2017-05-05 - production
04:21	<mutante>	scheduled long downtime for mailman I/O stats on fermium - until we find better ways to deal with the normal spikes causing alerts	[production]