Ok, we are merging with another company and as a result are having to change nodes from being ICMP only at first, to being SNMP and ICMP at a later point in time.
I tried doing it in one big group, but tons of the nodes started showing up as being down. Only those for which the SNMP community hadn't been set up yet. This confused me because I thought, unless you explicitly set it in "List Resources" that a node would use ICMP for general node up/down status, not SNMP. But, in doing a bit of investigation, the nodes that were set up correctly for SNMP are having their "Status & Response Time" set to SNMP in "List Resources" too. So, I'm guessing that this is the case for those that don't have the SNMP community set up correctly, but since I can't get in to "List Resources" to check I can't tell for 100% sure. I did go to "Manage Pollers" and it looks like it is that way, but I'm not positive. Nor am I 100% sure how to change it in bulk from this screen yet, a bit confusing...
Now, I did also test just adding a node with a valid SNMP community right off the bat, ie: not converting it from ICMP-only to SNMP, and it set it correctly to use ICMP for status. I also tried doing a conversion on a single node from ICMP-only to SNMP, and it set the node to use SNMP for status once again. I think it should work the same both ways...
Now, I do have a dev server running 11.5 RC and it appears to be working correctly, always getting the status via ICMP. So I'm wondering, is this a bug I ran across that was fixed in 11.5, or do I have a setting somewhere in my config that is causing this to happen?