Users are reporting multiple nodeDown messages on various nodes. OpenNMS should only report a single nodeDown message until the device is back up with a nodeUp message.
It appears this problem is related to reloading the pollerd configuration with:
send-event.pl -p 'daemonName Pollerd' uei.opennms.org/internal/reloadDaemonConfig
Create a nodeDown message by disconnecting or powering off a device (either monitored by OpenNMS or by a minion).
Send the above event
See the attachment. The original nodeDown happens at 10:02 and the duplicates at 10:05.