Sometimes you are required to re-deploy the ESXi host that was part of a vSAN cluster. If this unfortunate situation arises, and you find that you are not able to add the local disks to the vSAN cluster, it is because they are already registered to the old vSAN UUID? So, what to do?
During my new lab deployment, I found myself crashing my ESXi hosts several times due to many reasons; but everything seemed to be OK as I recovered them, because I was quickly improving my skills with vSAN and NSX-T and other technologies.
But when I completed the re-deployment and tried to add the ESXi hosts to a vSAN cluster, I realised my vCenter could not recognize the disks as valid disks to reconnect to the vSAN, despite the new ESXi deployment.
What had changed? I did a check with the following command to see what was going on.
The SSH command esxcli vsan storage list revealed that the old vSAN was still holding my local disks. You can see this at the screen shot below (VSAN UUID):
To remedy this situation and set your disks free, you require the following simple command: esxcli vsan storage remove -u <UUID>
After you have run the commands we talked about here, you will be able to run through the VSAN configuration interface without further issues and add your disks to the existing vSAN, or a new one.