Disk Replacement

[larst1]

I have a FAS2040 (7-mode). Software version 8-1-4P2. One of my partner disks shows (prefail). I'm fairly new to netapp storage.

partner         0d.02.4         0d    2   4   SA:B   0   SAS 15000 560000/1146880000 560208/1147307688 (prefail)

Here are my spares:

RAID Disk       Device          HA  SHELF BAY CHAN Pool Type  RPM  Used (MB/blks)    Phys (MB/blks)
---------       ------          ------------- ---- ---- ---- ----- --------------    --------------
Spare disks for block checksum
spare           0d.01.1         0d    1   1   SA:B   0   SAS 15000 560000/1146880000 560208/1147307688
spare           0d.02.8         0d    2   8   SA:B   0   SAS 15000 560000/1146880000 560208/1147307688

Can I issue the following command to replace the disk?

disk replace start 0d.02.4 0d.02.8

Thanks,

Larry

[AlexDawson]

Hi there,

ONTAP 8.1 would require you to run commands on the node owning the disk at the time, so potentially two SSH sessions.

If you log in to the controller with the prefailed disk, and run "aggr status -r", I assume that there are no spare disks listed and the pre-failed disk is still part of the aggregate?

To begin - if you plan to run this controller longer term, you should download this document - https://library.netapp.com/ecm/ecm_download_file/ECMP1141781 - we no longer support ONTAP 8.1 and the documentation may be removed at any time.

On the controller with the spare disks, follow the procedure on page 43 - turn off disk auto assign and remove_ownership of the spare disk. Then on the controller that needs the disk, run "disk show -an", and verify the spare disk is visible and unowned, then run "disk assign <diskname>". At that point it should automatically recognise it as a spare, but I can't remember the logic for 8.1 - it may not automatically replace it since it wasn't present when the disk failed, so then yes, run the "disk replace" command.

Good luck!

[Ashun]

hi，

I don't think you need to worry about the state of the drive being prefail, because when the state of the drive becomes failed, the spare drive will automatically replace the failed drive.

Of course, you can advance the replacement process by using the command you mentioned

But if you want to replace it manually, you need to do it at the node of this prefail drive

[akiendl]

The disk is belonging to the partner node. not the node you are logged in to. In order to address this login to the partner and identify the disk. If it is not pre-failed there do a disk fail on the drive wait for it to be evacuated and then do a disk unfail.  That should set the status correct.

As said above you run a very old release that has no support any more.