[Updated] 6 Lots 5x 400GB SanDisk Optimus SAS 6Gbps SSDs $240/lot

arglebargle

H̸̖̅ȩ̸̐l̷̦͋l̴̰̈ỏ̶̱ ̸̢͋W̵͖̌ò̴͚r̴͇̀l̵̼͗d̷͕̈
Jul 15, 2018
656
233
43
Edit: Seller listed 6 new lots of 5x drives/ea:

LOT OF 5 - SanDisk Optimus SSD TXA2D2 400GB 2.5" | eBay

Performance on these is really good for 400GB SAS2 drives.

Edit:
These drives are pulls from M$ Azure StorSimple appliances (on-prem Azure storage buckets) so wear is going to vary depending on the use of the appliance your drives came from. I got *exceedingly* lucky with my first batch, they had less than 9TWB and less than 6mo power on hours. Your mileage there will vary heavily.

If anyone works with one of these and wouldn't mind trying to pull a current firmware from one of the drives could you PM me? I'd love to get my hands on the most recent release.

---- Original Post:

This looks like a pretty decent price for some smaller sized drives. If you're looking to fill up a bunch of un-used 2.5" bays with flash this isn't a bad option at ~48$/drive. They're a little older but I highly doubt they've had 7+PB written to them during their time in service.

--

Warranted for 10dwpd x 5Yr (~7.3PB)
500MB/s Sequential R/W
100k/50k IOPS

Storage review liked them: SMART Storage Systems Optimus SAS Enterprise SSD Review | StorageReview.com - Storage Reviews

Got a batch of them in today, they're in excellent shape:

Code:
Manufactured in week 49 of year 2013
 Specified cycle count over device lifetime:  0
 Accumulated start-stop cycles:  129
 Elements in grown defect list: 0

 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        0         0         0          0      84447.642           0
 write:         0        0         0         0          0       9945.846           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                   -   26220                 - [-   -    -]
 # 2  Default           Completed                   -      78                 - [-   -    -]
 
Last edited:

BLinux

cat lover server enthusiast
Jul 7, 2016
2,525
971
113
artofserver.com
i also bought several batches of these and received them yesterday. although the SMART data looked great on all of them:

- these are SAS drives, so SMART data is not the same as SATA drives. but looking at the error counters, I saw no corrected or uncorrected errors at all.
- power on hours was > 33K on most of them, so they've been in use for a few years.
- GB written varied a lot. some were very low, like < 1 DW in total, some were higher.
- endurance used metric was between 0%~3%

All that said, out of the 20 units I got, 4 of them seem to have problems even though the SMART data didn't indicate an issue. When I tested them out, I got errors like:

Code:
[ 4499.250172] sd 1:0:7:0: [sdp] FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 4499.250187] sd 1:0:7:0: [sdp] Sense Key : Hardware Error [current]
[ 4499.250195] sd 1:0:7:0: [sdp] Add. Sense: Internal target failure
[ 4499.250201] sd 1:0:7:0: [sdp] CDB: Read(10) 28 00 00 00 00 00 00 00 20 00
[ 4499.250205] blk_update_request: critical target error, dev sdp, sector 0
[ 4499.250531] sd 1:0:7:0: [sdp] FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 4499.250544] sd 1:0:7:0: [sdp] Sense Key : Hardware Error [current]
[ 4499.250551] sd 1:0:7:0: [sdp] Add. Sense: Internal target failure
[ 4499.250557] sd 1:0:7:0: [sdp] CDB: Read(10) 28 00 00 00 00 00 00 00 08 00
[ 4499.250561] blk_update_request: critical target error, dev sdp, sector 0
[ 4499.250611] Buffer I/O error on dev sdp, logical block 0, async page read
[ 4499.250832] sd 1:0:7:0: [sdp] FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 4499.250837] sd 1:0:7:0: [sdp] Sense Key : Hardware Error [current]
[ 4499.250842] sd 1:0:7:0: [sdp] Add. Sense: Internal target failure
[ 4499.250846] sd 1:0:7:0: [sdp] CDB: Read(10) 28 00 00 00 00 80 00 00 08 00
[ 4499.250849] blk_update_request: critical target error, dev sdp, sector 128
[ 4499.251081] sd 1:0:7:0: [sdp] FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 4499.251087] sd 1:0:7:0: [sdp] Sense Key : Hardware Error [current]
[ 4499.251092] sd 1:0:7:0: [sdp] Add. Sense: Internal target failure
[ 4499.251097] sd 1:0:7:0: [sdp] CDB: Read(10) 28 00 00 00 00 80 00 00 08 00
[ 4499.251100] blk_update_request: critical target error, dev sdp, sector 128
[ 4499.251148] Buffer I/O error on dev sdp, logical block 16, async page read
[ 4499.251387] sd 1:0:7:0: [sdp] FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 4499.251392] sd 1:0:7:0: [sdp] Sense Key : Hardware Error [current]
[ 4499.251396] sd 1:0:7:0: [sdp] Add. Sense: Internal target failure
[ 4499.251400] sd 1:0:7:0: [sdp] CDB: Read(10) 28 00 2e 93 90 00 00 00 08 00
[ 4499.251403] blk_update_request: critical target error, dev sdp, sector 781422592
[ 4499.251593] sd 1:0:7:0: [sdp] FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 4499.251599] sd 1:0:7:0: [sdp] Sense Key : Hardware Error [current]
[ 4499.251604] sd 1:0:7:0: [sdp] Add. Sense: Internal target failure
[ 4499.251608] sd 1:0:7:0: [sdp] CDB: Read(10) 28 00 2e 93 90 00 00 00 08 00
[ 4499.251611] blk_update_request: critical target error, dev sdp, sector 781422592
[ 4499.251664] Buffer I/O error on dev sdp, logical block 97677824, async page read
[ 4499.251866] sd 1:0:7:0: [sdp] FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 4499.251873] sd 1:0:7:0: [sdp] Sense Key : Hardware Error [current]
[ 4499.251879] sd 1:0:7:0: [sdp] Add. Sense: Internal target failure
[ 4499.251884] sd 1:0:7:0: [sdp] CDB: Read(10) 28 00 2e 93 90 a0 00 00 08 00
[ 4499.251890] blk_update_request: critical target error, dev sdp, sector 781422752
[ 4499.251942] Buffer I/O error on dev sdp, logical block 97677844, async page read
[ 4499.252108] sd 1:0:7:0: [sdp] FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 4499.252114] sd 1:0:7:0: [sdp] Sense Key : Hardware Error [current]
[ 4499.252118] sd 1:0:7:0: [sdp] Add. Sense: Internal target failure
[ 4499.252124] sd 1:0:7:0: [sdp] CDB: Read(10) 28 00 2e 93 90 a8 00 00 08 00
[ 4499.252127] blk_update_request: critical target error, dev sdp, sector 781422760
[ 4499.252179] Buffer I/O error on dev sdp, logical block 97677845, async page read
[ 4499.252691] sd 1:0:7:0: [sdp] FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 4499.252702] sd 1:0:7:0: [sdp] Sense Key : Hardware Error [current]
[ 4499.252708] sd 1:0:7:0: [sdp] Add. Sense: Internal target failure
[ 4499.252713] sd 1:0:7:0: [sdp] CDB: Read(10) 28 00 2e 93 90 a8 00 00 08 00
[ 4499.252717] blk_update_request: critical target error, dev sdp, sector 781422760
[ 4499.252769] Buffer I/O error on dev sdp, logical block 97677845, async page read
[ 4499.252951] sd 1:0:7:0: [sdp] FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 4499.252959] sd 1:0:7:0: [sdp] Sense Key : Hardware Error [current]
[ 4499.252965] sd 1:0:7:0: [sdp] Add. Sense: Internal target failure
[ 4499.252971] sd 1:0:7:0: [sdp] CDB: Read(10) 28 00 00 00 00 00 00 00 08 00
[ 4499.252974] blk_update_request: critical target error, dev sdp, sector 0
[ 4499.253020] Buffer I/O error on dev sdp, logical block 0, async page read
[ 4499.253220] Buffer I/O error on dev sdp, logical block 0, async page read
[ 4499.253397] Buffer I/O error on dev sdp, logical block 0, async page read
[ 4500.995556] Buffer I/O error on dev sdp, logical block 97677824, async page read
[ 4501.044536] sd 1:0:7:0: [sdp] Unit Not Ready
[ 4501.044544] sd 1:0:7:0: [sdp] Sense Key : Hardware Error [current]
[ 4501.044549] sd 1:0:7:0: [sdp] Add. Sense: Internal target failure
[ 4501.046715] Dev sdp: unable to read RDB block 0
The seller accepted return for refund. But I would suggest testing them out a bit more thoroughly and not just rely on the SMART data.
 

arglebargle

H̸̖̅ȩ̸̐l̷̦͋l̴̰̈ỏ̶̱ ̸̢͋W̵͖̌ò̴͚r̴͇̀l̵̼͗d̷͕̈
Jul 15, 2018
656
233
43
Hmm, fortunately the batch that I got didn't have any issues like that. I've definitely received "tested, working 100%!!" batches with 1-2 bad drives in the past (not these) so yeah, I would absolutely run some stress tests against them when you get them in the mail.

For convenience I've just been running "badblocks -b 32768 -wsv -p 5 /dev/sdX" against all of mine. That's the badblocks write test, 2M sized write transactions, run until 5 passes (4 patterns, full drive write per pattern, per pass) with 0 errors. On a healthy drive that'll be ~20 full drive writes with zero errors occurring, on a bad drive the drive will do it's best not to kill itself before you stop the test by hand. Either way it'll be obvious that the drive is good, or bad. On a 400GB drive that takes around 3 hours per "pass" so the test should be complete in around 15-16hrs.

Re: testing SAS SSDs -- watch for increasing numbers of grown defects (bad sectors). I typically have "watch smartctl" running on a drive (or drives) while I'm badblocks testing it/them using "-AH" to look at health status (which will trip if the drive firmware detects health status issues) and grown defects:

Code:
root@bigboy:~# smartctl -AH -dscsi /dev/sdf
smartctl 7.0 2018-12-30 r4883 [x86_64-linux-4.18.0-20-generic] (local build)
Copyright (C) 2002-18, Bruce Allen, Christian Franke, www.smartmontools.org
=== START OF READ SMART DATA SECTION ===
SMART Health Status: OK
Current Drive Temperature:     38 C
Drive Trip Temperature:        75 C
Elements in grown defect list: 2
That's a drive from a batch of STec I purchased last month, it has two grown defects over its 5 year service but is stable and hasn't grown more over 20 drive passes with badblocks. That batch (of STec) did have one bad drive that I sent back to the seller, it grew an additional 4k grown defects on the first badblocks pass and started throwing "imminent failure" SMART health warnings.

You can also watch the error log too:
Code:
root@bigboy:~# smartctl -l error -dscsi /dev/sdh
smartctl 7.0 2018-12-30 r4883 [x86_64-linux-4.18.0-20-generic] (local build)
Copyright (C) 2002-18, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF READ SMART DATA SECTION ===
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        0         0         0          0      88314.045           0
write:         0        0         0         0          0      13394.528           0

Non-medium error count:        0
Edit:
Oh BTW... yes, performance is really good on these for 6Gbps... easily 500MB/sec read and write.
Performance on these drives is *great* for what they are. What blew me away is that they don't just have high performance peaks for a short duration during a benchmark, they perform at that level constantly 24/7 with very little deviation.

Edit:
"smartctl -x -dscsi /dev/sdX" includes all useful drive information too, I've been dumping both that and with the -j flag before I start stressing a drive and after then diffing them to look for anything fishy.

Edit:
These drives are pulls from M$ Azure StorSimple appliances (on-prem Azure storage buckets) so wear is going to vary depending on the use of the appliance your drives came from. I got *exceedingly* lucky with my first batch, they had less than 9TWB and less than 6mo power on hours. Your mileage there will vary heavily.

If anyone works with one of these and wouldn't mind trying to pull a current firmware from one of the drives could you PM me? I'd love to get my hands on the most recent release.
 
Last edited: