HUH728080AL5201 (HGST 8TB SAS Helium) Slooooooow writes (ebay purchase)

[Mithril]

Picked up a few of these from ebay, doing my normal `run badblocks on HDD from ebay` thing and I noticed they seemed to be crawling. I checked the speed and 2 drives were writing at ~7MB and 2 at ~25MB. Even doing a DD to the disk is slower than it should be at ~77MB, reads seem fine at over 200MB.


I've never seen this behavior before in new or used drives, not seeing uncorrected errors in what smart data does come up, an example:

=== START OF INFORMATION SECTION ===
Vendor:               HGST
Product:              HUH728080AL5201
Revision:             B7J0
Compliance:           SPC-4
User Capacity:        8,001,563,222,016 bytes [8.00 TB]
Logical block size:   512 bytes
Physical block size:  4096 bytes
LU is fully provisioned
Rotation Rate:        7200 rpm
Form Factor:          3.5 inches
Logical Unit id:     
Serial number:       
Device type:          disk
Transport protocol:   SAS (SPL-3)
Local Time is:        Mon May 16 20:59:32 2022 PDT
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:     28 C
Drive Trip Temperature:        85 C

Accumulated power on time, hours:minutes 40380:28
Manufactured in week 23 of year 2016
Specified cycle count over device lifetime:  50000
Accumulated start-stop cycles:  25
Specified load-unload count over device lifetime:  600000
Accumulated load-unload cycles:  844
Elements in grown defect list: 0

Vendor (Seagate Cache) information
  Blocks sent to initiator = 33224553580724224

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:          0        6         0         6    1913184      77027.823           0
write:         0        0         0         0    1412200     138881.252           0
verify:        0       11         0        11     411586     167278.153           0

Non-medium error count:        0

 

[UhClem]

Try sdparm --set=WCE --save /dev/sdX

 

[Mithril]

I've currently got them in a TrueNAS box running a long self test, checking for the equivalent command...

Darn, checking with "camcontrol modepage /dev/da1 -m 0x08 -e" it looks like thats enabled on all the drives

IC:  0
ABPF:  0
CAP:  0
DISC:  0
SIZE:  0
WCE:  1
MF:  0
RCD:  0
Demand Retention Priority:  0
Write Retention Priority:  0
Disable Pre-fetch Transfer Length:  65535
Minimum Pre-fetch:  0
Maximum Pre-fetch:  65535
Maximum Pre-fetch Ceiling:  65535
FSW (Force Sequential Write):  0
LBCSS (Logical Block Cache Segment Size):  0
DRA (Disable Read-Ahead):  0
Number of Cache Segments:  8
Cache Segment Size:  0

 

[UhClem]

...
Darn, checking with "camcontrol modepage /dev/da1 -m 0x08 -e" it looks like thats enabled on all the drives

...
WCE:  1
...

Yeah, too bad. That's the typical low-hanging fruit.

 

[Mithril]

Really really weird. I did a short and long self test on all the drives, one did fail. Out of curiosity I did a quick DD write test and now it seems like all the drives are performing more as expected. I'm kicking off a new run of badblocks on all of them. I know that SAS drives and smart info vary quite a bit. Short test passed after writing to the first part of the drive, but no increase in what I think is this drive's "bad sector count"

=== START OF INFORMATION SECTION ===
Vendor:               HGST
Product:              HUH728080AL5201
Revision:             B7J0
Compliance:           SPC-4
User Capacity:        8,001,563,222,016 bytes [8.00 TB]
Logical block size:   512 bytes
Physical block size:  4096 bytes
LU is fully provisioned
Rotation Rate:        7200 rpm
Form Factor:          3.5 inches
Logical Unit id:     
Serial number:       
Device type:          disk
Transport protocol:   SAS (SPL-3)
Local Time is:        Wed May 18 10:11:21 2022 PDT
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:     29 C
Drive Trip Temperature:        85 C

Accumulated power on time, hours:minutes 40417:39
Manufactured in week 23 of year 2016
Specified cycle count over device lifetime:  50000
Accumulated start-stop cycles:  25
Specified load-unload count over device lifetime:  600000
Accumulated load-unload cycles:  846
Elements in grown defect list: 0

Vendor (Seagate Cache) information
  Blocks sent to initiator = 33236690151669760

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:          0        6         0         6    1913191      77027.823           0
write:         0        0         0         0    1426042     138976.081           0
verify:        0       11         0        11     411586     167278.153           0

Non-medium error count:        0

SMART Self-test log
Num  Test              Status                 segment  LifeTime  LBA_first_err [SK ASC ASQ]
     Description                              number   (hours)
# 1  Background short  Completed                   -   40417                 - [-   -    -]
# 2  Background long   Failed in segment -->       3   40381                 - [0x1 0x5d 0xfd]
# 3  Background short  Failed in segment -->       3   40380                 - [0x1 0x5d 0xfd]

 

[Mithril]

Update: Drives seem to be chugging along now, once the badblocks verification is over I'll run a latency scan on the disks to see if the "grown defect list" actually reflects the presence or lack of reallocated sectors for these disks.