if you get this alert from opsmgr, it’s more than likely a problem with the management group(s) your agent is assigned. i didn’t catch on to this initially because of the myriad distractions, but the clue is in the alert description.
“The health service on myServer is attempting to communicate with a management group that does not exist on this server. The management group Id the agent is requesting is <guid>. Please verify that the management group specified on the agent and server is correct.”
so that is in fact, exactly what i did. in the registry, i found the entries to the old management group still plugged in there under this path:
hklm\software\microsoft\microsoft operations manager\3.0\agent management groups.
after that, the agent streamed down all the assigned management packs. looks like i might to get look forward to some clean up scenarios when deploying to production. if so, i’ll be scripting it. :)
update: unfortunately the management servers continued to complain about the agent. further investigation lead me to another path. after removing the key, restarting the health service on the health agent cleared up the problem. get rid of that problem group from the path here: