I have two drives attached to the same Mac mini via USB, where a secure erase using Diskutil (which writes 3 times to the whole disc) seems to work consistently with no errors on either disc, however a SoftRAID Certify on either disc results quite quickly in errors like this:
Jan 30 07:52:18 - SoftRAID Application: The certify disk command for disk disk6, SN: 0123456789ABCDEF, USB3 bus 0, id 0, lun 0 hung while writing (offset 59,835,940,864, i/o block size = 1,048,576). This disk should be replaced immediately.
...
Jan 31 07:31:26 - SoftRAID Application: The certify disk command for disk disk5, SN: 0123456789ABCDEF, USB3 bus 0, id 0, lun 0 hung while writing (offset 87,350,575,104, i/o block size = 1,048,576). This disk should be replaced immediately.
I've tried this a few times now and I find it strange that diskutil can seemingly do this ok, but SoftRAID shows errors.
Note: I haven't been able to enable SMART via USB (yet - subject of another thread/SoftRAID bug), so I have no idea if the drives themselves are reporting any errors.
Also does it matter to SoftRAID that the USB enclosure returns "SN: 0123456789ABCDEF, USB3 bus 0, id 0, lun 0" for all 4 discs in the same enclosure?
So now I'm not sure if the discs are really both bad or something else is afoot.
Stephen
SoftRAID uses an OS X tool to certify disks. There is no SoftRAID driver involved. So a pass that fails certify, in theory should fail the secure erase. the main difference is SoftRAID uses larger IO sizes, so perhaps that is the trigger.
It is not terminal that your USB enclosure does not give each disk a unique ID, but it should. That is the sign of a budget enclosure that may not be as robust as it should. Worse are enclosures that have identical USB Global Unique identifiers. That happens when the manufacturers are too cheap to pay the modest USB royalty fee per device.
My guess is a better enclosure would not have this problem.

