Case:
Customer storage model is IBM STORWIZE V5000, and RAID5 made with 24 300GB SAS disks, no heat preparation.There are 3 LUNs on the storage, one where LUN is located for the Oracle database backup file.Lun’s usage environment is RHEL5; the other two Luns are LUN where the Oracle production library of the ERP system is located.Because the two disks are offline, RAID crashes, the data is lost, and the hard disk is detected. A large number of bad sectors are found in many hard disks.
Solution:
After the engineer processed the bad disk of each hard disk, it was fully mirroring into the secure storage of North Asian data centers.All the original disk disks of the customer to ensure that it is not damaged by the outside world.Analyze the original RAID combination parameters from the disk mirror.Analyze the first offline disk, because the first offline disk is not involved in the storage of the data after the offline disk, and the data written after the disk has nothing to do with this disk.Virtual reorganization RAID, restore all data.After verifying the Oracle database, it was found that a table space file was damaged. This was caused by disk bad sectors. After the engineer was repaired, it was normal.