Server data recovery Background:
The background data recovery case is a normal server with 20 hard drives, for unknown reasons the sudden collapse of the upper layer service, the server room administrator checks found that the main cause of server crashes are three hard disks on the server offline, administrators will all hard disks in the server according to the existing order to carry hard disk to a data recovery center in Beijing for server data recovery operations from the slot after removal.
Server data recovery process:
After the server data recovery engineer to the customer's hard drive data recovery testing equipment to detect the hard disk 20, and found that all hard disk data recovery under the equipment can identify, which avoids the process of repair of hardware and physical damage is too severe because the hard disk can not be repaired due to server data recovery risk, is a fortunate thing, then all the mirrored hard disks in the server, mirror server found during the original three prompts offline disk mirroring a very slow process, which is also before there is a certain relationship between the causes of hard disk offline, the majority of the reason is because these three hard disk memory in a large number of bad sectors or unstable sector, so there are offline in the normal server environment, data recovery professional equipment you can recognition, mirroring the situation will be very slow process in the image by adjusting the mirror policy to adjust too bad sectors hard disk to be adjusted until all hard drives are mirrored successfully completed. 
After all the hard drives have been successfully mirrored, engineers continue to use the data recovery server data recovery tool to perform all of the underlying data to expand the image file analysis, the server hard disk and parity information in accordance with the reverse sequence analysis ext3 file system analysis here I do not go into details. Finally, use this information to analyze out restructuring raid array, to extract some of the oracle dmp files and user communication, in the course of dmp recovery, database reports for the imp-0008 error, through careful analysis of import dmp file log files, there is a problem caused by the recovery of dmp dmp file import data failed. Raid immediately re-analyzed the structure, and to further determine the extent of ext3 file system is damaged, and after hours of work, restore the original dbf dmp file and library files will be restored out of the dmp file transfer user data to import test, the test results successfully found no problems, indicating that the data recovery is successful, then the recovery dbf out of the original library file for verification testing, all files can pass the test.
Server data recovery engineers contact customers for data recovery results of the verification, customer-proven All data has been successfully restored, then on the server and set up a new set of raid array, the data recovery engineer with the server data migration all recovery success on the server back to the client. In principle, the customer's data in our free save for three days, three days after the destruction of data.