I have recently have several problems with Alerts that won't clear, even though the monitored component is recording an UP status.
I currently have two open alerts claiming one application is critical, and one is down - although both show as up from all their recent polls.
The problem seems to be that the data in the database table "ContainerMemberSnapshots" is still showing these components on their original alert status, and is not refreshing.
How can I force this data to refresh?
One thing I did notice is that these alerts seemed to fire about a minute after the issue actually resolved, e.g. today I had:
TIME OF EVENT MESSAGE
18/07/2013 10:28 ALERT: NetPerfmon - Application X critical
18/07/2013 10:27 Group A is Up
18/07/2013 10:26 NetPerMon Event Log - Component Z on application X is Up
18/07/2013 10:25 Group A is in a crigtical state due to a member status. Application A is in a Critical State
18/07/2013 10:25 NetPerMon Event Log - Component | on Application X is Critical
I have several applications in a group (e.g. Group A), which are polled every minute.
The alert in question is configured on "Group Member Status" with a "trigger after condition exists for more than 2 minutes" set (to avoid any single poll interuptions) to fire when a member is down or critical.
The alert firing is a pain - but I'd like to be able to clear the incorrect ones - can anyone help? I've restarted the services with no effect, and previously I've had to unmanage/remanage the applications to force it to clear, but there must be a better way.
Tim