I have been trying to 'fine tune' our Orion's and reviewing the Data Processor logs I see there a lot of errors where it is trying to get SNMP data from down nodes. While I fully understand why it is doing what it is doing but to reduce the overhead on the Data Collector process it would be nice if there was a mechanism in place that Orion could say "hey, that device is down so I don't need to poll for SNMP data until it comes back up".
One of the driving forces behind this is we have quite a few "part time" or "On Demand" devices that we need to monitor, ICMP status and SNMP polling - when up.
The reason I targeted the Data Processor log is I routinely see where the Data Processor service is using a lot of CPU - I assume to process SNMP requests for down nodes.
Am I way off here??