Description of problem:
In a cluster configured for HA resource reservation, with a single host in state up which is running HA vms, and all the other hosts are in statuses Maintenance/Down, the check of HA resource reservation is ended successfully.

Version-Release number of selected component (if applicable):


How reproducible:
always

Steps to Reproduce:
1. Mark cluster for HA resource reservation
2. Have 2 hosts in the cluster:
2.1. host 1 is down
2.2. host 2 has 2 vms running on
3. wait up to 5 minutes where the following appears in the engine.log:
2014-01-24 13:51:24,315 INFO  [org.ovirt.engine.core.bll.scheduling.HaReservationHandling] (DefaultQuartzScheduler_Worker-34) HA reservation status for cluster Default is OK

Actual results:
Resource check ends successfully

Expected results:
If there are no hosts when the check is executed that can guarantee the resource reservation for the HA vms, an error should be reported both to engine.log and mostly to the event log.

Additional info: