A recent case in a failed EMC storage array with over 40 drives confirmed that multiple disk failures can happen even in an enterprise class architecture. When a large RAID array fails it is recommended you get advice from Andy at R3, pulling and replacing drives and then allowing drives to resync may appear a normal procedure. But it is the most likely time data will be corrupted and the data be lost.
The problem stems from the fact that if one drive has failed and not been noticed then another fails just replacing the failed drives will often lead to finding a 3rd or 4th failing drive, especially in large scale environments.
The problem is the next drive to fail might only do so because of the resynching of the replacements of the first drives to fail. It is at this point it get messy and can become a full disaster recovery situation requiring data recovery.
Call to Andy at R3 for advice on a failed RAID array or SAN - his lab and team of data recovery engineers are one of the best ever assembled in the UK.