Created attachment 930541 [details] engine log Description of problem: when configuring power management health check and adding a power management agent with false credentials to a host, a warning will be issued, but afterwards no similar warning will be issued. Version-Release number of selected component (if applicable): ovirt-engine-3.5.0-0.0.master.20140821064931.gitb794d66.el6.noarch How reproducible: always Steps to Reproduce: 1. with engine-config, configure: PMHealthCheckEnabled=true, PMHealthCheckIntervalInSec = 20. 2. restart engine. 3. add two hosts, one with a pm agent. 4. configure the pm agent with invalid credential (wrong username/pass/port) 5. wait for health check alert to be issued. 6.a. configure the pm agent with the correct credentials and wait until a check is done and successful. 7.a. configure the pm agent with invalid credential again (wrong username/pass/port) 6.b. remove the alert from the alerts tab Actual results: in either 6.a-7.a or 6.b, the result is a message regarding the issued STATUS command every 20 sec, but no alert is issued again regarding the health check's failure (which does appear in engine.log) Expected results: After removing a health check alert, either manually or by correcting the initial problem, we expect to get an alert again in case the same power management agent is faulty again. Additional info:
Verified with rhevm-3.5.0-0.12.beta.el6ev.noarch according to scenarios described on the description, also tested scenarios checking all different error messages (sequential primary, sequential secondary, concurrent primary, concurrent secondary) are reacting the same and are reappearing after first occurrence.
oVirt 3.5 has been released and should include the fix for this issue.