This problem occurs because of an error in the Dmio.sys module. The Dimo.sys module does not process bad blocks correctly as specified in the "SCSI Sector Sparing" technology specifications. Therefore, some reassignment requests fail unexpectedly. See ME928124
for a hotfix applicable to Microsoft Windows Server 2003.
Definitely this is an information that must warn you to have frequent backups and to consider replacing the hard drive if things get worst.
On the other hand I have seen hard drives functioning without problems for years after such an error. Depends of what caused the bad block, how extensive the damage is (usually are more bad blocks), where the bad blocks are located.
You may consider using a low level utility from the manufacturer of your hard drive, for example I used with good results PowerMax from Maxtor (see link). These utilities are able to "clear" the bad block. Actually it is not repairing the bad block, it's just reallocating the bad block to the spare track in order to give the same storage capacity and make the customer happy that the hard drive has no bad blocks. As these programs know everything about the particular hard drive, they are able to perform best testing, sector by sector. If the utility "re-certify" the hard drive, you may use it with enough confidence for a while.
Use Scandisk to fix it temporarely but anyway, the drive is on it's way to fail.
In our case, this message seemed to occur randomly when we broke up a mirror and used the disk for other purposes (e. g. for a volume or a mirror of another volume). There seemed to be no hardware defect. Only Adaptec Controllers have been involved in this yet. If possible, perform a low level format using the SCSI utilities of the controller.