Currently when a poller is taken down for maintenance if the maintenance period is longer than the polling period of the devices it is monitoring it will send alerts that all of the monitored devices are down when it comes back up. This flood of alerts is confusing and unneeded. If the poller is down for less than the polling interval this is not an issue. Being able to select an option from the control panel to stop monitoring everything on one poller prior to doing maintenance on the poller would solve this issue.
This has been seen when nodes are taken down for upgrades, patches etc.