@User you can randomise that process for a statistically relevant check but it wont be a 100% tested. if you want to develop such an algorithm
The Theory of the Design of Experiments D.R. COX Honorary Fellow Nuffield College Oxford, UK AND N. REID Professor of Statistics University of Toronto, Canada
01/24/2022, 8:33 PM
Yeah, probably even probabilistic sampling will work. Or EMA over error status. 🙂
Ideally, we don’t trace healthchecks, but trace migrations and their status only when they are explicitly running.