There is a bug somewhere that causes this problem on a host.
The host thinks the resource pool is not there, VC says, go ahead add it, but it already exists. You will go into a loop.

Do NOT turn off DRS, this will remove all your resource pools, but put it on partial automated.
The only way to fix this:

1. Check the host’s configuration (with the client direct on the host) Virtual machine startup/shutdown.
Check if the box at ‘Allow virtual machines to start and stop automatically with the system’ is NOT selected.
2. Disconnect the host from the cluster.
3. restart the mgmt-vmware (service mgmt-vmware restart)
4. wait for a couple of minutes.
5. reconnect the host to the cluster.
6. Enter maintenance mode, go to drs en generate recommendations, check them, and apply them.
The hosts will be migrated to the other machines.
7. remove the host from the cluster.
8. remove the resource pools on the host (using the vmware client)
9. restart the host.
10. Import back into the cluster.
11. Check if everything is working. (try a migrate by hand.)
12. Put drs back on automated.