Description of problem: Distinguish between manual fence resulted from user action and manual fence called internally by auto fence Manual fence currently can be executed in 2 flows 1) As part of the non responding treatment when host is restart via its PM agent. 2) As a result of a manual host stopping and choosing "confirm Host has been rebooted" from the right-click popup menu The messages generated by both to the log are of the format: Manual fence for host ${VdsName} was started. Manual fence for host ${VdsName} failed. When this occurs in scenario 1) above, teh term "manual" in the audit log is confusing and should be changed to reflect the fact that was done internally by the engine code and not by the user Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1.Have 2 Hosts in the Clustet/DC one with tested PM configuration 2.Block connection to the host from engine 3.After host is rebooted check the engine log for "Manual fence for host ${VdsName} was started." message Actual results: Messages with the term "manual" are logged when the fence operation is internal Expected results: Messages with the term "manual" are not logged when the fence operation is internal, engine should provide a different suitable message for that Additional info:
Verified on rhevm-3.4.0-0.16.rc.el6ev.noarch. with hosts with vdsm-4.14.7-0.2.rc.el6ev.x86_64. 1. stopped network on host with pm. host was rebooted using power management and the audit log didn't show the original "manual fence" message. 2. if I use power management through ui, get the same result as expected 3. shut-down the host. host becomes non-responsive. then start the host again and chose 'confirm host has been rebooted' in the ui. in this case got the message 'Manual fence host rose07 was started as expected
Closing as part of 3.4.0