I have a RAID 5 where a disk failed, and I am using Veritas. There was a spare declared, and everything was automatically relocated and recovery over to the spare completed successfully...see below
---------------------------------------------------
Failures have been detected by the VERITAS Volume Manager:??failed disks:? ora104??failed plexes:?vol01-02?vol02-02?vol03-02?vol04-02
Attempting to relocate subdisk ora104-01 from plex vol01-02.?Dev_offset - length 8467464 dm_name ora104 da_name -.?The available plex vol01-01 will be used recover the data.??Attempting to relocate subdisk ora104-02 from plex vol02-02.?Dev_offset - length 8467464 dm_name ora104 da_name -.?The available plex vol02-01 will be used recover the data.
Attempting to relocate subdisk ora104-03 from plex vol03-02.?Dev_offset - length 8467464 dm_name ora104 da_name -.?The available plex vol03-01 will be used recover the data.
Attempting to relocate subdisk ora104-04 from plex vol04-02.?Dev_offset - length 8467464 dm_name ora104 da_name -.?The available plex vol04-01 will be used recover the data.
Volume vol01 Subdisk ora104-01 relocated to ora105-01,?but not yet recovered.
Volume vol02 Subdisk ora104-02 relocated to ora105-02,?but not yet recovered.
Volume vol03 Subdisk ora104-03 relocated to ora105-03,?but not yet recovered.
Volume vol04 Subdisk ora104-04 relocated to ora105-04,?but not yet recovered.
Recovery complete for volume vol01 in disk group ora1.
Recovery complete for volume vol02 in disk group ora1.
Recovery complete for volume vol03 in disk group ora1.
Recovery complete for volume vol04 in disk group ora1.
---------------------------------------------------
I used vxdiskadm options 4 and 5 to mark the failed disk as going to be replaced, then I replaced it and it was initialized and is now online...ie
4 Remove a disk for replacement
5 Replace a failed or removed disk
What I wonder now is, should I keep the original "spare" disk (oa105) labelled as "spare", even when everything has been relocated to it? Should I mark it as "no spare", and make the original (ie replaced drive ora104 ) the "spare"?
I have read through various docs, and it is never made clear what one should do "after" a successful failover, and after the bad drive is replaced....
---------------------------------------------------
Failures have been detected by the VERITAS Volume Manager:??failed disks:? ora104??failed plexes:?vol01-02?vol02-02?vol03-02?vol04-02
Attempting to relocate subdisk ora104-01 from plex vol01-02.?Dev_offset - length 8467464 dm_name ora104 da_name -.?The available plex vol01-01 will be used recover the data.??Attempting to relocate subdisk ora104-02 from plex vol02-02.?Dev_offset - length 8467464 dm_name ora104 da_name -.?The available plex vol02-01 will be used recover the data.
Attempting to relocate subdisk ora104-03 from plex vol03-02.?Dev_offset - length 8467464 dm_name ora104 da_name -.?The available plex vol03-01 will be used recover the data.
Attempting to relocate subdisk ora104-04 from plex vol04-02.?Dev_offset - length 8467464 dm_name ora104 da_name -.?The available plex vol04-01 will be used recover the data.
Volume vol01 Subdisk ora104-01 relocated to ora105-01,?but not yet recovered.
Volume vol02 Subdisk ora104-02 relocated to ora105-02,?but not yet recovered.
Volume vol03 Subdisk ora104-03 relocated to ora105-03,?but not yet recovered.
Volume vol04 Subdisk ora104-04 relocated to ora105-04,?but not yet recovered.
Recovery complete for volume vol01 in disk group ora1.
Recovery complete for volume vol02 in disk group ora1.
Recovery complete for volume vol03 in disk group ora1.
Recovery complete for volume vol04 in disk group ora1.
---------------------------------------------------
I used vxdiskadm options 4 and 5 to mark the failed disk as going to be replaced, then I replaced it and it was initialized and is now online...ie
4 Remove a disk for replacement
5 Replace a failed or removed disk
What I wonder now is, should I keep the original "spare" disk (oa105) labelled as "spare", even when everything has been relocated to it? Should I mark it as "no spare", and make the original (ie replaced drive ora104 ) the "spare"?
I have read through various docs, and it is never made clear what one should do "after" a successful failover, and after the bad drive is replaced....