This works great for monitoring circuits that are "burstable". With many of our circuits, we have "purchased" a set amount of bandwidth but which are able "burst" over that set amount.
To that end we monitor for the purchased amount but configure for the bust amount.
Example - Circuit 10/100, The purchased amount is 10 meg of a 100 meg circuit.
First the bandwidth on the device interface is set to 100 meg (the maximum burstable rate).
Second the Custom Tx and Rx bandwidths are set within Orion for that interface to the "purchased" amount.
Third the change mentioned [<add key="ValidateCounters" value="False" />] is set on ALL the polling engines.
With this set-up we are able to
1) Track the amount of bustable bandwidth used (which we are of course we are billed for), but also this allows us to see if we need to increase the amount of purchased bandwidth
2) Alerting on utilization over 100% - Bust is being used
3) With netflow, we can see who and what is causing the overage.
Prior to this setup we did not see the bust. Now we are able to see the bust / overage utilization of a circuit and alert
So if you have an environment that has bustable circuits, I would suggest you looking into this type of setup.
Thanks for vote up