log levels

[debrouwere]

It might be interesting to have the option to log extensively (when setting up a cluster and seeing how it behaves for the first couple of days) or less extensively (the most important diagnostic metrics) when everything seems to be in order. After all, AWS charges $0.05 per custom metric, and currently we have about 10 custom metrics per instance, not all of which are truly essential for daily monitoring. (On the other hand, when shit goes down, it's nice to have that extra level of detail, so I'm not entirely convinced yet that less logging should be encouraged.)