we found a way to use this to find systems that were not responding to SNMP. but it seems that it is not 100% up-to-date on the new v9.1sp5. We are finding that there are devices that have interfaces in unknown state that are being reported as group status of Up instead of the correct reporting of group status Up-Unknown.
A second issue we are seeing is that current status is not being filled properly when a device enters SNMP failure. example is that we had a server that was reporting 100% CPU and shortly thereafter stopped responding to SNMP but even hours later the device shows a current cpu usage of 100% even though that data is hours old. This use to revert to the -2 placeholder when this happened.
Guess my question is why are these changes being made and possibly how do you see this new implementation being more useful?
