absent vSAN Disk unable to remove

[niceguy001]

as shown in figure below:

my infrastructure is using vSphere 6.7U3 now and a disk in a vSAN host became absent.

things i've tried:

1. use command "esxcli vsan storage remove" to make it out of the vsan but failed(see figure below)

2. put the host(which the disk located on) into maintenance mode, failed because of this disk...

3. remove the vsan disk group which contain this disk from the host in vsan configuration page, still failed because the disk is absent....

4. reboot the host and re-check this vsan disk: still absent.

should I just pull this disk out of server? this is my final solution and I don't want to do that for now cause i treasure the data on it.

[TheBobkin]

Hello niceguy001​

What are you trying to achieve here?

Are you trying to get that disk into a healthy state so that you can use the data on it?

or

Are you trying to remove the stale entries relating to this disk so it can be re-used as a new disk? (provided it isn't physically faulted).

So that you are aware, 'esxcli vsan storage remove' is essentially a command that deletes the data on a Disk/Disk-Group so if you are trying to achieve the first option then this is not the path to do this. Secondly, you are using that command with the incorrect syntax - it is expecting a disk (e.g. naa.XXXXXXXXXXXXXXX) not a vSAN UUID if using the -d or --disk option, if you are trying to delete the references to this device with the vSAN UUID then use the -u option.

If you need the data that is on the device then you need to figure out why it got marked as failed from the vmkernel.log and vobd.log, if you attach them here I can take a look but from your screenshot it looks like that device is/was PDL.

Bob

[frandelgado93]

I had this issue happened to me before in several customers sites. I normally fix it with this procedure.

1. Put esxi server in to maintenance mode and then disconnect from vcenter(disconnect button in vcenter). If you can't put the esxi in maintenance mode move all vm's out of the host and disconnect. (change drs to manual)
2. Move the esxi outside of the cluster
3. (optional)Add or remove any faulty drive, and I always format all the disk in this steps using the built in format option on esxi or booting the server with gparted or any other live linux distribution.
4. Move esxi host back to vsan cluster
5. Add disk group and reclaim disk.

[niceguy001]

Guys, thank you somuch for the help~!

eventually, i removed and examined the absent disk and found that it is failed.

thanks again, your information is truly helpful.