My department has a LAB environment with Cisco UCS blades and a NetApp FAS2240 storage appliance. We noticed a problem on one of the disks; an orange led was visible instead of green.
When we opened NetApp OnCommand System Manager, we saw the disk had the status “broken”, and “init failed” on the broken details properties.
Please note, before reading further – Let’s make clear I’m by no means a storage expert! The following steps might not be the formal way to replace a NetApp disk, but it worked in my situation.
We replaced the broken disk with a new one. When I looked in OnCommand System Manager, the new disk had the status “unowned”
There’s actually no option in OnCommand System Manager to assign an owner to the new disk. Start a SSH session (with Putty for example) to one of the Out of Band Mgmt Interfaces of the NetApp appliance.
Logon and type “system console“. Logon again.
With the commando “disk show -v” we see the status of all the disks, including the “not owned” disk.
You can assign the disk to an owner with the following command: “disk assign <disk id> -o <owner id> – s <system id>”
After this step I received a “bad label” error, as shown in the following screenshots.
To fix this, we type in the following command lines:
- priv set advanced
- disk unfail -s <disk id>
The disk is now added as “spare” and still not part of the pool and aggregate. To add the disk to the aggregate type in the following command: “aggr add <aggregate id> -d <disk id>”
In the warning message you will read the disk will start zeroing automatically. When this is completed the disk is successfully added to the disk pool!