Shulou(Shulou.com)06/02 Report--

1. Description of server data recovery

Let's first give a brief introduction to the server failure. The customer's server is a Huawei OceanStor S5600T model, the server is located in Beijing Daxing computer room, using the way of MD1200 disk cabinet + raid card, the server has a total of 15 hard drives to create a set of RAID5 array, assigned two LUN a total of about 80T, at the system level, the LUN is divided into three partitions: sdc1, sdc2 and sdc3, in which the sdc1 partition size is 4T, through the way of LVM expansion, the sdc1 partition is added to the root_lv The remaining sdc2 and sdc3 partitions are formatted for XFS file system use.

After reinstalling the system on the server, the administrator found that the sdc disk partition had changed, and the original sdc3 partition was lost and inaccessible.

Second, server data backup

The server data recovery engineer carries the backup server to the customer site for the initial inspection of the data, and backs up the data of all disks to the backup server with the help of the data recovery tool. For later analysis and recovery of data.

3. Server data recovery steps

1. Use the mirror disk to analyze raid's disk order, stripe size and other information. Here, you can search and view the FILE ID number with the help of the data recovery tool, and you can determine the disk order, block size, loop direction, homoasynchronous, etc. (you need to be careful. If the data is very important, you'd better find a professional data recovery company for operation).

Through the virtual reorganization of the analyzed raid information, the raid structure is obtained.

2. Navigate to the starting position of the partition of the xfs file system; XFS INODE number: the number of digits with variable length, which is composed of three parts: starting block group number + starting block number + INODE number within the block. The bit length of the starting block number and the INODE number within the block is specified by the frombyte.com parameter in SUPERBLOCK.

3. Check the integrity and correctness of the xfs file system; through verification, it is found that the super block and some nodes and directory items in the head of the file system are missing.

4. Repair the super block structure of the xfs file system, repair and restore the super block according to the super block backup and the directory tree structure in the file system.

Figure 1:

5. Repair the lost nodes and directory items in the xfs file system, repair and reconstruct the lost nodes and directory items.

Figure 2:

Figure 3:

6. Write a program to parse the xfs file system and extract the data after the repair is completed.

4. Server data recovery results

Because the customer did not do any other destructive operations such as writing to the server after the data loss, most of the data and file system information remained intact, and the data was recovered 100% successfully.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.