Need help with next steps after disk issue
I am running a Thunderbay with 4 drives over Thunderbolt to my Mac running SoftRAID. Two of the drives are configured as a Raid 1 Volume. A couple days ago my Mac started locking up and crashing repeatedly. Additionally I was seeing many errors reported on one of the drives of the Raid 1 volume. After some Internet investigation, I tracked it down to the Thunderbolt connection to the Thunderbay (when I disconnected the Thunderbolt cable, the crashes stopped).
After this I then powered down both the Mac and Thunderbay then restarted both. Since then I have had no crashes but SoftRaid reports one of the drives as having I/O errors.
I ran verify on the drive and it reported no problems. I reset the I/O error counter and the drive is currently showing failed as part of the volume. SoftRAID log shows warnings that the disk is 20-60x more likely to fail over 2-6months.
It seems to me that the original problems I has having with the drive may not have been the drive but something wrong in the Thunderbay that was fixed by the power cycle. (Of course, I suppose it could have been the disk failing that caused the problems resulting in the MAC crashing but would be surprised that a disk going bad would cause these crashes).
Given that a verify passed does this mean the drive is ok? If not would a certify tell me that the drive is ok? If the drive is not ok, should I just power down, yank the drive, put a new one in? Thoughts on the right approach would be appreciated.
Look at the disk tile. Does the disk have reallocated sectors? (even one?)
Then replace it.
Does it have "pending reallocations"? or "unreliable" sectors?
then "certify" the disk to determine if these sectors are good or bad.
(destroys data on the disk)
Its possible the cable connections did damage the drive, but those are the parameters I would use to determine a failed drive.
if it just has IO errors, clear the errors and don't worry about the disks.