How many bad sectors is real bad? HDD talk

[Xeogred]

I've got a Seagate 3TB External that's popping up "yellow caution" in Crystal Disk Info and the sector count problems look ridiculous:

Reallocated Sector Count: 704
Current Pending Sector Count: 592
Uncorrectable Sector Count: 592

Seems like a lot. In comparison one of my external drives has almost always been in the caution zone (old drive I think), but for years has just had 16 pending and 10 uncorrectable. I keep tabs on it and it's never gotten worse so I hope that one is good. I defrag manually off and on (and I think it does it automatically as the "last run" dates are always curretn), I'm not sure if it can be repaired though?

But yeah, that Seagate... bad news? It runs at 53 degrees right now too, which is also highlighted yellow. Perhaps my first and last Seagate, because I have an older 2TB WD Elements that is still rocking, and in the past I never had severe issues with other WD externals either. Heck, this Seagate has a power on count of 110, my WD Elements says 3,252!

Right now I have a 6TB WD Green internal that has about 3TB free, and I'm going to swap out a 1TB internal for a 6TB WD Black (hope the access speeds are much better than the green). So I more than likely can salvage anything from this Seagate, bit of a headache... but you guys let me know what you think. I'll probably handpick some stuff and back things up here and there once I get that WD Black in.

Now what's extra sketchy here though, is that I can't find this Seagate drive on Amazon anymore, I can't even find my order. I just went through my huge 10 year some history and then tried to name search it and it's not coming up at all. I'm sure Amazon delists things, but this is the first time I've not been able to find previous order history. WTF?

[samsonlonghair]

Maybe you ordered it from newegg not Amazon?

[Tanooki]

Amazon doesn't remove order history that's just too weird. I just had a Seagate get taken out probably by windows 10 recently when I was down for a few weeks. Win10 would consistently keep hitting that common problem of it doing the disk at 100% again and again. Eventually it just took its toll trying to write and read all over doing normal operations and one sunday night kicked up a nice pile of bad cluster errors like you got. I could have had it back up in 2 days but warranty made me lose it like 2 weeks to get it replaced and restored. Seagate 1TB 7800 RPM HDD in this. It takes a dump again I'll hope it's after the expiration of the warranty, replace it with a 1TB or 2TB SSD at that point.

[Xeogred]

I'm a little confused, was your drive okay after that or you get it replaced?

Maybe I am going a little senile and didn't get it off Amazon. I've gotten some externals from Best Buy off and on over the years. I swear I thought I remember getting this one specifically off Amazon because it had a good price years ago, but maybe not. I can't find a recent or the box anymore.

[Tanooki]

Me? It was toast. It was around 10pm~ on a Sunday night about a month ago. Windows yet again got into 100% Disk mode, and I was like f-it and restarted. It never came back. I ended up forcing it into a command prompt and ran a specific checking/repairing chkdsk and after an hour it had coughed up a good dozen busted files and quite a few bad sectors and I just flipped the thing off and called it dead. Then by the middle of the week it was off to the west coast for replacement and a base windows 8.1 reinstall with drivers.

[Anapan]

From what I've heard, it's fairly rare that a consumer computer owner ever finds out about any bad sectors, but every single mechanical hard-drive has some.
Back when hard drives first came out they came with hand-written notes that told the operator of the bad sectors. Later, this list was communicated to the computer internally and those sectors were never written to. Some time in the 90s, hard drives had a pretty powerful CPU and as sectors went corrupt, the CPU and some non-volatile onboard memory worked together to find data that was hard to read (weak magnetic read) and it was re-written on the fly using a cache independant of the PC. If the data was hard to read again, the onboard CPU marks it bad and is re-written elsewhere (some space is never reported and is only used for this purpose) without the computer operator being informed.

A raw-data reading utility will typically report errors on all factory-new drives. How many is OK? I don't know. Once I suspect a drive of being faulty, I give anything 120 a *Bad!* and move my data to a new drive (ya, doesn't matter the size). Incidentally I did see a drive I marked "*bad!*" in my own hand writing again years later and it still sustained a full 100% write and read for a friend that I gave it to (it was a 3TB and is probably still 100% fine)... I read that they can keep working for a long time just marking sectors bad and moving data to the outer rim of the plattens? I am usually in charge of getting data back and I usually see them in a bad state. I still wouldn't trust my data to a perfectly fine seagate.

Seagate went bad many years ago as that's over 80% of my mechanical data recovery work. I certainly never use that brand anymore. All of mine have failed - over 8 TB of my own data lost. None made it over 2 years. Good thing I kept some backups on Western Digital of much of the important stuff.
That trick where you open them up while they're running and help the reader arm back on to the plattens would've saved over half of that data had I known (I recognise the sound now) . That shouldn't be a viable recovery solution, but it's saved many customers photos and videos. Inquiries for recovery from a clean-room recovery service are typically over $2K and all of my clients chose drive opening dust exposure or complete data loss (I'll open a drive for a 6-pack if they stay and drink with me - that kinda danger is exciting to me).

[Xeogred]

Yeah I'll probably avoid Seagate from here on.

I got my WD 6TB Black. I kind of want to get another one, which would give me leeway to move most of my stuff from my two externals (5TB total) to an internal drive. Only problem is I don't think I have an extra power cable available for another drive. I think there's still one open SATA slot, underneath my huge GPU haha.

I have a 1TB internal that was where I originally installed my Windows 7 OS, but then I transferred the OS over to a SSD. But I still think there are programs and such installed to that drive, so I've been weary of messing with it.

I use CrystalDiskInfo for checking drives health. Maybe there's similar programs that are worth looking at too?

It looks like my Seagate is reporting something else this morning, compared to above lol:

Reallocated Sector Count: 704
Current Pending Sector Count: 144
Uncorrectable Sector Count: 144

What's a little annoying on a side note is that I just discovered Windows Home Premium only supports up to 16GB of ram (my motherboard can do 32GB). But most other versions support up to 192GB. So... guess I shouldn't bother adding more ram with this rig.

[Tanooki]

Damnit that doesn't inspire confidence as I'm sure they put the same model/make seagate back in since it was warranty work.

Hopefully in about 2 years SSDs of the 1-2TB in size will be cheap as the HDD is currently. I'll replace it with something far less crappy, or if I did stick with older tech HDDs again back to WD as I've almost used them exclusively since the 90s.

[ElkinFencer10]

Damn, that's weird - Seagate is my go-to brand for hard drives. I've got six or seven Seagate hard drives if you include internals and externals with various systems, and I've never had a problem with it. I wonder if I'm unusually lucky with them or if you're just unlucky.

[Tanooki]

Christmas 1990 to present, only drive ever to die on me was a Seagate.

I've had 2 of their over time, 2 maxtor's, and everything else WD. We do have one portable seagate in the other room which has lasted at least 5 years, it's I think a 500GB drive.