I'm moving this to the appropriate feature request forum and marking this for the PM to review.
Jeff, I've logged your request to have more granular control over node status.
Today, node status is determined with ICMP (ping). If SNMP is not working for a node, then the only way we really know this is we aren't able to poll the interface information (and therefore you get the gray "Unknown" box). Would you prefer the node be down if SNMP polling doesn't work? Can you tell me more about why SNMP status is important? We are also looking at adding SNMP only status. Under this scenario, status would be tied to SNMP information (as opposed to how it works know).
Actually node status via snmp for those devices configured for SNMP probably works the best. One of the issue's we have is polling devices on the other side of a firewalls. Firewall admins like to block icmp because DNoS attacks. Of couse you still have to deal with ICMP for devices that are just polled with ICMP. Truth is we don't have an issue of with using both ICMP and SNMP only the way it alerts. In todays NPM enviroment when SNMP stops working on a node the alert looks just like a node that is still responding to ICMP and SNMP but has an interface that has disappeared. To us an interface that has disappered is not as high a priority as a node that has SNMP that's stopped working. Most of the issues that stop SNMP from working are far more serious than and interface disappearing.
Keep to jeffnorton!
I think that is necessary to change the state of the Node on the 'Warning' when going to work 'Advanced Alert' by SNMP.
For example because in the Group with includes the Node not shown status 'Advanced Alert' of the Node.
And since at least the status can be seen.