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

[failure analysis of server data recovery]

In the data recovery industry, we often encounter data loss caused by raid module hardware damage caused by accidental power outage or raid module damage such as the loss of riad management information. Normally, once the raid array of the server is created, it will no longer change the information in the management module, but the information of the raid management module can actually be modified. One or more accidental power outages may cause this information to be tampered with or lost, and too many power outages may even lead to damage to the meta-devices on the raid card. It indirectly causes the host to lose the middle-tier module for RAID management of multiple physical hard disks. This is the case with this server today.

[server data recovery failure description]

Customer server belongs to HP brand EVA4400 series storage. There are 6 hard drives at the bottom of the server to form a raid5 array, and a single hard disk is 1TB and SAS hard disk. The operating system is WINDOWS 2003 SERVER.

The main role of the server is the company's internal file server, in the normal working state due to the computer room voltage problems encountered two consecutive accidental power outages. The administrator restarted the server, but the raid array prompted "unable to find the storage device" and tried to enter the raid management module to check, but when entering the raid management module, the server crashed, and the failure still existed after many attempts, so the administrator had to contact the data recovery company for server data recovery.

[server data recovery scheme]

1. First of all, all the hard drives in the server are physically tested to determine whether there is a physical fault, and the hard disk reading state is normal.

two。 Make a mirror backup of all the hard drives of the server. Note: in order to ensure data security, the mirror target storage needs to select an array with redundant functions.

3. After the backup of the original server is completed, the server data recovery engineer analyzes the data of the backup file, determines the data block size, stripe information, disk sequence and check mode of the 6 hard disks in the server through the file system storage rules, and rebuilds the virtual raid group in the virtual environment.

4. After logically checking the data in the constructed RAID to ensure that the parameters applied in the refactoring RAID are correct, complete verification is carried out against the data that users are most concerned about.

5. After the user confirms that the data recovery result has fully reached the expected (the data is restored to the pre-failure state), all the user business data are migrated to the user storage, and the data recovery is complete.

[server storage security recommendations]

1. Try to ensure a stable power supply in the computer room, so as to reduce the impact of abnormal power supply on the host and storage

two。 It is best to configure UPS for important servers and storage, which can ensure that the core business system can continue to work normally for a certain period of time in case of accidental power outage in the computer room, thus winning valuable time for enterprises to seek emergency solutions.

3. The security status of the server that has been in service for a long time should be checked regularly, and its overall operation status should be evaluated to determine whether to upgrade the hardware and system comprehensively, and to formulate an emergency plan to deal with sudden data disasters in advance to reduce business losses caused by data disasters.

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.