I'm a Noob to vRops. How to achieve this?
A bit rough and ready but one way could be to set a symptom definition with an appropriate wait cycle for powered down machines. So for 3 days (assuming default interval of 5 minutes per cycle) it would be 864 cycles.
Then set an alert with that symptom:
Of course there is probably a better way but its what I thought of first.
What version of vROps are you using?
Great answer above from pauljmcconnon!
Related threads:
Reporting on VM powered off duration. Where to set thresholds?