Description of problem:

On RHEV 3.3, once a host becomes unreachable via rhevm/ilo, it moves to non responding and an alert is displayed on the log file


Version-Release number of selected component (if applicable):
rhevm-backend-3.3.1-0.48.el6ev.noarch

How reproducible:


Steps to Reproduce:
1. Have one host in Up state and other hosts working in the DC
2. disconnect both LAN and ilo network
3. Wait for the host to become connecting/non responsive
4. watch for fencing attempts
5. watch for logging events to appear

Actual results:
Alert is not cleared

Expected results:

Alert should be cleared when status is fixed as per https://bugzilla.redhat.com/show_bug.cgi?id=840495
Additional info:

Attaching shortly a screenshoot