Configuring a Historical Alarms Step

The Historical Alarms step reads alarm and event history from an alarm history source, such as an Alarm Historian collection, into a data flow. Use it when you want to analyze alarm history in BI Server, for example, to build a data table of the alarms raised over the last day. Before you begin, open the data flow that will read the alarm history. Learn more

The step reads the events between a start time and an end time for each configured source. You choose which event fields to read, and each field becomes a column in the output of the step.

To add and configure a Historical Alarms step:

  1. Open Workbench and in Project Explorer, expand your project > Analytics > BI Server > Data Flows, and then open the desired data flow.

  2. In the Steps panel, select Click here to add new step, and in the drop-down list, select Data Sources > Historical Data > Historical Alarms. The step is added to the end of the list of steps, and its Time Settings, Historical Alarms Properties, and Data Sources appear next to it.

  3. In Data Sources, select Click here to add new item, and then enter or browse the point name of the alarm history source, for example an Alarm Historian collection such as hae:/Plant Collection. Repeat for each source that the step should read, or select Click to add multiple tags to browse and add several sources at once.

    • To take the point name from a data flow parameter instead, select Use Parameter and select the parameter.

  4. Under Time Settings, set the Start Time and the End Time of the period to read. The start time must be earlier than the end time.

    • To take either time from a data flow parameter, select the Use Parameter check box next to it and select the parameter. Parameters let a data table refresh the data flow for a moving time window. Learn more

  5. Under Historical Alarms Properties, next to Event Fields, select the Event Fields Editor button, select the fields that the data flow needs, and then confirm. The field list appears as the event type and field name of each selected field, separated by semicolons, for example BaseEventType.SourceName;BaseEventType.Time;BaseEventType.Message.

    Select only the fields that you need. BI Server requests exactly the selected fields from the source. Reading every field and then discarding columns with a Remove Column step reads all of the data from the source first.

    If you leave Event Fields empty, BI Server asks the source for its list of event fields each time the data flow runs, and it requests all of them: every field of the base event type and of its subtypes, except binary fields. If the source does not return that list, the read fails and a data table that loads from the data flow shows the Error status. Selecting the fields explicitly avoids that dependency.

  6. Leave Qualify Column Names selected to name each output column after its event type and field, such as BaseEventType.Message. Clear it to name the columns after the field alone, such as Message.

    The data type of each column is set from the first event that the step receives. A field whose first value is empty produces a String column, and a status code field produces a UInt32 column.

  7. Select Apply Changes. The Data Flow Preview panel shows the events that the step read, one column per event field.

  8. Add any further steps that shape the data, and then select Apply to save the data flow.