RAID 5 Server does not equate to a fail-safe backup.
RAID which stands for Redundant Array of Independent Disks or sometimes alternatively Redundant Array of Inexpensive Disks - is a technology that employs the simultaneous use of two or more hard disk drives to achieve greater levels of performance, reliability, and larger data volume sizes.
Redundancy is a way that extra data is written across the array and organized in a manner such that the failure of one (sometimes more) disks in the array will not result in loss of data.
RAID 5 arrays are used by many companies because they are cost-effective and provide a level of fault tolerance and protection against drive failure.
Short comings of Raid 5
However, companies have to know that RAID 5 servers can fail. When one disk fails, a RAID 5 server can still function without any data loss. However when two hard disk fails at the same time, all data is lost. In short RAID 5 can survive only one disk failure, but not more than one at any one time.
Companies that utilize RAID 5 servers must be aware that RAID 5 is not a good alternative to backing up data. RAID 5 cannot prevent data loss – when more than one disk fails.
Data may also become damaged or destroyed, even when the drive on which they are stored is intact. For instance, part of the data may be overwritten by a system malfunction; a file may be damaged or deleted by user error and not be noticed for days or weeks.
RAID 5 servers also suffer the issue of correlated failures – when one disk fails, there is a high chance that the second one will fail. The theory behind the error correction in RAID assumes that failures of drives are independent. In practice, the drives are often the same ages, with similar wear. This means the failure of these drives are statistically correlated and the chances of a second failure after the first failure is significantly higher.
A part from outright disk failure, RAID 5 servers can experience failure in many ways, including corrupt data, degraded raid 5, deleted files, missing partitions, RAID 5 controller failure,
In summary, companies have to realise that RAID 5 servers are configured to only withstand the failure of only one disk at a time. At the same time, the likelihood of a second disk failing is significantly higher when the first disk fails , which may result in data loss.
What to do if your Raid system crashes
The first step is to identify the whether the crash is due to one disk failure or multiple disk failures. For a situation involving one disk failure, the RAID server is running at a degraded mode. Critical data should be copied out as soon as possible before any rebuild attempt is performed. Once the critical data is copied out, standard rebuild process can then be carried out.
In the event that the RAID server is used as an application and data server, it may not be sufficient to recover just the data as re-configuration of application may not be possible.
