Hello @Ellery , This can be achieved by workflows. Ultimately you will need to build the metric in either the Davis Anomaly Detector or in Metric Events. So you could look at the number of requests, being below a count of 1. Then set your sliding window for 1 hour. Additionally you'll want to set your violating sample count to the same as the sliding window. This will satisfy your requirement of 0 Requests for a full 60mins within a 60 min evaluation period. For Example, 8:00 - 9:15 There were no requests observed. At 9:00 a full hour of 0 Requests happened in a full hour window, so an alert was triggered. Now 9:15 we saw data, so if there is no more data after 9:15 till 10:15, then at 10:16 an alert will trigger as it will hit the 60 min full time period of 0 Data.
Now as it relates to the rule being on/off at a given time on given days of the week. You'll want to make a workflow that enables/disables the Metric Event at your given time frame. You can also use 3rd party tools and just post in an updated Json at the target time when you want the rule on/off.
-Chad
