Description of problem:
non-responding host is selected as a proxy for fencing operations
Version-Release number of selected component (if applicable):
How reproducible:
Steps to Reproduce:
1. Have Host1 with PM and Host2 under cluster C1
2. Have Host3 under cluster C2 on the same DC as Host 1 and Host2
3. block communication from engine to Host2
4. wait for Host2 to become non-responsive
5. block communication from engine to Host1
6. Host2 is selected as a proxy for fencing operation and only after it fails Host 3 is selected and perform the operation successfully
Actual results:
non-responding hots Host2 is selected as a proxy
Expected results:
non-responding hosts should be skipped while searching for a proxy
In the scenario above, proxy selection algorithm should return Host3 for the first time
Additional info:
this bug status was moved to MODIFIED before engine vt5 was built,
hence moving to on_qa, if this was mistake and the fix isn't in,
please contact rhev-integ