How bad can your SMART numbers be on a used HD before you refuse it?

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

Scott Laird

Active Member
Aug 30, 2014
312
144
43
I ordered a batch of 7 HUS726060ALS641 6T drives off of eBay. They arrived slightly padded in a bulging Priority Mail box. Of the 7, 4 failed to even spin up, and the 3 remaining drives have impressively large error and byte counts. This one has ~6.4 million corrected errors and 910 TB read:

Code:
# smartctl -a /dev/sdf
smartctl 7.1 2019-12-30 r5022 [x86_64-linux-5.4.0-33-generic] (local build)
Copyright (C) 2002-19, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Vendor:               HGST
Product:              HUS726060ALS641
Revision:             B283
Compliance:           SPC-4
User Capacity:        6,001,175,126,016 bytes [6.00 TB]
Logical block size:   512 bytes
LU is fully provisioned
Rotation Rate:        7200 rpm
Form Factor:          3.5 inches
Logical Unit id:      0x5000cca23223312c
Serial number:        xxxxxxx
Device type:          disk
Transport protocol:   SAS (SPL-3)
Local Time is:        Mon Jun  8 00:37:49 2020 UTC
SMART support is:     Available - device has SMART capability.
SMART support is:     Enabled
Temperature Warning:  Enabled

=== START OF READ SMART DATA SECTION ===
SMART Health Status: OK

Current Drive Temperature:     31 C
Drive Trip Temperature:        85 C

Manufactured in week 43 of year 2014
Specified cycle count over device lifetime:  50000
Accumulated start-stop cycles:  21
Specified load-unload count over device lifetime:  600000
Accumulated load-unload cycles:  1380
Elements in grown defect list: 0

Vendor (Seagate Cache) information
  Blocks sent to initiator = 15151315965444096

Error counter log:
           Errors Corrected by           Total   Correction     Gigabytes    Total
               ECC          rereads/    errors   algorithm      processed    uncorrected
           fast | delayed   rewrites  corrected  invocations   [10^9 bytes]  errors
read:    5124364       15         0   5124379    3519750     910485.278           0
write:         0        0         0         0    1398994      37177.230           0
verify:        0        0         0         0      32822          0.000           0

Non-medium error count:        2
Does this look acceptable to anyone? I'm going to return the 4 failed drives, and I'm strongly leaning towards returning the 3 "good" drives as well.
 
  • Like
Reactions: Samir

EffrafaxOfWug

Radioactive Member
Feb 12, 2015
1,394
511
113
Unless you were explicitly informed of their FUBAR-ness beforehand, I wouldn't consider them acceptable in any way, shape or form.

My only regret in returning them would be knowing that they're likely to keep getting re-sold until they find a buyer that doesn't know how to run smartctl.
 
  • Like
Reactions: Samir

Whaaat

Active Member
Jan 31, 2020
301
157
43
hdd seems to be in perfect shape and has only 37 TB written during 5 years of operation. This is nothing compared with typical annualized workload rate of 550 Tb/year for enterprise class hdd
 
  • Like
Reactions: Samir

vl1969

Active Member
Feb 5, 2014
634
76
28
Unless you were explicitly informed of their FUBAR-ness beforehand, I wouldn't consider them acceptable in any way, shape or form.

My only regret in returning them would be knowing that they're likely to keep getting re-sold until they find a buyer that doesn't know how to run smartctl.
Well my issue would be not ,not able to run smartctl, but to Interpret the results
 
  • Like
Reactions: Samir

nephri

Active Member
Sep 23, 2015
541
106
43
45
Paris, France
So from my perspective

I'm really aware from the "Elements in grown defect list"
A no-zero value is a no go for me. In your case it's OK.

For the high corrected ECC read values, i would perform such things like badblocks (write+read) and rerun a long smartctl to see how theses values progresses.
For instance theses errors never output any I/O error into the host. so in a point of view of the computer, this HDD didn't give any error.

But the "Non-medium error count" to 2 is more suspect to me ! It's things i didn't like....
 
  • Like
Reactions: Samir

pricklypunter

Well-Known Member
Nov 10, 2015
1,708
515
113
Canada
Those disks really are shot, send them back and explain to the seller that they all tested faulty. The errors you are seeing are not going to get any better, no matter how many times you run badblocks and smartctl. The dangerous scenario is having errors on a disk that the OS doesn't know about, or tries to work around, that's when you lose critical data. It's simply not worth the effort and pain :)
 
  • Like
Reactions: Samir

UhClem

just another Bozo on the bus
Jun 26, 2012
433
247
43
NH, USA
Nothing in those stastics indicates any problem. For HDDs, "errors corrected by ECC" are totally normal. Because of the bit density, and track density, along with rotation/vibration/servos, the extensive ECC used is a critical part of the design. Do not jump to a wrong conclusion based solely on the word "error" without fully understanding the context--maybe they should have written "ECC recoveries" instead. [And, while 5.12 million seems like a lot, it is only one for every 350K LBAs read (on that disk).]

[Keep in mind that all of those drive statistics are resettable.]

If you want to return the 3 good drives, don't do it for the wrong reason. Assuming that the vendor did not send you 4 "dead" drives, and that their demise was due to the inadequate packaging, you would be justified in returning all 7, since the other 3 incurred the same shock/stress, and you should not bear the "future risk" for that.
 
  • Like
Reactions: beren and Samir

Samir

Post Liker and Deal Hunter Extraordinaire!
Jul 21, 2017
3,257
1,445
113
49
HSV and SFO
When something is insufficiently packed like that, that's the first thing that makes me mad--because now there's a good chance the goods are damaged or will need a damage claim, etc.--a unnecessary hassle since packing material isn't expensive have to be fancy. I've had things arrive wrapped in newspapers perfectly fine and things just throw into a box with a sheet of bubble wrap bounce its way to me. :eek:

Usually, cases like this is where the seller can afford to either eat it or insurance will cover it, and more than likely you will still be able to keep the goods, so I'd press for a refund. If I get free drives in this shape, I'm never upset. ;)
 

Evan

Well-Known Member
Jan 6, 2016
3,346
598
113
4 dead drives I guess that box has been dropped , I would just return them all.
 
  • Like
Reactions: Samir