombiephysicist]

How many bad blocks and how fast is it growing. A small number of bad blocks is no big deal. It could become a big deal if it’s rapidly increasing. Rapid increase can be a warning sign that it’s getting time to replace the drive.

 

[Jheroen]

How many bad blocks and how fast is it growing. A small number of bad blocks is no big deal. It could become a big deal if it’s rapidly increasing. Rapid increase can be a warning sign that it’s getting time to replace the drive.

It went from 80 in dec 2023 to 4232 in feb 2024, it seems to grow fast.. strange that Ironwolf health indicates that everything is fine
About choice for replacement, i can choose for WD or the same as i have now, any suggestion?

 

[Rusty]

Personally I would stick with same model as the current one but def get a new drive and have it on hand. Even better it would be best to replace it before it fails, but in any event you will still need to rebuild the array once your replace the drive (sooner or later).

 

[Telos]

It went from 80 in dec 2023 to 4232 in feb 2024

1. Ensure your off-NAS backups are current NOW!
2. If your system is btrfs formatted, run a scrub
3. Replace that drive as soon as the scrub completes

 

[zombiephysicist]

It went from 80 in dec 2023 to 4232 in feb 2024, it seems to grow fast.. strange that Ironwolf health indicates that everything is fine
About choice for replacement, i can choose for WD or the same as i have now, any suggestion?

That is not a lot of blocks. I had a drive similarly start throwing errors. 4232 bad blocks is miniscule. It's basically a touch over 2kilobytes out of 8TB.

BTW what RAID type are you using? RAID 5, SHR? Or is this is a single drive with no redundancy?

I have had several drives start throwing similar levels of bad blocks (all iron wolf drives which is why I stopped using them and and have also had nothing but woes with WD drives. I now use only Seagate EXOS or Toshiba Enterprise drives). Out of the 5 drives that started throwing errors like that for me, 4 of them 'settled down' while 2 progressed. The 4 that settled down increased bad blocks for about 6 months and then the number just stayed stuck there. However, two of the drives, the bad block count just kept creeping up consistently for over 6 months. The number of bad blocks was not large, but I just didn't like the regular creeping, so I replaced them.

It's important to note that it's not at all unusual for such drives to have some small number of bad blocks and for them to be just fine. It's more telling as to the velocity of growth (or lack thereof) of bad blocks over time.

@Rusty advice is sound and it depends on your risk tolerance. Since my RAID had one drive of redundancy, and I had great backups, I chose to risk it and watch those drives. More so for my own education on how these things go bad. Some of those drives are now near 7 years old and still working just fine!

Also, I kind of disagree with Rusty on using the same exact type of drive. Not that it's a bad idea. If you are very conservative, it's a fine strategy. But with SHR I look at replacing a drive with a larger capacity drive to be an 'upgrade opportunity'. I still use high quality drives like Seagate EXOS enterprise drives, but I do look to expand the capacity of the RAID seeing what the best bang per TB high quality drive is going for, and upgrade the capacity.

BTW, if you have room (a spare bay) and want the ultimate conservative 'peace of mind' but still want to watch how this develops, you can throw in a drive as a hot spare, so if something goes really bad, the NAS will automatically grab the hot spare and immediately start rebuilding.

TLDR I do not think you are in any imminent danger of losing the drive or your data. But much of life depends on your risk tolerance. If you are conservative, sure, get a new replacement drive and enjoy peace of mind. That said, it is a very low number of bad blocks and very little growth over 3 months, so it's not clear if the drive is going bad.

 

[Coop777]

That is not a lot of blocks. I had a drive similarly start throwing errors. 4232 bad blocks is miniscule. It's basically a touch over 2kilobytes out of 8TB.

BTW what RAID type are you using? RAID 5, SHR? Or is this is a single drive with no redundancy?

I have had several drives start throwing similar levels of bad blocks (all iron wolf drives which is why I stopped using them and and have also had nothing but woes with WD drives. I now use only Seagate EXOS or Toshiba Enterprise drives). Out of the 5 drives that started throwing errors like that for me, 4 of them 'settled down' while 2 progressed. The 4 that settled down increased bad blocks for about 6 months and then the number just stayed stuck there. However, one of the drives, the bad block count just kept creeping up consistently for over 6 months. The number of bad blocks was not large, but I just didn't like the regular creeping, so I replaced it.

It's important to note that it's not at all unusual for such drives to have some small number of bad blocks and for them to be just fine. It's more telling as to the velocity of growth (or lack thereof) of bad blocks over time.

@Rusty advice is sound and it depends on your risk tolerance. Since my RAID had one drive of redundancy, and I had great backups, I chose to risk it and watch those drives. More so for my own education on how these things go bad. Some of those drives are now near 7 years old and still working just fine!

Also, I kind of disagree with Rusty on using the same exact type of drive. Not that it's a bad idea. If you are very conservative, it's a fine strategy. But with SHR I look at replacing a drive with a larger capacity drive to be an 'upgrade opportunity'. I still use high quality drives like Seagate EXOS enterprise drives, but I do look to expand the capacity of the RAID seeing what the best bang per TB high quality drive is going for, and upgrade the capacity.

BTW, if you have room (a spare bay) and want the ultimate conservative 'peace of mind' but still want to watch how this develops, you can throw in a drive as a hot spare, so if something goes really bad, the NAS will automatically grab the hot spare and immediately start rebuilding.

TLDR I do not think you are in any imminent danger of losing the drive or your data. But much of life depends on your risk tolerance. If you are conservative, sure, get a new replacement drive and enjoy peace of mind. That said, it is a very low number of bad blocks and very little growth over 3 months, so it's not clear if the drive is going bad.

If he has direct access to the NAS, a cold spare makes much more sense. Why put wear and tear on a drive that spends its time mostly doing nothing?

The old school Network admin in me agrees with Telos - make sure backups are current. Assume the worst can happen, hope for the best, but being prepared for it makes things much less stressful.

 

[Telos]

It went from 80 in dec 2023 to 4232 in feb 2024

it is a very low number of bad blocks and very little growth over 3 months, so it's not clear if the drive is going bad.

It is frightening high number of bad blocks in short order.

 

[zombiephysicist]

It is frightening high number of bad blocks in short order.

If that's your view, fair enough. I've had similar movements. Wasn't a big deal to me. My experience is it starts with around 80 blocks. It goes up relatively large number of blocks more, but still and overall small number of actual bad blocks. Then it levels off. So for the times I've had it, that was the pattern, and then it died down where after the 2nd bump, it just didn't go up much more at all after that.

So the progression for me was something like this. 100bad blocks, next report next month 5000 bad blocks, then next month say 6000 bad blocks, and then months after that it just stays the same. That's been my experience over 6 drives experiencing bad drives. 2 of which just kept increasing, but 4 of which settled down like that.

Your view that it's frightening is totally fair and understandable. My view is it's kind of par for the course and can level off. But we're talking WELL under 1 Megabyte out of 8Terabytes of data, so on that relative scale, it's really marginal. We're talking less than 3 kilobytes. It's relatively very little IMO.

That said, your assessment is perfectly reasonable. I don't begrudge your view.

 

[Jheroen]

Thank you all for your responses, I have ordered a new drive of the same brand and type so that in the event of a continued increase in bad blocks I can replace the drive.
The array is RAID 5 and consists of 4 8TB drives.
Given your comments, it seems to work that the number of bad blocks does increase, but Ironwolf health does not (yet) see this as a problem. Margins will probably be higher.
I leave the drive on the shelf so I can replace it immediately if things suddenly go wrong.

 

[zombiephysicist]

Thank you all for your responses, I have ordered a new drive of the same brand and type so that in the event of a continued increase in bad blocks I can replace the drive.
The array is RAID 5 and consists of 4 8TB drives.
Given your comments, it seems to work that the number of bad blocks does increase, but Ironwolf health does not (yet) see this as a problem. Margins will probably be higher.
I leave the drive on the shelf so I can replace it immediately if things suddenly go wrong.

Please let us know how it develops and progresses. Hopefully you won't see much more progression. If you get too nervous, just do the drive swap. And since it's raid5 getting the exact same drive is the right way to go.

Knock wood, hope it all works out smoothly.

 

[zombiephysicist]

Thank you all for your responses, I have ordered a new drive of the same brand and type so that in the event of a continued increase in bad blocks I can replace the drive.
The array is RAID 5 and consists of 4 8TB drives.
Given your comments, it seems to work that the number of bad blocks does increase, but Ironwolf health does not (yet) see this as a problem. Margins will probably be higher.
I leave the drive on the shelf so I can replace it immediately if things suddenly go wrong.

By the way. I would do a regular SMART health check in addition to the specific iron wolf health check.

 

[Jheroen]

By the way. I would do a regular SMART health check in addition to the specific iron wolf health check.

Yes that checks run automatically so I'm informed about the (changing) status of the disk. Ordered a 'new' disk on amazon, got the disk yesterday but it shows refurbished so sent it back to the seller. Also warranty showed until july 2024... Now buying another one from another store.
[automerge]1708692944[/automerge]

Please let us know how it develops and progresses. Hopefully you won't see much more progression. If you get too nervous, just do the drive swap. And since it's raid5 getting the exact same drive is the right way to go.

Knock wood, hope it all works out smoothly.

4336 bad blocks today, the number goes up

 

[zombiephysicist]

Yes that checks run automatically so I'm informed about the (changing) status of the disk. Ordered a 'new' disk on amazon, got the disk yesterday but it shows refurbished so sent it back to the seller. Also warranty showed until july 2024... Now buying another one from another store.
[automerge]1708692944[/automerge]

4336 bad blocks today, the number goes up

Does the increase seem to be settling down? It went from 80 to 4232 over 3 months. And now to 4336 in how many days?

The thing to check is kind of plotting a curve. Is the rate of bad blocks per week increasing, roughly the same, or going down. If it’s going down you can probably watch it for a while and it may settle down and stop. It’s still a small number of blocks.

If it’s increasing to more many more blocks each week, you may want to swap the drive. At those small numbers of blocks I personally kept watching for about 6 months to figure out the trajectory and I was morbidly curious how the failure might come about or not.

But for most people, if the numbers are still increasing per week after 3 months, then you probably will want to put in the new drive..

 

[EAZ1964]

My 2 cents: replace.
It will only get worse, better replace the disk before issues appear on a second disk and you will loose the whole volume during rebuild.
Using 4 of the same disks increases the risk that they fail in the same time frame. I always recommend to mix disks both in brand and age, but not in RPM.

 

[Telos]

4336 bad blocks today, the number goes up

Maybe I've overlooked the RAID type here... however if it is SHR (with one drive redundancy), and one chooses to delay drive replacement... Should another drive fail, realize that the integrity of the RAID group now hinges on a drive with 4000+ bad blocks, a drive that must live through a rebuild cycle... just saying...

Good to hear a replacement drive is on the way.

 

[Coop777]

Maybe I've overlooked the RAID type here... however if it is SHR (with one drive redundancy), and one chooses to delay drive replacement... Should another drive fail, realize that the integrity of the RAID group now hinges on a drive with 4000+ bad blocks, a drive that must live through a rebuild cycle... just saying...

Good to hear a replacement drive is on the way.

yep, that is going down the road of playing with Fire, which likely will not end well.

 

[zombiephysicist]

To give you a sense of scale of how few blocks this is, many drive manufacturers would refuse warranty service for this.

A program like hd sentinel can tell you the health of your drive but will let you know that even more blocks in a continuous spot won’t qualify as warranty replaceable.

Most dives have 3-5% blocks available as spares.

How many spare sectors are in a hard drive?

Answer: Depends on the hard drive. Some have 0, the most I've seen is 12% of the size of the drive. I think 3-5% is typical for remapping on a mechanical drive, and 6-9% on a SSD. If on the other hand you mean how much free space is on some drive that you happen to have without giving any deta...

www.quora.com

So we’re talking like 0.0001% here. So we don’t need to freak out. Again, the important thing is the trajectory of the bad block reporting.

The pattern is almost always, report 1 is like 100 blocks, report 2 jumps relatively to say 5000 blocks, and then the increases after that are where you start plotting the trajectory. If the rate of increase after that is starts leveling off, odds are the drive will be fine. If the rate of increase stays the same or is increasing, you really want to act fast in replacing.

This, imo, is educational. Your personal preferences are all reasonable. Some may want flawless drives with no bad blocks. That’s fine. Some like me, don’t care as long as the bad block count dissipates and stays relatively low relative the number of spare blocks.

But the point is to diagnose is the drive one that is kind of within spec or one that is exhibiting imminent failure. The curve and rate of bad block increases (starting with the 2nd and 3rd bad block count reports and then the 4th bad block starts to establish the initial trajectory, because the first report tends to be misrepresentatively/anomoly-like low). It’s more an issue in how stable is the drive and errors. Are they minimal and staying minimal and are the additional new bad blocks on a downward trend, if so, it’s not that bad and may range from “you got plenty of time to fix this” to “the drive is stable and fine and can last years longer”…or if the bad block reports increase in relative numbers, are large in absolute numbers, or show large erratic deviations (also coupled with reports of lots of spin up/downs) then you know time is really of the essence.

The point being, it’s fine to replace the drive immediately if your risk tolerance is more conservative, but it’s still important to properly diagnose, just how dire the condition of the drive truly is. And it’s fine to be conservative and act like any imperfection is dire, sure, better safe than sorry etc etc, but it’s still important to note, even with that philosophy, some conditions are still way more dire than others. And on that scale, this doesn’t scream 10 on a 1-10 scale. It says 1 to me. It’s a bit like tires. Some get a flat and plug the tire and go on to have years more of use. Others who may drive super cars at super speed immediately change all 4 tires because they want perfect and wear matched tires to be able to feel confident driving at hyper speeds. So a scale of defect even at level 1 for them is not acceptable, where for others, it’s just fine. Depends on many factors, your use case, and personality, all being factors. Still, knowing the scale of the problem is important. For example a sidewall puncture versus one directly in the tread side or a bent rim all are hugely different in scale of the problem.

Anyway, here’s another thread with similar reports:

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

View: https://www.reddit.com/r/synology/comments/mnbtu6/my_disk_has_9528_bad_sectors_and_somehow_is_still/

Anyway, appreciate your reporting on the progress here. I think it’s a great bit of experience to share with others.

 

[Coop777]

well, generally speaking, leaks may start small...but they generally only get worse as they won't fix themselves.

 

[zombiephysicist]

well, generally speaking, leaks may start small...but they generally only get worse as they won't fix themselves.

I can’t speak on leaks, but I see your analogy. I have no statistics on this and you may well be right. It might be cool to see if drive blaze or anyone keeps statistics on this, bad block growth. As far as I know they just keep actual total drive failure statistics.

That said, my totally unscientific stats are I’ve had 6 drives out of probably 50 that reported bad blocks. 4 of them stabilized after several months and have gone on to be just fine (they are now over 7 years old and still running). 2 of them, the block counts kept increasing linearly for too long for my comfort so I pulled them. I can’t tell if they would have failed, or stabilized, but after the 7th report (i think past 6 months) with the block counts continuously throwing more bad blocks I had had enough and replaced them.

Of course that is highly anecdotal and statistically irrelevant, but just my experience. But it does show that drives are not like leaks at least in some cases and can “fix themselves” and it’s the entire point of why modern drives come with spare blocks and literally have functions to remap bad blocks (ie to fix their own leaks).

Having suffered water damage in the home, I dearly wish water pipes had a similar capability!
[automerge]1708713105[/automerge]

Hello everyone, One of the disks indicates that the number of bad sectors is increasing. This can also be seen in the graph in storage management, but Ironwolf health indicates that everything is fine (normal status). How should you view/assess this deviation?

Btw, how old is this drive? How many hours of use. Does it exhibit any clicking noise or do you have a way to see or hear that it randomly has spin downs and spin ups?

If you hear any clicking or hear excessive unexplained spin up/downs, seek to replace the drive immediately.