Case:
A Infortrend ESDS-S12F-G1440 storage, 12tb 2TB hard disks form RAID6, and all spaces of the entire RAID6 are divided into a LUN and mapped to the Windows system.On the Windows system, a GPT partition is divided, with a size of 18.2TB, and the original space is about 16.5TB.After using it for a period of time, it was found that the storage was unable to access. The manager checked the storage and found that the three hard disks were offline. After that, the storage was launched through some unconventional means, and the Rebuild began.However, during the host interview, it was found that the partition could not be opened, and all data could not be accessed.After consulting the professional agency, it was learned that this situation showed that the data had been destroyed. In order to avoid further deterioration of the state, Rebuild was suspended.
Solution:
In order to avoid the expansion of faults, engineers first do a complete sector -level cloning on the fault hard disk.If some hard disks have physical faults, the hardware recovery group resolves the hardware failure.Analyze the RAID6 algorithm used in the storage, and then use this algorithm to combine the combination of 66 possibilities for 2 hard disks for 12 hard disks.Artificial or procedure to determine the most correct possibility of lack.Build a virtual RAID platform through RAID data recovery software or third -party data recovery software, and adds to the analyzed dissection status, sequence, block size, verification direction, and RAID6 algorithm construction.For virtual RAID, explain the GPT partition structure, and then explain the file system to determine whether the algorithm is correct.If it is incorrect, adjust the algorithm until the best structure.Migrate data to another storage according to the file or sector method to complete the recovery work.