Checking drives before putting into “production”
See also Even ”Enterprise Grade” Drives Fail.
According to Tim Standing of SoftRAID, initial verification of a drive can preclude a large percentage of relatively early drive failures.
While SoftRAID has a certify command, it does not graph the performance, and so I would add this point: in my experience, drives that show aberrant behavior are also more likely to fail, even if they pass certification.
The diglloydTools DiskTester fill-volume command can test 99% of the drive and graph the behavior, as shown below, where 5 samples were tested simultaneously, and then graphed together to verify consistent performance—important for RAID setups.
Over the years I have found that aberrant performance behaviors (obvious in a graph) are often an excellent predictor of flaky drives. The test-reliability command is good too, with the major benefit of being able to operate on in-use drives—no need to take a system down for days to certify (which means having to completely wipe) the drives.
Show below is drive performance for five samples of the HGST 8TB Ultrastar He8 hard drive. All samples deliver the same pattern of declining performance as the drive fills up—exactly as expected. Drives with excessive remapped sections tend to show weird spikes in speed in the wrong places. This is bad for RAID performance as well (when drives do not perform the same in the same area).
