Data recovery failure description
The need to restore a company's data is an information management platform for customers to use three virtual machines to share a single enterprise storage devices for enterprise use, a large number of companies to store important data files. Administrators found that when in normal operation for the storage network and connected to a Windows2003 server, this result can not be used to store a sudden, storage administrators to troubleshoot virtual memory disk is lost, the partition table is lost. After restarting the storage failure is still not resolved. Since the data is stored in a very important and not backed up, the administrator can not be allowed to try to repair, but to restore the data by means of data recovery. 
Photos from the network, intrusion deleted
Storing data recovery analysis
Due to a hardware failure is not stored in the collapse, but the range of data recovery process we require hardware engineers help customers store hard to detect, it really all hard disk status is normal, there is no physical failure. It can determine the cause of failure is stored in the server administrator Windows2003 connected to the exclusive operation of storage resulted in destruction of the entire storage VMFS volume. Data recovery engineers to analyze the underlying data store found the partition table is cleared; there are effective end of the flag 55AA; hard disk ID logo. In the backward analysis it found that the store has a NTFS volume, but there is no data, data recovery engineers had no choice but to continue the analysis of the volume BITMAP find all the space and the size of the contents of the storage size is almost the same. It has occupied part of the volume at different locations, but all the space occupied by less than 110MB.
Finally, data recovery engineers found that the customer's storage there are two partitions, partition first group accounted for 80%, the second group is the first group of extended partition partition partition, and not related to the ntfs partition when the source data is destroyed the second partition. So the focus should be on data recovery partition on the first set, after analysis and inquiry found that important information of the first set of partitions are in, there is the chance of data recovery. 
Photos from the network, intrusion deleted
Storing the data recovery process
Since the detection of the hardware at the same time has been mirroring for all the hard drive before the analysis is also based on the operation of the image file, so there is no need separate image data of the customer. Two direct attached storage VMFS partition according to the partition organization may be extracted directly vmdk files and configuration files. After extracting the files can be carried out by nfs to move back to the data recovery self-test results, inform the client to the scene to verify the results of data recovery engineer self-correct. After successful authentication, the transfer of data.
Data Recovery Postscript:
The reason for this loss of data is very simple, it is because a fiber optic environment due to improper mutually exclusive volume in the Windows system partition and re-do the operation NTFS formatted and deleted partitions. Because esx vmfs mutex is hardware-independent Ming has a separate operating system dependent, where also remind the administrator when the other storage network access server storage should pay attention to assign permissions to avoid data loss.