I used to have RAID (or at least some variation of it) for my main storage. For redundancy, in case of disk failure. I started with some crazy LVM mirrors done on two disks of different size. Sync job was starting on every boot 😄

Then came time for RAID5 on mdadm + LVM for volume management. It was working nice until the moment when disks became bigger. Long array rebuilds or checks, required my PC to stay turned on overnight just to validate if stuff works still.

Then in work, I had my first contact with ZFS and it was love at first sight ❤️ Biggest improvement was, that after creation of pool there’s no need to re-sync whole disk. Both resilvering or scrub were much faster than standard mdadm check.

My old disks are 2TB big and almost each of them is different now. I started with 3 x ST2000DL003 2TB, but after failure of one I was unable to buy exact replacement. Closest possible was: ST2000DM001. Since that moment, I had disks running at variable speed around 5400RPM and one on 7200RPM. I decided to add another drive to the pool as spare, it was Samsung HD204UI. After it failed, I added WDC Black WD2003FYYS 2TB.

Today, oldest disk is around 10 years old, youngest are around 8. Just this is pretty awful but what pushed me to replace them, was actual lack of space 😄

I was hoping to find 4 x 8TB drives which will quadruple my capacity and should provide enough space and safety for years to come, but I couldn’t find any reasonably priced offers. I even bought 4 x Toshiba X300 8TB. I started copying data to them… And they started dying one by one 😞 Actually, they didn’t die, but SMART was reporting “pre-fail” condition, when they were working heavily for too long. What if I have to re-sync/scrub whole pool? Will they die in the middle. I returned all of them. It was first time I gave a chance to Toshiba. It was last time 😉

Then I saw a movie on Youtube or article - someone was buying at bulk cheap external HDDs and “shucking” them 1. It was pretty close to how I understand building RAID as “Redundant Array of Inexpensive Disks” :simple_smile: Those disk are usually pretty good disks - either WD RED which are actually designed for NAS/RAID workload, or so called “white labels” which are just variations of HGST drives (sometimes slower, with less cache, etc).

I noticed a promo on Amazon on WD Elements 14TB and I bought 3 of them. Wanted 4, but they were limiting orders. They were 30% cheaper than similar internal HDDs.

What’s inside?

I connected them as they’re provided, in those plastic coffins and started playing. First I stress tested them with badblocks 2 3.

Run badblocks
badblocks -b 4096 -wsv /dev/sde

After around 1 hour, all of them were quite warm. Having between 53~55°C. Theoretically those drives can operate in up to 60°C, but at the end of 2nd day, two disk turned off and I have to turn them off and on again to make them discoverable, which just reminds me:

As SMART didn’t notice anything worrying, I was quite sure it’s just because of temperature (coldest one stayed alive). I finished my torture testing here and decided to “shuck” them.

There’s really good instruction on how to do that on iFixit 4, so I won’t be explaining it here.

I ended up having 3 new WDC WD140EDGZ-11B1PA0 14TB 5400RPM disks, with average transfers around 170~200MB/s. What’s interesting SMART recognise them as 7200RPM, but I heard it’s not worth to trust it in case of WDC/HGST drives.

smartctl output
smartctl 7.2 2020-12-30 r5155 [x86_64-linux-5.13.0-21-generic] (local build)
Copyright (C) 2002-20, Bruce Allen, Christian Franke, www.smartmontools.org

Device Model:     WDC WD140EDGZ-11B1PA0
LU WWN Device Id: 5 000cca 2adcc7fcb
Firmware Version: 85.00A85
User Capacity:    14 000 519 643 136 bytes [14,0 TB]
Sector Sizes:     512 bytes logical, 4096 bytes physical
Rotation Rate:    7200 rpm
Form Factor:      3.5 inches
Device is:        Not in smartctl database [for details use: -P showall]
ATA Version is:   ACS-2, ATA8-ACS T13/1699-D revision 4
SATA Version is:  SATA 3.2, 6.0 Gb/s (current: 6.0 Gb/s)
Local Time is:    Fri Nov 12 15:59:23 2021 CET
SMART support is: Available - device has SMART capability.
SMART support is: Enabled

SMART overall-health self-assessment test result: PASSED

General SMART Values:
Offline data collection status:  (0x80)	Offline data collection activity
                    was never started.
                    Auto Offline Data Collection: Enabled.
Self-test execution status:      (   0)	The previous self-test routine completed
                    without error or no self-test has ever
                    been run.
Total time to complete Offline
data collection: 		(  101) seconds.
Offline data collection
capabilities: 			 (0x5b) SMART execute Offline immediate.
                    Auto Offline data collection on/off support.
                    Suspend Offline collection upon new
                    Offline surface scan supported.
                    Self-test supported.
                    No Conveyance Self-test supported.
                    Selective Self-test supported.
SMART capabilities:            (0x0003)	Saves SMART data before entering
                    power-saving mode.
                    Supports SMART auto save timer.
Error logging capability:        (0x01)	Error logging supported.
                    General Purpose Logging supported.
Short self-test routine
recommended polling time: 	 (   2) minutes.
Extended self-test routine
recommended polling time: 	 (1383) minutes.
SCT capabilities: 	       (0x003d)	SCT Status supported.
                    SCT Error Recovery Control supported.
                    SCT Feature Control supported.
                    SCT Data Table supported.

SMART Attributes Data Structure revision number: 16
Vendor Specific SMART Attributes with Thresholds:
  1 Raw_Read_Error_Rate     0x000b   100   100   001    Pre-fail  Always       -       0
  2 Throughput_Performance  0x0004   100   100   054    Old_age   Offline      -       0
  3 Spin_Up_Time            0x0007   092   092   001    Pre-fail  Always       -       0 (Average 328)
  4 Start_Stop_Count        0x0012   100   100   000    Old_age   Always       -       5
  5 Reallocated_Sector_Ct   0x0033   100   100   001    Pre-fail  Always       -       0
  7 Seek_Error_Rate         0x000a   100   100   001    Old_age   Always       -       0
  8 Seek_Time_Performance   0x0004   100   100   020    Old_age   Offline      -       0
  9 Power_On_Hours          0x0012   100   100   000    Old_age   Always       -       0
 10 Spin_Retry_Count        0x0012   100   100   001    Old_age   Always       -       0
 12 Power_Cycle_Count       0x0032   100   100   000    Old_age   Always       -       5
 22 Unknown_Attribute       0x0023   100   100   025    Pre-fail  Always       -       100
192 Power-Off_Retract_Count 0x0032   100   100   000    Old_age   Always       -       5
193 Load_Cycle_Count        0x0012   100   100   000    Old_age   Always       -       5
194 Temperature_Celsius     0x0002   062   062   000    Old_age   Always       -       25 (Min/Max 20/27)
196 Reallocated_Event_Count 0x0032   100   100   000    Old_age   Always       -       0
197 Current_Pending_Sector  0x0022   100   100   000    Old_age   Always       -       0
198 Offline_Uncorrectable   0x0008   100   100   000    Old_age   Offline      -       0
199 UDMA_CRC_Error_Count    0x000a   100   100   000    Old_age   Always       -       0

SMART Error Log Version: 1
No Errors Logged

SMART Self-test log structure revision number 1
No self-tests have been logged.  [To run self-tests, use: smartctl -t]

SMART Selective self-test log data structure revision number 1
    1        0        0  Not_testing
    2        0        0  Not_testing
    3        0        0  Not_testing
    4        0        0  Not_testing
    5        0        0  Not_testing
Selective self-test flags (0x0):
  After scanning selected spans, do NOT read-scan remainder of disk.
If Selective self-test is pending on power-up, resume after 0 minute delay.

That’s enough for today. Next time, I will build ZFS array with them.