K 10
svn:author
V 3
mav
K 8
svn:date
V 27
2011-01-17T18:58:28.424046Z
K 7
svn:log
V 488
Add FAILED disk and subdisks states. Disk falls into that state after
some number of fatal I/O errors. I/O to this disk should be avoided.
Same time RAID0 still can try to use it, as soon as there is no other.
Disk in this state is a candidate to be replaced with spare, if present.
Other option I had is to just drop failed disk out of the array, but
implemented way seems more alike to what Intel MatrixRAID BIOS does.

Add method to allow transformation modules claim disks as failed.

END