vROPs reports that there are reclaimable CPU, Memory and Disk Capacity in several clusters.
I know that Memory and CPU can be decreased provided that the VM is powered off. With downtime but doable.
The question is, how about the disks? how can it be decreased? Some are production VMs.
The admin wants to execute the recommendations provided by vROPs.
UP!
With disks it's not a straightforward process and there's no great way to go about it short of rebuilding things or provisioning a new disk and migrating data inside the guest onto it.
Then what is the process to do this?
Add a new disk to the VM of the desired size, then migrate data within the guest to the new drive.
THis would be applicable if disks are added independently.
how about for VMs with a single disk but multiple partitions?
You'll have to recreate the partition structure in that case. Generally speaking, though, there is little need for multi-partitioned virtual disks so you may want to use separate VMDKs and just change your data layout.