Select, Detect, Act, or why I consider Reflex being a watchdog
Reflex is many things next to one of the workloads of Microsoft Fabric. Before I delve into these things in more detail in later articles (yes, maybe this is the birth of another series of articles), I want to say this: Reflex is cool. It was never that simple to watch your data in your Power BI datasets (and this is only one of the capabilities of Reflex). Because I need images whenever I try to understand things, I start with a simple image of Reflex: I consider Reflex a watchdog! Reflex is watching something and alarms me or someone else when something happens - a defined condition is met. It will be likely that the first thing you ask Reflex to watch is a measure in one of your Power BI datasets. The dataset I use in this article is the dataset that holds the status of my on-premises data gateways. I store the status of each on-premises data gateway in a delta table-you can read about this here: (https://www.minceddata.info/2023/10/16/power-bi-gateway-monitoring-and-more-using-microsoft-fabric/) I defined a measure that returns 1 if the status of the gateway is “NotReachable.” This status indicates that something has happened, and the Power BI Service cannot use the gateway to support import and Direct Query solutions. This measure is simple: a new Measure = var maxActivityDateTime = CALCULATE( MAX( 'view1'[activityDateTime] ) ) return CALCULATE( COUNT( 'view1'[GatewayStatus] ) , 'view1'[GatewayStatus] = "NotReachable" , 'view1'[activityDateTime] = maxActivityDateTime ) The table ‘view1’ holds the status information about all the on-premises data gateways. The measure returns a value that is greater than 0 when the status of one the gateways becomes “NotReachable.” This measure returns the value only for the latest available status. The timestamp is created inside the notebook that is retrieving the status information and stored to the delta table.