The RAID group was set degradation after one of its members lost efficacy. Before its recovery, if the IO of another member had bad sector, the system would set the LUN where the bad sector was failure while there was no influence to other LUN.
1. If there is a unused hot spare disk in system, change its type into spare, avoiding the system to start reconstruction automatically in step 3.
2. Recover all the member disks which have bad sectors one by one(maybe some), in this example, there was only one disk(0,5) which has bad sector,
a. Collect the smart information of the disks which have bad sectors;
b. Check the smart information, find the Pending Sector Table Entries of the disks, and ensure the LBA of bad sectors;
In this example, the LBA of the bad sector is 0x00002710.
c. Recover all the bad sectors with the disktool;
d. Back to step a till the record of Pending Sector Table Entries in disks is empty.
3. If the first disk which lost efficacy has associated with a hot spare disk, pull out the hot spare disk first to set the RAID group degradation, then use the command “revive lun<LunID>” to set the LUN which lost efficacy normal;
4. Replace the disk which lost efficacy, if so, swap the disk manually, and the system will start reconstruction automatically;
5. After the reconstruction is done, check the state of LUN and RAID are normal or not, recover the hot spare of the system;
6. Scan LUN in the host again, check the file system and recover the service after the LUN which lost efficacy was found.