Correct method of server data recovery/solution of server hard disk failure

[Server data recovery reason inference]

Server data loss is a lot, often can not clear the cause of server data loss, common causes of loss are: Server hard disk failure, Administrator or server automatic fsck operation, this operation may cause more serious data loss or make the server data recovery more difficult. Server data loss after performing MKFS operations, if the operation is not completed, it is easy to cause some block group all 0. Or the administrator in the server data loss after the attempt to recover data, data recovery company improper operation, etc., may further damage to the data within the server, increasing the difficulty of data recovery.

[Correct method for server data recovery]

First of all the data of the failed server to perform a full mirror backup operation, data recovery should not be in the original disk for data analysis and recovery (including the server, all storage media, such as data recovery operations). Analysis of the image file determines the disk partition composition of the server, the remaining space allocation, what the file system type is, whether the root partition's Super block state is normal, and whether the description of the first group is normal according to the Super Block state, and whether the state of the node area is 0.

Since the first group has been queried for the description of the state, and then according to the first group of the description table for the other block group analysis, query rules, such as we found in the server data recovery in the first 26 blocks of the user server is all 0, the last part of the block group of the source data area is also all 0, but the middle block group of Nodes, bitmap and other information are normal state, block group before and after the data area has obvious user data existence phenomenon.
Try to check the root directory of the XXX and take this as a clue to the server root node area to restore, combined with root records to generate a file system tree, (server data recovery to this can already see some data) check the structure of the file system is normal, note whether the file or folder node is 0, The trace file node finds the location of the node area in the file system. It is often difficult to repair and export this part of the data if a file or folder with a node area of 0 is found in the file system structure check.

[Server Data Recovery recommendations]

If the data inside the server is very important, try to use a RAID array of multiple hard disks as storage environment, avoid using a single hard disk for storage environment; The data in the server must be backed up regularly, the backup package should not be stored in the same storage body, if the condition is not allowed to be stored in the unified partition , the server disk should be done as soon as possible after the backup work, to avoid the non-professional repeated attempts to repair, select the data recovery company as far as possible to choose a professional company to operate, to avoid intentional or unintentional two damage.

Correct method of server data recovery/solution of server hard disk failure