Datastore conflicts with an existing datastore

• Multiple ESXi hosts in the same cluster may become unresponsive or go to a Disconnected or Not Responding state as a result of the attempt to add the ESXi host

Cause

This issue occurs due to a duplicate entry of the datastore UUID in the vCenter Server database

Impact / Risks

This process interacts directly with the vCenter Server Appliance (VCSA) Postgres Database, and it is recommended to have a snapshot of the VCSA prior to proceeding.

Resolution

Ensure a snapshot is taken of the vCenter VM. Then, to resolve this issue:

1. Connect to the VCSA through SSH.
2. Access the Bash Shell with the command shell.set --enabled true or just type shell.
3. Stop the vpxd service with the command: service-control --stop vmware-vpxd
4. Access the Postgres DB using this command: 

/opt/vmware/vpostgres/current/bin/psql -d VCDB -U postgres

5. vCenter 6.5: Run the query select id,storage_url from vpx_datastore;
6. vCenter 6.7/7.x: select * from vpx_datastore;
7. Look for the affected datastore (DS) UUID and find the ID.
8. To confirm the correct ID as needed, run select * from vpx_entity where id=(ID found on previous step);
9. Delete the entry from vpx_ds_assignment, vpx_vm_ds_space, vpx_datastore and vpx_entity using query delete from vpx_ds_assignment where DS_id=XXXX; 

Sample were the ID = 461:

delete from vpx_ds_assignment where ds_id=461;
delete from vpx_vm_ds_space where ds_id=461;
delete from vpx_datastore where id=461;
delete from vpx_entity where id=461;

10. To quit from database use \q
11. Start the vmware-vpxd service in order for changes to be reflected:

 service-control --start vmware-vpxd

12. Once the vmware-vpxd service is up again, proceed to add the ESXi host again back to the vCenter Server.