#Issues with Disk replacement on version 8.3

1 messages · Page 1 of 1 (latest)

vivid umbra
#

We had a failed disk that was originally showing amber on a Dell DS2246 using Ontap 8.3, which we eventually will be replacing. However, when we removed the drive and replaced with a new one there was an error of "bad label". We had giving it time to register but it still shows up. We are working with an MSP provider that will note run the following command:
::> priv set advanced
::> storage disk unfail -s <disk_name>

There is a warning that states:
The following action plan assumes that the disks are not critical data disks being migrated to the new system. Verify that the disks in question do not store production data.

Will running this command for one disk cause processing power and affect the other disks in the Array to take down the entire storage array on that cluster?
Anyone else run into an issues with naming a bad label with the following steps? With it affect any of the other disks in the cluster?

  1. Initialize the Disk:
    • Command Prompt: disk initialize -f <disk_name>
    • Replace <disk_name> with the name of the replacement disk.
  2. Label the Disk:
    • Command Prompt: disk label <disk_name>
    • Replace <disk_name> with the name of the replacement disk.
  3. Check Disk Status:
    • Command Prompt: disk show -n
    • Ensure that the replacement disk now appears with a proper label and status.
  4. Add to Aggregate or RAID Group:
    • Depending on your configuration, you may need to add the replacement disk to an aggregate or RAID group.
    • For example, to add the disk to an aggregate: aggr add <aggregate_name> -d <disk_name>
    • Replace <aggregate_name> with the name of the aggregate and <disk_name> with the name of the replacement disk.
hollow ice
#

Dell?

#

But I typically saw this when a customer had a 3rd party replacement and the disk was just bad. I’d ask for a new disk before you run anything.

#

#4 is something I’ve never had to do at all and doesn’t make sense. When it comes to replacing a failed disk.

vivid umbra
#

Yes a Dell expansion shelf. We replaced it with two new drives and still faced the same issue the same issue.

lyric idol
#

ot sure about the Dell part, is it a rebranded system of some kind?
But to answer the question.. no, running the commands will not cause processing power to be used.
It 'should' not have any effect on the other disks, but stranger things have happened, if that disk somehow is broken and you force it online and it gets added to an aggregate/etc it could cause some issues, but that is unlikely.

I have not had problems with bad labels in a very long time but it's usually from a non-netapp supplied disk, disk being used on a different system with a different version of ONTAP or firmware needing to be updated.
Not sure you need to label the disk, as the system will do that itself.

Typically it goes like
1): priv set advanced
2): disk unfail -s xx.xx
3): disk zero spares

you can have the system zero the disk when it gets added to an aggregate, but I tend to do it when they are first put in the system.

if that doesn't work, the other option is
1): priv set diag
2): disk assign DISKID
3:) labelmaint isolate DISKID
4:) label wipe DISKID
5:) lbale wipev1 DISKID
6:) label makespare DISKID
7:) labelmaint unisolate

vivid umbra
#

Thank you their biggest concern is they wont doing anything without NetApp saying it is ok. Unfortunately since the support with them for that device was 2019, yes that old, we can get them to say yes or no on those commands. They believe because of the notification on the KB article says the warning they wont proceed https://kb.netapp.com/onprem/ontap/hardware/FILESYSTEM_DISK_BAD_LABEL_-_AutoSupport_Message. The MSP believes this will wipe all disks in the cluster and wont touch it unless verification from NetApp. Yeah I know very difficult to process.

unreal aurora
#

if the disk shows as bad label, it could well be that it was written to using a newer version of ONTAP and someone just forgot to mark it as spare and do a physical zeroing on it. In that case, and especially on 8.3, I'm not sure you can make the disk usable even with labelmaint. But it never hurts to try I guess, just don't kill your data by accident (labelmaint is a dangerous tool after all)