Question 10TB HGST SMART warning - bad drive or power issue?

[Shmee]

Hello, I recently received a refurb 10TB HGST data center drive, from Amazon/GoHDD. I was testing it earlier, by plugging it in to USB 3.1 dock, which is 12V DC powered by AC adapter. Good news, the drive shows up in storage management/device manger, and it spins up. Bad news, it fails to initialize GPT. I checked the SMART data, and it has a warning status. From what I can tell, a bunch of bad sectors, specifically Uncorrected Sector Count, IIRC.

Basically, I would just like to get a second opinion on this drive. It seems to me that it is not healthy, with the bad sectors, and that explains it not initializing. So this is return / warranty time. I just wanted to be sure I am not missing something. I don't think the adapter could be the issue, as the drive does seem to spin up, is detected, and I can read its SMART data. But I wanted to be sure, as I know these drives come with the special power off feature, and sometimes do not work properly with some desktop PSUs, what with the drives being designed for datacenter. Thoughts?

 

[mikeymikec]

Disclaimer: I don't know anything about enterprise drives.

Post the SMART stats? Though if the drive is in warranty and SMART stats say nope, chances are this is a simple warranty issue.

 

[igor_kavinski]

I was testing it earlier, by plugging it in to USB 3.1 dock

Plug it directly into a SATA port to be sure.

 

[manly]

Doesn't sound like anything to do with power disable, but you can block off PIN 3 with kapton or electrical tape. All reports are that goHardDrive is very amenable to an RMA. I guess you do have to pay 1-way shipping.

You could try some Linux-fu as well (sgdisk -Z), but as always, be careful.

 

[Shmee]

Here is the SMART screenshot for the drive.

 

[manly]

I would tell goHardDrive that there are a high number of errors in the SMART data, and that you would like them to provide a call tag for RMA.

It looks like the raw count is 30, which isn't a small number. We have no way of knowing what the progression is (i.e. you might be okay if it's been at 30 and stays there).

My search-fu says that this counter incrementing is fairly predictive of an unhealthy drive, so I wouldn't want to chance it.

https://kb.acronis.com/content/9134?ckattempt=1

 

[bigboxes]

You just bought it from Amazon. I'd just return it to Amazon. You shouldn't have to pay return shipping for a defective part. I'm assuming you're within 30 days. Have them send you out a healthy drive. That's what you're paying for. I'd first try it in a SATA port on the mobo first, just to make sure it's not the dock causing the issues.

 

[mikeymikec]

Here is the SMART screenshot for the drive.

That's bizarre but IMO a slam-dunk of an RMA job. I'd like to know under what circumstances that 'uncorrectable sector count' occurs, and I didn't get much joy from a google search; it gave me the answer I was logically aware of in that when a drive runs out of reserve sectors as a result of correcting many bad sectors then any further ones would be uncorrectable, but your drive doesn't have any corrected or pending sectors, it just went to a bad sector handling fail.

 

[aigomorla]

@Shmee

The controller on the drive is either toasted, failed to get its SMART info resetted properly, or the wrong firmware was probably loaded.

Does it make clicking noise? I noticed that is also a fair source of debugging if it clicks.

Eitherway, how old is the drive?
I would think of RMAing to back to the vendor, given that most He drives have a 3 yr refurbised warrenty unless you bought it from a really bad vendor.

 

[Shmee]

Yeah I will be returning it.

 

[fzabkar]

I notice that the drive has recorded 36,214 hours, but has only lost 5 points from the normalised Current value (5 = 100 - 95). At this rate, it is projected to hit the threshold at 82 years.

Weird.

 

[aigomorla]

I notice that the drive has recorded 36,214 hours, but has only lost 5 points from the normalised Current value (5 = 100 - 95). At this rate, it is projected to hit the threshold at 82 years.

Weird.

Nah, its a bad reset on the SMART, which is probably what shmee see.

Hence why i recommend him just returning it.

 

[fzabkar]

Another weird POH count:

https://forums.tomshardware.com/thr...t-values-within-reason.3633149/#post-21898678

 

[jamesdsimone]

When I got my 12TB drives they shipped with special SATA power cables. Not sure exactly what they were needed for. I quickly looked but my drives worked fine so I didn't pay much attention but I can dig them up and check.

 

[Shmee]

When I got my 12TB drives they shipped with special SATA power cables. Not sure exactly what they were needed for. I quickly looked but my drives worked fine so I didn't pay much attention but I can dig them up and check.

As mentioned, earlier and in the information manuals from GO HDD, these enterprise drives have a power off feature that sometimes is not compatible with certain desktop PSUs, so they have these SATA power adapters to mitigate the issue. Sounds like your 12TB drives worked fine though, so no issues.

Just to clarify I also am using 12TB drives, in my NAS, and they work fine, all 4 of them. I have since returned the 10TB with the SMART errors.

 

[igor_kavinski]

so they have these SATA power adapters

Is there a special name for them?