It has happened us several times now that rather critical Qos values are no longer updated but still show (old) values in our dashboards. The result is that we and our customers are thinking things are running ok, but in fact are not ok.
When incidental (*) hovering the mouse over that specific item it shows a value with a SampleTime from days or even weeks ago! In fact all Qos values shown in custom dashboard does not have any validation check, it just shows just the last value (as it appears after investigating it, no matter if it is a day or a week old)
(*) note: custom dashboards are used on wall-display monitors, so checking qos values is not a common or standard procedure here.
As it was mentioned in an other and earlier post on this forum - alternatively you can use an actual reading/value using a direct call to the specific probe. For instance we use this for temperature readings in the MERS (based on snmpget) The risk of a wrong or obsolete temperature value here could have a big impact on many systems, even when normal alarm-notification is active.
What really needed is, in my opinion, is a robust way of validation when using Qos values in custom dashboard. It could be something like: NA (Not Available) or anything else, instead of showing an obsolete Qos value.
Does anyone have a thought or solution for this issue- or are we the only one with this kind of problems.?