Seagate 3TB bad sectors, advice?

Howdy,

CrystalDiskInfo showed some warnings this morning regarding my ST3000DM001, think it was regarding the "Current Pending Sector Count".

Long story short, I ran SeaTools in DOS and the Long test found bad sectors and was able to repair them.


--------------------

--------------------


CrystalDisk after long test:


I was expecting to see the value of "Reallocated Sectors Count" to increase after the repairs done by SeaTools. (Didn't it basically reallocate the bad sectors it caught?)

I've already backed up my data but can anybody with previous experience shed some light on the situation here? What appears to be the problem and what should I look for under SMART values? Why the "Reallocated Sectors Count" is still at 0?

Thanks,

I would RMA the drive. You could hold onto the disk and see if the bad sector count is increasing. Theoretically, if the count stays the same and the other attributes don't show signs of getting worse, you could use the drive with some confidence. However, I suggest you don't take that risk unless the data on it is worthless.

EDIT: To clarify, SMART is showing quite a lot failing at once: Read error rate, seek error rate, and reported uncorrectable errors.

Thanks for the clarification thideras

Think I'll keep using the drive until it gives its last breath, I'm certainly not leaving anything critical on it.

I just noticed that HDDExpert shows a warning while CrystalDiskInfo doesn't


I'll be monitoring the values for the next 7 days to see if they increase or remain the same.

Any idea what should I keep an eye for? "Reported Uncorrectable Errors"? Or basically all of the attributes that report 'errors' in general?

Thanks,

Different thresholds between the programs. If you installed HDTune, it'd probably flag other options, as well.

Any of the ones I listed are Bad Bad Bad. You could try another cable, but I really think the drive is going out, at this point.

I can't speak to the other errors that thideras commented on but the reallocated sector count is the one I hang my hat on. I have an ancient 200GB Seagate drive that went from zero to 16 (IIRC) very early in life and never changed since. I think it has about of 7 years of service now. I had a 2TB Seagate that had a bunch and the count went up a couple times/week. This was a drive in a remote system that started up and ran a couple hours on a daily basis. The number continued to grow to over a thousand and at some point SMART reported that failure was imminent (or something similar.) At that point I retired it as it was out of warranty and replaced it with a WD drive.

It was in a RAID1 setup (mirror) and on a secondary backup system so I could afford to watch it degrade. At no time did it seem to malfunction from the viewpoint of the OS nor was it ever dropped from the RAID.

On most of my systems the reallocated sector count remains at zero.

I'm surprised people still buy seagate products

I did some reading and here are my findings:

1) Seek Error Rate: This attribute is not really critical and the data is formatted in a certain way (at least for Seagate). (link)

According to that, my "Seek Error Rate" Hex value is: 00000158B54F, which computes into 0 Seek errors within 22590799 seeks.

2) Read Error Rate: The raw value of this represents 'sector count' and not 'error count' (link)

3) Reported Uncorrectable Errors: Couldn't find much information about this, wikipedia says: The count of errors that could not be recovered using hardware ECC (see attribute 195).

Conclusion: Some of those attributes are totally nonsense in Decimal, also the equations/meaning of each value probably differs per manufacturer.

This doesn't change the fact that I won't be relying on this hdd anymore. Thanks everyone for the helpful comments

freeagent said:

I'm surprised people still buy seagate products

Click to expand...

I don't any more. At least not spinning rust. I did buy a Seagate SSD not too long ago.

I bought a couple 2TB drives several years ago based on the reliability that they earned with their older Barracuda drives. That turned out to be not so smart.

OTOH one of the Deskstars (Deathstar?) crushed IBM's reputation years ago with the 75GXP and these days under the HGST name (and owned by WD) that line shows one of the best reliability rates (according to one large user.)

If Seagate tightens up their QC and WD gets sloppy, this could change again.

At the moment I'm buying HGST and WD drives. Too bad Samsung sold the business.

And it finally died after a month and 3 weeks later. Mad head clicking happening, taking it out off the system right now.

During this period, SMART values kept getting worse, especially in the last couple of days, the numbers on uncorrectable sectors and some other attributes were increasing like never before.

Power on hours: 12501. That's a little short of 1.5 years, Seagate really is bad. This is the second hdd from them that dies this quickly. I have a 2TB Green WD drive in my system with 33819 hours, almost 4 years.. And it's still going like a champ. Have to say though I was hammering the Seagate drive with games, while the green is only for media storage.