Hi all!
I am trying to work out how long / often nodes are being put into Unmanaged mode - so we have a better idea of how accurate the availability statistics are (for aexmaple, if a node has 100% availability, but has been in Unmanaged mode for 80% of that time, how accurate is the availability?)
At the moment, I have an SQL query which gets me close to what I am after:
use SolarWindsOrion
go
Select AE.TimeLoggedUtc as 'DateTime', AE.AccountID,
CASE AE.ActionTypeID
when 26 then 'Remanaged'
when 27 then 'Unmanaged'
end as 'Management Action'
, Nodes.Caption
, Nodes.UnManageUntil,
CASE Nodes.UnManaged
when 1 then 'Currently unmanaged'
when 0 then ''
end as 'Unmanaged State'
from AuditingEvents AE
join Nodes Nodes on Nodes.NodeID = AE.NetworkNode
where AuditEventMessage like '%managed%'
order by Nodes.Caption
The problems with this query are:
1. If a node goes back into Managed mode because the time for Unmanaged mode expires, then that even is not logged (I suspect as it is done by the System account)
2. If someone re-unmanages a node, it destroys the existing UnmanageUntil data and makes the node look like it has been unmanaged for an extended period of time.
In order to resolve the issue, is it possible to log the Unmanage event when the time expires? This would resolve both problems, and make the reporting for the Unmanaged state much more accurate.
That said, has anyone else tried reporting on the amount of time that systems are in Unmanaged mode? Id be interested to find out how you did it, and the logic behind it. Thanks!