For many companies, RAID is a convenient, practical way of storing critical data. But if the RAID system crashes, users risk losing precious data – their most important asset. It could be important research materials, key accounts statements or even customer information. So what can they do to prevent losing crucial files when this happens?
For a start, do not attempt to recover the data on your own – even he most experienced system engineers who may be familir with standard configurations may not have the skill sets to rescue a corrupted or inaccessible RAID volume. Instead, there are other ways to minimise your loss of data or set the disk up for a better recovery.
So, determine which situation you are in:
Situation 1: There's one disk failure and the RAID server is running at a degraded mode. Only some critical data is required.
Do this: Copy out your critical data before any attempt to rebuild. At this point, the remaining disks that make up the rest of the RAID volume could be near failure. So when you start rebuilding , the IO-intensive process might stress these remaining disks to total failure. Only carry out the rebuilding process after the critical data is copied out. A tip: Before you copy the data, you may want to consider cloning out the disk image in case any additional disk should fail during the copying process. See the end of the article for a step-by-by procedure.
Note: If you are in Situation 1, thank your lucky stars because this is the most likely situation where you will be able to recover your data completely. There might not even be a need to send the RAID system to a data recovery specialist. This is because the RAID is still
accessible (but without any more fault tolerance redundancy). Only the failture
of the next disk will result in complete RAID server failure.
