Stress testing used hard drives

Introduction

I got a batch of used 18 GB SCSI hard drives this week, and wanted to test each drive to see if they were free of errors and still usable. I had to outfit each drive with a Sun spud bracket prior to installing them in a 6-bay Sun Multipack. Once the drives were installed, I used the following while loop to continuosly write 1 Mb blocks to disk drives attached to my Solaris 10 server:

while :
do
    dd if=/dev/zero of=/dev/rdsk/c1t6d0s2 bs=1048576 seek=1
done

I allowed this loop to run 20 times per drive, and used the smartctl utility to check the health of each drive once the tests completed:

$ smartctl -a /dev/rdsk/c1t2d0s2 
smartctl version 5.33 [sparc-sun-solaris2.9] Copyright (C) 2002-4 Bruce Allen
Home page is http://smartmontools.sourceforge.net/

Device: SEAGATE  SX318203LC       Version: B90B
Serial number: LR869054    102424W2
Device type: disk
Local Time is: Thu Jun 30 19:56:54 2005 EDT
Device supports SMART and is Enabled
Temperature Warning Disabled or Not Supported
SMART Health Status: OK
Vendor (Seagate) cache information
  Blocks sent to initiator = 220889811
  Blocks received from initiator = 118918432
  Blocks read from cache and sent to initiator = 8153478
  Number of read and write commands whose size <= segment size = 1354214
  Number of read and write commands whose size > segment size = 46873
Vendor (Seagate/Hitachi) factory information
  number of hours powered up = 65.45
  number of minutes until next internal SMART test = 106

Error counter log:
           Errors Corrected by           Total   Correction     Gigabytes    Total
               EEC          rereads/    errors   algorithm      processed    uncorrected
           fast | delayed   rewrites  corrected  invocations   [10^9 bytes]  errors
read:       4621        0         0      4621       4621        135.662           4
write:         0        0         0         6          6         68.820           0
verify:        0        0         0         0          0         15.207           0

Non-medium error count:        0

Error Events logging not supported

[GLTSD (Global Logging Target Save Disable) set. Enable Save with '-S on']
Device does not support Self Test logging

Smartctl detected several uncorrectable errors on disk c1t2d0s2, so I decided to return that disk to the individual I purchased it from.

Conclusion

Only two disk drives in the batch of drives I purchased failed my tests, and luckily I was granted a refund for the defective hardware. Smartmontools is a great piece of software for proactively detecting failed disk drives, and is freely available for all to use. If you have questions or comments on the article, please feel free to E-mail the author.

References