During a previous Guest Files restore an NFS Datastore named VeeamBackup_<mount.server.name> was mounted to the ESXi host. If that ESXi host then had an unexpected reboot the previously mounted share becomes unavailable after reboot. When the next Guest Files restore begins the Veeam software detects that the NFS Datastore needs to be remounted and attempts to send requests for a new datastore to be created. This operation fails because the ESXi host kept the share name in the cache and does not allow the mounting of a share with the same name.
A reboot does not clear the cache, so restarting ESXi does not fix the issue.