Mistakes can kill RAID
RAID Recovery - Mistakes can kill RAID
Users are expected to have rough and ready knowledge of all the functions of RAID to minimize risks and malfunctions in RAID drives. One of the best reasons to use it is to achieve state of excellent performance and enhance reliability with mirroring in operating systems; for example, RAID 1 and 10 or striping with parity such as RAID 5.
RAID is a complex process and any wrong command can cost all the stored important data in the array. But, one can prevent most of them and save the lifecycle of information by careful management.
Failed consistency checks
Users should confirm whether RAID arrays are functioning consistently or spot blocks have gone wrong. This assessment on a regular basis will offer users to know if the hot spares in the system are absolutely capable of functioning properly to take over the charge in the case of disk failure. The negligence about checking up the array can lead to a big problem such as disk failure.
Hard disk failure in RAID 5 array is the most problematic. If two or multiple disks fail together, the system may not be able to rebuild the array even after replacing the failed disk. On the other hand, RAID 6, which uses two parity stripes, is capable of protecting against this problem but at the expenses of increased hardware and reduced performance. It is far better to check up the system constantly and change the drives if needed. Some software are available that make the users free from daily monitoring the system. For example, NetRAID Assistant is commonly used to check disks once a week.
Write cache without battery backup
RAID controller come with caches that can be set to offer different amounts of memory for read or write caching. Checking up the battery backup of controller is very important before using write caching; otherwise, one can lose data or corrupt the system in case of power failure. RAID controller does not have battery backup and it is better to set the write cache size to zero.
