Is it possible to get more details on the opsDB - perhaps a worked example, with the most common types of observations/expectations. It’d be good to see some code too.. Samplers, perhaps some of the stored-procs that figure out expectation misses, etc.
Also, some management data on the opsDB approach. Time to develop. Time saved by Ops, etc. So we can have a rich discussion about the investment needed for such a thing.
As you may deduce, I’m quite interested and excited by this area.